Risk Stratification and Role of Implantable Defibrillators for Prevention of Sudden Death in Patients with Hypertrophic Cardiomyopathy Barry J

Circulation Journal REVIEW Official Journal of the Japanese Circulation Society http://www.j-circ.or.jp Risk Stratification and Role of Implantable Defibrillators for Prevention of Sudden Death in Patients With Hypertrophic Cardiomyopathy Barry J. Maron, MD

Hypertrophic cardiomyopathy (HCM) is the most common cause of sudden cardiac death (SCD) in young peo- ple, including trained athletes. It is now 30 years since the introduction of implantable cardioverter-defibrillators (ICDs) to clinical cardiovascular practice and coronary artery disease, and now device therapy represents the most significant therapeutic innovation and the only definitive strategy for prolonging the life of HCM patients. ICDs have proved effective in preventing SCD in young HCM patients with appropriate intervention rates of 11% for secondary and 4% for primary prevention, despite massive left ventricular (LV) hypertrophy, LV outflow obstruction, diastolic dysfunction or microvascular ischemia. Targeting candidates for prophylactic ICD therapy can be complex, compounded by the unpredictability of the arrhythmogenic substrate, the absence of a dominant risk factor, and difficulty in assembling randomized trials. However, a single major risk factor is often sufficient to justify an ICD, although additional markers and other disease features can resolve ambiguous decision-making. Nevertheless, the absence of all risk factors does not convey absolute immunity to SCD. The current risk factor algorithm, when combined with a measure of individual physician judgment (and patient autonomy considerations), is an effective guide to identifying high-risk HCM patients. ICDs have altered the natural history of HCM for many patients and provided an opportunity to achieve many decades of productive life, and the potential for normal or near-normal longevity. Indeed, prevention of SCD has now become a new paradigm in the management of HCM. (Circ J 2010; 74: 2271 – 2282)

Key Words: Defibrillation; Hypertrophic cardiomyopathy; Implantable cardioverter-defibrillator (ICD); Sudden cardiac death

isk of sudden cardiac death (SCD) in patients with be superior to conventional antiarrhythmic drugs, for second- hypertrophic cardiomyopathy (HCM) has been rec- ary and primary SCD prevention in high-risk patients with R ognized since the initial contemporary description CAD or nonischemic cardiomyopathy.19 of the disease 50 years ago.1–7 SCD is the most visible and devastating complication of HCM,4–7 but it is only in the ICD in HCM past 10 years that the aspiration to prevent SCD has been In their seminal 1980 paper, Mirowski et al reported 3 pa- realized with the systematic use of implantable cardioverter- tients at exceptionally high risk for SCD after surviving 2 or defibrillators (ICDs) in high-risk HCM patients.8–17 Indeed, more cardiac arrests.18 In the laboratory, the implanted de- SCD prevention represents a turning point in the management fibrillator operated spontaneously and automatically to abort of HCM, by altering the natural history of the disease.8–10 ventricular fibrillation (VF). Notably, 2 of the 3 original pa- With the rate of ICD implants in HCM increasing sharply, tients had HCM, but for the ensuing 2 decades patients with it is timely to review the overall experience with this therapy HCM were largely ignored with regard to ICD therapy. in this complex disease. Pharmacologic therapy has failed to provide HCM patients with reliable protection against SCD. Historically, such medi- cations have included β-blockers and verapamil, for which Historical Perspectives of ICD Therapy there are no prospective data, type IA antiarrhythmic agents General such as quinidine and procainamide (abandoned because of The ICD was developed by Drs Michel Mirowski and Morton the potential for proarrhythmia), and amiodarone,9,10,20 which Mower in the early 1980s for patients with coronary artery is impractical for long-term administration to young patients disease (CAD).18 Over the past 20 years a number of prospec- given its potential for important side-effects. For example, tive, controlled and randomized trials have shown ICDs to in a large retrospective study, almost 30% of patients taking

Received September 15, 2010; accepted September 16, 2010; released online October 16, 2010 Hypertrophic Cardiomyopathy Center, Minneapolis Heart Institute Foundation, Minneapolis, MN, USA Mailing address: Barry J. Maron, MD, Hypertrophic Cardiomyopathy Center, Minneapolis Heart Institute Foundation, 920 E. 28th Street, Suite 620, Minneapolis, MN 55407, USA. E-mail: [email protected] ISSN-1346-9843 doi: 10.1253/circj.CJ-10-0921 All rights are reserved to the Japanese Circulation Society. For permissions, please e-mail: [email protected]

Circulation Journal Vol.74, November 2010 2272 MARON BJ

Figure 1. Sudden cardiac death (SCD) and age. (Top) SCD is most common before approximately 25 years of age, whereas heart failure and stroke generally occur later in life.7 (Reproduced with permission.) (Bottom) Single most frequent cause of SCD in young competitive athletes in the U.S.. ARVC, arrhythmogenic right ventricular cardiomyopathy; AS, aortic valve stenosis; CAD, coronary artery disease; CHD, congenital heart disease; CM, cardiomyopathy; LAD, left anterior de- scending; MVP, mitral valve prolapse; WPW, Wolff-Parkinson-White. *Regarded as possible (but not definitive) evidence for hypertrophic cardiomyopathy at autopsy with mildly increased LV wall thickness (15–19 mm) and heart weight (447±76 g). †Includes most commonly, Kawasaki disease, sickle cell trait and sarcoid. (Reproduced from Maron BJ, et al. Circulation 2009; 119: 1085 – 1092, with permission.)

amiodarone (as an adjunct to ICD therapy) experienced appro- 10 years from less selected regional or community-based priate ICD interventions.9,10 Furthermore, in an Italian HCM cohorts placed annual HCM mortality rates at a much more cohort largely devoid of ICD implants, cardioactive and anti- realistic ≤1%.3,4,7,21 arrhythmic drugs were ineffective in protecting patients from Nevertheless, the traditional profile of SCD in HCM re- lethal ventricular tachyarrhythmias (VT), and SCD occurred mains unchanged; that is, it usually occurs without warning in 20% of those taking amiodarone.20 in asymptomatic or mildly symptomatic young patients (pre- The ICD era in HCM treatment essentially began in earnest dominantly <25 years of age)3,4,7,13 In this respect, HCM is with the highly visible publication in 2000 of the first size- the most common cause of sudden death in competitive able series of patients.8 This report triggered enthusiasm for athletes in the U.S. (Figure 1).6 device therapy in HCM patients and subsequently greater Although SCD risk is lower in midlife and beyond, achiev- numbers of ICDs were implanted prophylactically. Interest in ing this measure of longevity does not confer immunity to defibrillator therapy for HCM patients has stimulated more SCD7,13 (Figure 1). No relationship between SCD risk and widespread application to other genetic heart diseases (ie, gender has been identified,22 and although no differences in ion channelopathies (long QT and Brugada syndromes) and risk according to race are known, SCD is common in African- arrhythmogenic right ventricular cardiomyopathy).13 American competitive athletes with HCM.6

Profile of SCD in HCM Estimates of the SCD rate in HCM unavoidably emanate from Selection of HCM Patients for ICDs hospital-based cohorts, and in the older literature were as high Conventional Risk Markers as 6%/year, which we now recognize as overestimates based There has been a substantial investigative effort in ICD deci- on tertiary center data contaminated by the preferential re- sion-making toward reliably identifying the relatively small ferral of high-risk patients.3,4 However, reports over the past subset of patients within the broad HCM disease spectrum

Circulation Journal Vol.74, November 2010 Implanted Defibrillators in HCM 2273

Figure 2. Risk stratification for sudden cardiac death (SCD). (Top) Pyra- mid profile currently used to identify patients at highest risk for SCD who are potential candidates for an implantable cardioverter-defibrillator (ICD). BP, blood pressure; LV, left ventricular; LVH, left ventricular hypertrophy; NSVT, nonsustained ventricular tachycardia; VT, ventricular tachycardia. *Following alcohol septal ablation, sustained VT has been reported in a significant minority of patients (≈10%) over the short term.36 (Reproduced with permission.) (Bot- tom) Direct relation between magnitude of LV hypertrophy (maximum [max] wall thickness by echocardiog- raphy) and SCD risk. Mild hypertrophy generally conveys lower risk and extreme hypertrophy (wall thickness ≥30 mm) is associated with the highest risk.23 (Reproduced with permission.)

who are at unacceptably high risk for SCD.3,4,7–10,13,23–27 There ≥1 episode of unexplained, recent syncope; (3) massive LV is universal agreement concerning the prudence of second- hypertrophy (thickness ≥30 mm); (4) nonsustained VT on ary prevention with ICDs for patients who have survived a serial ambulatory 24-h (Holter) ECGs; and (5) hypotensive cardiac arrest with documented VF or an episode of sustained or attenuated blood pressure response to exercise (Figure 2). VT, because of the relatively high recurrence rates in this However, the exercise blood pressure response is tested less subgroup.4 commonly than other risk factors13 and it rarely represents the However, selection of patients most likely to benefit from sole indicator for a prophylactic implant in clinical practice,3,13 ICD therapy for primary prevention is less resolved, with guide- although it is often used as an arbitrator when the risk assess- lines an evolving process. Such stratification of risk in HCM ment is otherwise ambiguous. Also, as a matter of practice, is predicated on generally accepted noninvasive markers, isolated brief runs of nonsustained VT on single, random usually in clinically stable patients, that have emerged from 24-h ambulatory (Holter) ECGs have not usually triggered observational studies.13,15,22,23–27 Of note, this strategy differs decisions for a prophylactic ICD, although frequent and/or conceptually from that in patients with CAD, in whom pri- prolonged (>10 beats) bursts of nonsustained VT, or repeti- mary prevention is based largely on a single predominant risk tive runs over extended serial monitoring periods, intuitively marker emanating from randomized trials, in which a major carry greater weight. clinical event (myocardial infarction) led to LV remodeling and systolic dysfunction (ejection fraction ≤30–35%), usu- Potential Arbitrators ally associated with disease progression.19 Certain features of HCM can contribute to resolution of ICD The conventional primary prevention risk factors in HCM, decisions when there is uncertainty regarding assignment of which assume greater weight in younger patients <50 years risk level, on a case-by-case basis (Figures 2,3). of age, are: (1) family history of ≥1 HCM-related SCDs; (2) LV Apical Aneurysms This recently reported subset of

Circulation Journal Vol.74, November 2010 2274 MARON BJ

Figure 3. Morphology of HCM patient subgroups associated with possible risk for sustained ventricular tachyarrhythmias. (A) Massive hypertrophy with ventricular septal (VS) thickness of 55 mm. (B) Akinetic thin-walled LV apical aneurysm with mid- cavity muscular apposition. D, distal (cavity); LA, left atrium; P, proximal (cavity); (B1) Contrast-cardiovascular magnetic resonance shows delayed enhancement (ie, scar) involving the thin aneurysm rim (arrowheads) and contiguous myocardium (large arrow); small apical thrombus is evident (small arrow). (C) Large transmural ventricular septal scar (arrow) resulting from alcohol ablation45 (arrow) (reproduced with permission). (D) “End-stage” heart showing extensive and transmural septal scarring, extending into the anterior wall (arrowheads)13 (Panels A,B,B1 and D) (reproduced with permission).

patients has been associated with a substantial (10%) annual gation showed no such relationship.33 Other obstacles to out- event rate, largely because of the arrhythmogenic substrate flow obstruction as a primary risk factor include its dynamic created by the fibrotic thin-walled aneurysm, as well as the nature, and the high frequency of gradients encountered in myocardial scarring identifiable by contrast cardiac mag- the HCM patient population (ie, 70%, at rest or with physi- netic resonance (CMR) in the contiguous distal LV wall27,28 ologic exercise).34 Reduction of obstruction by myectomy (Figure 3). Successful mapping and ablation of the VT focus (or alcohol ablation) is not considered a primary strategy for associated with the aneurysm has been reported.29 mitigating SCD risk. End-Stage Phase In a small minority of HCM patients, Delayed Enhancement (DE) Because current risk strati- approximately 3%, widespread (often transmural) LV scarring fication cannot guide SCD prevention in precise terms for evolves to irreversible systolic dysfunction, often associated each HCM patient, and SCD is known to occur in occasional with wall thinning and cavity dilatation30 (Figure 3D). Inevi- patients without any conventional risk factors, there is the tably, an adverse clinical course with progressive heart fail- ongoing aspiration to identify more sensitive and specific ure and often atrial and ventricular tachyarrhythmias ensues. clinical markers. Ideally, this would lead to a single, non- ICDs are used in the end-stage as a bridge to heart transplan- invasive, repeatable and accurate quantitative test (without tation. adding to patient risk). LV Outflow Obstruction A gradient ≥30 mmHg at rest is There is considerable interest surrounding in vivo detec- a quantitative measure of elevated intraventricular pressure tion of LV myocardial fibrosis on contrast-CMR as DE, and and wall stress. In 2 studies, obstruction had a very modest, its relationship to determinants of SCD and disease progres- though statistically significant, relationship to SCD risk (posi- sion.35–39 DE has been linked to the underlying and unstable tive predictive value, only 5–10%),31,32 but another investi- electrical substrate in HCM by the observation that non-

Circulation Journal Vol.74, November 2010 Implanted Defibrillators in HCM 2275

Figure 4. Delayed enhancement (DE) on cardiovascular magnetic resonance (CMR) images, as an arrhythmogenic substrate. (Top) Ventricular tachyarrhythmias on ambulatory (Holter) ECG, including nonsustained VT (NSVT), are significantly more frequent in the presence of DE. PVC, premature ventricular contraction; SVT, supraventricular tachycardia.32 (Reproduced with permission.) (Bot- tom) A 21-year-old man with hypertrophic cardiomyopathy (HCM) and septal scarring who survived an episode of ventricular fibrillation (VF) because of ICD intervention (A). Con- trast-enhanced CMR image showing transmural DE with high signal inten- sity occupying substantial proportion of septum (arrows). (B) Without contrast, showing moderate asymmetric hypertrophy of the ventricular septum (VS; 21 mm). (C) Intracardiac electrogram showing VF interrupted by defibrillation shock (arrow). AML, anterior mitral leaflet; FW, free wall; LV, left ventricular. (Reproduced from Maron BJ, et al. Am J Cardiol 2008; 101: 544 – 547 with permission of Elsevier.)

sustained VT on ambulatory Holter ECG is more common compared to myectomy; more than 20% of ablation patients in patients with DE than in patients without DE36 (Figure 4). experienced sudden or other cardiac death, aborted sudden Whether DE can be regarded as a bona fide risk marker in death, and/or appropriate ICD discharge, resulting in an HCM will ultimately require adequately powered studies in annual event rate of 4.4% following ablation.43 This experi- large populations with sufficient numbers of events accrued ence led the Rotterdam investigators to recommend myec- over many years.35 However, solely the presence of DE in tomy as the preferred intervention for obstructive HCM.43 a given HCM patient is unlikely to become a reliable risk These findings are similar to those at Massachusetts General marker, given that fully 50% of all patients with this disease Hospital, where an event rate of 5%/year (for VT, VF, and/or show some DE. appropriate ICD discharges) was reported.40 Furthermore, Alcohol Septal Ablation Percutaneous alcohol septal ab- high-risk HCM patients implanted with ICDs after alcohol lation is an alternative therapeutic strategy for selected HCM ablation have appropriate interventions for VT/VF of 3– patients to the preferred option of surgical myectomy,3,4 with 10%/year,9,13,42 which is substantially higher than the extreme- the capability of reducing the LV outflow gradient and heart ly low incidence of sustained VT reported after myectomy failure symptoms,40–44 by creating a sizeable transmural scar (0.2–0.9%/year).48 Indeed, the multicenter HCM-ICD regis- occupying on average 30% of the ventricular septum and try demonstrated the rate of appropriate ICD therapy among 10% of the overall LV chamber45,46 (Figure 3). alcohol ablation patients with primary prevention ICDs to be There is accumulating evidence and concern that the poten- 3-fold more frequent than in other patients (10.3%/year vs tial arrhythmogenicity of the scar created by alcohol septal 3.6%/year).9,10 It is not entirely certain how commonly such ablation can augment risk in the HCM population. Several arrhythmias are caused by the alcohol ablation procedure studies have documented the occurrence of sustained ven- itself or alternatively to the underlying disease. tricular arrhythmias and SCD following septal ablation in Whether HCM patients should have routine ICD im- approximately 10% of patients with or without SCD risk fac- plantation following alcohol septal ablation is unresolved. tors.9,13,40–43,47 In a recent single center experience, tenCate However, based on these considerations and concern that et al from the Thoraxcenter reported that long-term outcome alcohol-induced infarcts could compound preexisting and and survival was 4-fold less favorable with alcohol ablation underlying myocardial electrical instability, some practitio-

Circulation Journal Vol.74, November 2010 2276 MARON BJ

Figure 5. LAMP2 cardiomyopathy.53 (A) From a 14-year-old boy with sudden cardiac death and a septal thickness of 65 mm (heart weight 1,425 g). (B) Clusters of myocytes with vacuolated sarcoplasm (stained red) embedded in an area of scar (stained blue; Masson trichrome). (C) Disorganized arrangement of myocytes most typical of sarcomeric hypertrophic cardiomyopathy. (D) Intracardiac electrogram. The implantable cardioverter-defibrillator elicited 5 defibrillation shocks that failed to interrupt ventricular fibrillation (280 beats/min). (Reproduced with permission.)

ners have regarded alcohol ablation as a risk arbitrator13,44,47 not achieved the initial expectation that it would become a and prudently implanted ICDs in selected patients with other reliable strategy for predicting prognosis or selecting patients risk markers post-procedure (Figure 2).42 for primary prevention ICDs.13,52 The gene-based hypothe- sis for risk stratification is clinically untenable, because of Uncertain Contributors to Risk the heterogeneity of HCM, now with >1,000 mutations in Atrial Fibrillation (AF) AF is the most common sustained ≥11 genes, including many mutations for which the pathologic arrhythmia in HCM (20–25% of patients),49,50 associated with significance remains unresolved.44 Selected clinical situa- progressive heart failure and embolic stroke.16 However, there tions in which molecular diagnosis may predict prognosis is no compelling evidence that paroxysmal AF is a predictor have recently emerged, including LAMP2 cardiomyopathy,53 specifically of SCD in cohort analyses, although reported as an X-linked lysosomal storage disease, and possibly double an occasional trigger for VT.49 sarcomere mutations54 (Figure 5). Genotyping It was 20 years ago that there was initial Programmed Stimulation Laboratory electrophysiologic recognition that mutations in genes encoding proteins of testing, which directly probes the electrical properties of the the cardiac sarcomere cause HCM, which created substantial heart, has proved an impractical prognostic strategy now enthusiasm for identifying malignant or benign genetic sub- abandoned in HCM clinical practice as nonspecific, expensive, strates in order to facilitate assignment of SCD risk level.51 irrelevant with respect to clinical arrhythmias, and without Although now commercially available, genetic testing has advantage over current noninvasive risk stratification.13 Also,

Circulation Journal Vol.74, November 2010 Implanted Defibrillators in HCM 2277 there is insufficient evidence to regard specific 12-lead ECG patterns, T-wave alternans, or coronary arterial bridging as risk markers in HCM.13

Modifiable Risk Markers Participation in intense competitive sports can represent a potential risk factor in athletes with HCM, even when conventional markers are absent.6 The generally accepted rec- ommendation of Bethesda Conference #36 for trained athletes with HCM to reduce SCD risk is the withdrawal from most competitive sports.55 In older HCM patients, coexistent obstructive CAD may increase the overall SCD risk, potentially modifiable by coronary arterial intervention. The risk stratification algorithm in HCM (Figure 2) has proved to be a useful guide to selecting patients for ICD therapy, but a few SCDs have been reported in young HCM patients judged to be at low risk without any of the acknowl- edged risk markers, underscoring that the risk stratification strategy in HCM remains imprecise.13,56 Specifically, in pre- liminary data from the Minneapolis Heart Institute over the past 15 years, SCD events occurred in approximately 2% (0.5%/year) of young apparently low-risk patients, without primary prevention ICDs.

The Risk Factor Debate There has been a measure of controversy concerning the precise selection of HCM patients for prophylactic ICD strategies. Clinical practice in the European cardiovascular community has ascribed to a virtual mandatory minimum of 2 risk factors before recommending a prophylactic defibrillator.5,32 Certainly, the presence of multiple risk factors creates a clinical environment in which ICD decision-making is more intui- Figure 6. Number of risk factors.9 (Top) Appropriate im- tive and easier. Nevertheless, there is also considerable evi- plantable cardioverter-defibrillator (ICD) intervention rates dence that a single strong marker of increased risk within the (per 100 person-years) are not significantly different with re- clinical profile of an individual patient is sufficient to estab- spect to 1, 2 or ≥3 risk factors. (Bottom) Cumulative rates lish SCD risk as unacceptably increased, sufficient to raise the for first appropriate device intervention in patients with 1, 2 option of a primary prevention ICD.8–10 Indeed, an important or ≥3 risk factors. (Reproduced with permission.) proportion of appropriate ICD interventions for VT/VF occur in patients implanted because of only 1 risk factor (ie, 35%) and device therapy interventions do not differ significantly subsets of high-risk patients, ICD interventions have proved between patients with 1, 2 or ≥3 risk markers (Figure 6). to be both frequent and highly effective in aborting VT/VF. Rigid adherence to a minimum of 2 risk markers before rec- The largest and most recent of these studies is an interna- ommending a primary prevention ICD raises the possibility tional, multicenter retrospective/prospective registry assess- that some deserving 1-risk factor patients will be relegated to ment of 506 HCM patients from 42 centers in the U.S., Europe, a lower level of consideration for ICD therapy, or left essen- and Australia9 (Figure 7). tially unprotected. Patients in this ICD registry were predominantly asymp- On the other hand, not all 1-risk factor patients deserve tomatic or with only mild limiting heart failure symptoms, consideration for an ICD. HCM is a heterogeneous genetic and of relatively young age (average, 42 years old). Over a disease and inevitably there are numerous complex clinical mean follow-up of only 3.7 years, appropriate device dis- scenarios involving ambiguities and gray areas with respect charges (defibrillation shocks or antitachycardia pacing) for to the presence, strength or number of risk factors that ulti- VF or rapid VT occurred in 20% of patients (Figure 7), an mately affect ICD decision-making. One specific example is implant-to-life-saved ratio of 5:1. The overall appropriate the elderly patient with syncope as a single risk factor. Such ICD discharge rate was 5.5%/year, at an average age of only a patient may not be a candidate for primary prevention, 44 years. Cumulative probability of an appropriate ICD inter- given that HCM-related SCD is uncommon in this age range, vention 5 years after implant was almost 25%, indicating the survival to advanced age itself declares lower risk status in potential for ongoing SCD risk in the absence of an ICD. this disease, and syncope is relatively common in elderly The intervention rate for secondary prevention (implant individuals. after cardiac arrest) was highest (ie, 11%/year), and for primary prevention was 4%/year in those patients implanted based only on conventional risk factors (Figure 7). Consistent ICD Experience in HCM with the predilection of SCD in young patients with HCM, Efficacy almost 30% of those receiving ICDs were ≤20 years of age, There is now substantial evidence for the reliability of ICDs largely for primary prevention, and they experienced appro- in preventing SCD in HCM patients.8–12,14–17,48,56–63 In selected priate ICD interventions at an average of 18±4 years of age

Circulation Journal Vol.74, November 2010 2278 MARON BJ

Figure 7. Prevention of sudden cardiac death (SCD).8 (Top) Intracardi- ac electrogram obtained at 1:20 am while asleep 5 years after placement of an implantable cardioverter-defibrillator (ICD). From a 35-year-old man with hypertrophic cardiomyopathy who received prophylactic ICD because of family history of SCD and marked ventricular septal thickness (31 mm). (A) Ventricular tachycardia (VT) be- gins abruptly, at 200 beats/min. (B) Defibrillator senses VT and charges. (C) VT deteriorates into ventricular fibrillation (VF) (D) defibrillator issues 20-J shock (arrow), restoring sinus rhythm immediately. Virtually identi- cal sequence occurred 9 years later during sleep; patient is now 52 years old and asymptomatic. (Reproduced with permission.) (Bottom) Flow-chart of ICD-related outcome in 506 high- risk HCM patients from an interna- tional, multicenter ICD registry.9

(ie, 7%/year). Intervention rates in the multicenter study are sence of systolic dysfunction).8–10 However, a notable excep- consistent with those reported individually from other coun- tion in the registry of 506 patients was a 21-year-old student tries, including Spain,16 Poland,17,58 Canada,12 Portugal,60 with extreme LV hypertrophy, preserved systolic function, United Kingdom,61 Germany,59 Australia,11 Italy9 and other and syncope, who died suddenly when a mechanically defec- U.S. centers.48,62,63 tive ICD failed to deliver an appropriate shock because of It is notable that the ICD has performed so reliably in ter- massive electrical overstress resulting from short-circuiting,9,66 minating VT/VF in patients with a disease such as HCM, a design flaw known only to the manufacturer. Another ex- which is associated with unique pathology and hemodynamic ception is LAMP2 cardiomyopathy, a HCM phenocopy with features such as a substantial increase in LV mass (wall thick- massive LV hypertrophy, which appears largely refractory ening up to 3- to 5-fold greater than normal),23,64 dynamic to ICD therapy (Figure 5).53 obstruction to LV outflow with elevated intraventricular pres- sures,31,34 diastolic dysfunction and microvascular myocardial Event Occurrence ischemia due to small vessel disease.65 The timing of SCD events in HCM is largely unpredict- In our large registry experience the ICD proved effective able8–10,13,57,67 (Figures 7 and 8). The interval from ICD im- in preventing SCD in virtually every patient with the typical plant to first appropriate device intervention is quite variable, clinical and phenotypic expression of HCM (ie, in the ab- and often considerable in length (ie, as long as 10 years be-

Circulation Journal Vol.74, November 2010 Implanted Defibrillators in HCM 2279

Figure 8. Unpredictability of arrhythmogenic substrate in HCM.57 (Top) Time interval between implantation of implantable cardioverter-defibrillator (ICD) and first appropriate intervention. Variable time delay after implantation is evident, with some device discharges occurring relatively early and others 5–10 years later (blue bars). (Reproduced with permission.) (Bottom) Hourly distribution of appropriate ICD interventions over the 24-h day for 126 ventricular tachycardia/ventricular fibrillation events in 63 patients with HCM. (Reproduced with permission.)

fore the first life-saving intervention).8–10 In addition, several Data on ICD performance in HCM patients are assembled patients have survived >10 years (and up to 30 years) after out of necessity in cohorts judged to be at high risk in clinical cardiac arrest with or without the aid of a defibrillator inter- practice, and therefore the reported appropriate ICD discharge vention.67 rates may not be entirely representative of a truly general Therefore, initial clinical recognition of high-risk status and HCM population with a more benign profile.3,4,21 Also, reli- ICD implantation can precede by many years the occurrence ance on retrospective, observational studies is unavoidable, of potentially lethal VT that the defibrillator is required to because it is virtually impossible to design randomized ICD interrupt.8–10,13,57 trials in this disease given its unique features, including the Notably, development of disabling heart failure symptoms relatively low prevalence of HCM in cardiovascular practice, following major arrhythmic events appear to be uncommon its heterogeneous clinical expression, and infrequent events in HCM,67 and at present there is no evidence that ICDs dispersed over decades. Also, there are ethical and other con- merely shift the mode of demise from SCD to progressive siderations related to randomizing patients in a study involv- heart failure, as suggested in CAD.68 This observation raises ing a life-saving device. the distinct possibility of achieving normal or near-normal longevity with (and because of) the ICD (Figure 9). Once the potential risk is recognized and presented to the HCM vs CAD patient, it is difficult and probably imprudent to temporize or High-risk HCM patients are much younger on average with delay potentially preventive treatment. When the decision extended periods of future SCD risk (average 42 years at to implant an ICD in a high-risk HCM patient is made, it is implant and 44 at first ICD intervention),3–17,23–28,48 com- likely to represent a life-long preventative measure. pared to patients with CAD who are of relatively advanced

Circulation Journal Vol.74, November 2010 2280 MARON BJ

Figure 9. Profiles in prognosis. Patients with increased risk of sudden cardiac death often represent a distinct subgroup in which long-term prognosis is largely or completely dependent on this disease complication, exclusive of other potentially adverse consequences. In such patients, the ICD has the potential to convey near- normal or normal longevity. ICD, implantable cardioverter-defibrillator; HCM, hypertrophic cardiomyopathy.

ages at the time of implantation (average 65 years, or 25 years older), often with progressive heart failure as a consequence of prior myocardial infarction and LV dysfunction.19,68 In sharp contrast, ICDs are usually recommended to HCM patients who are asymptomatic with preserved LV systolic function. Therefore, although the annual appropriate ICD intervention rates are lower in HCM patients than in CAD patients, they are nevertheless particularly significant given the much younger patient population largely free of limiting heart failure symptoms.

Other Considerations Affecting ICD Therapy Complications The decision to implant an ICD prophylactically for SCD prevention in HCM patients always includes the consideration of the potential complications and inconvenience incurred by a permanent device vs the obvious life-saving benefit should the ICD effectively terminate a lethal arrhythmia. Neverthe- less, these 2 scenarios are not of equal weight, given the power of ICDs for preserving life. However, a measure of hesitancy toward life-long prophylactic ICDs may arise in pediatric cardiology practice, confronted by the clinical par- adox in which active and apparently healthy young HCM patients are exposed to the greatest SCD risk (by age), but also device complications. ICD-related complications are well documented, including infection, pocket hematoma, pneumothorax and venous thrombosis in a small minority.8–10,13,41,58–63 However, 25% of HCM patients8–10 experience inappropriate shocks (5%/ year),62 resulting from lead fracture or dislodgement, over- Figure 10. Primary prevention decision tree. When assess- sensing, double counting and programming malfunctions, or ment of risk status lies in a gray zone of uncertainty and when triggered inadvertently by sinus tachycardia or AF. The ambiguity, prudent clinical decisions rely on transparency, highest rate of complications is in younger HCM patients, full disclosure, and informed consent, utilizing a large measure of autonomy on the part of the fully informed patient. largely because their activity level and body growth places a continual strain on the leads, which are regarded as the weakest link in the system.63,69 Extended lead survival is par-

Circulation Journal Vol.74, November 2010 Implanted Defibrillators in HCM 2281 ticularly crucial for young patients because they will have 2212 – 2218. their ICDs for decades (probably most of their lives). 6. Maron BJ. Sudden death in young athletes. N Engl J Med 2003; Industry-related ICD problems have disproportionately 349: 1064 – 1075. 7. Maron BJ, Olivotto I, Spirito P, Casey SA, Bellone P, Gohman affected HCM patients, in which prominent recalls have in- TG, et al. Epidemiology of hypertrophic cardiomyopathy-related cluded defective generators resulting in several deaths,66 and death: Revisited in a large non-referral-based patient population. small-diameter high voltage leads prone to fracture.69 The im- Circulation 2000; 102: 858 – 864. plant procedure has been largely free of significant risk, with- 8. Maron BJ, Shen WK, Link MS, Epstein AE, Almquist AK, Daubert JP, et al. Efficacy of implantable cardioverter-defibrillators for the out reported deaths, although selected patients with extreme prevention of sudden death in patients with hypertrophic cardio- hypertrophy, or who have been administered amiodarone, may myopathy. N Engl J Med 2000; 342: 365 – 373. require high-energy output generators or epicardial leads.10,13 9. Maron BJ, Spirito P, Shen WK, Haas TS, Formisano F, Link MS, et al. Implantable cardioverter-defibrillators and prevention of sudden cardiac death in hypertrophic cardiomyopathy. JAMA 2007; ICDs Worldwide: Cultural and Other Factors 298: 405 – 412. 10. Maron BJ, Spirito P. Implantable defibrillators and prevention of The rate of prophylactic defibrillator implants in HCM pa- sudden death in hypertrophic cardiomyopathy. J Cardiovasc Elec- tients is related in part to the country of origin as determined trophysiol 2008; 19: 1118 – 1126. 11. Jayatilleke I, Doolan A, Ingles J, McGuire M, Booth V, Richmond by differences in healthcare systems and a number of cultural, DR, et al. Long-term follow-up of implantable cardioverter defi- 13 societal, and economic factors. The considerable variabil- brillator therapy for hypertrophic cardiomyopathy. Am J Cardiol ity among nations and cultures with respect to ICD implants 2004; 93: 1192 – 1194. undoubtedly affects clinical decision-making and, in particu- 12. Woo A, Monakier D, Harris L, Hill A, Shah P, Wigle ED, et al. lar, the threshold for recommending primary prevention de- Determinants of implantable defibrillator discharges in high-risk patients with hypertrophic cardiomyopathy. Heart 2007; 93: 1044 – vices to HCM patients. ICDs are most common in the U.S., 1045. more than 2-fold that of Germany and 5–10-fold greater than 13. Maron BJ. Contemporary insights and strategies for risk stratifica- in most European countries, including the UK, where the man- tion and prevention of sudden death in hypertrophic cardiomyopa- datory 2-risk factor model for primary prevention prevails. thy. Circulation 2010; 121: 445 – 456. 14. Begley DA, Mohiddin SA, Tripodi D, Winkler JB, Fananapazir L. In Japan, other Asian countries, the Middle East, and Efficacy of implantable cardioverter defibrillator therapy for primary Eastern Europe, HCM implant rates have been generally quite and secondary prevention of sudden cardiac death in hypertrophic low, probably related primarily to cultural factors. These cardiomyopathy. Pacing Clin Electrophysiol 2003; 26: 1887 – 1896. observations raise the distinct possibility that HCM patients 15. Bos JM, Maron BJ, Ackerman MJ, Haas TS, Sorajja P, Nishimura RA, et al. Role of family history of sudden death in risk stratifica- with the same level of risk in different countries may not have tion and prevention of sudden death with implantable defibrillators the same access to prophylactic ICDs and the opportunity for in hypertrophic cardiomyopathy. Am J Cardiol 2010 (in press). SCD prevention. 16. Marin F, Gimeno JR, Paya E, Garcia-Alberola A, Perez-Alverz L, Fernandez X, et al. The implantable cardioverter-defibrillator and hypertrophic cardiomyopathy: Experience at three centers. Rev Esp ICD Decision-Making Cardiol 2006; 59: 537 – 544. 17. Syska P, Przybylski A, Chojnowska L, Lewandowski M, Sterlin´ski Clinical dilemmas inevitably arise concerning ICD recom- M, Macia˛g A, et al. Implantable cardioverter-defibrillator in patients mendations, because many HCM patients fall into ambigu- with hypertrophic cardiomyopathy: Efficacy and complications of ous gray zones in which the level of risk cannot be assessed the therapy in long-term follow-up. J Cardiovasc Electrophysiol 2010; 21: 883 – 889. with precision using conventional risk factors (or arbitrators), 18. Mirowski M, Reid PR, Mower MM, Watkins L, Gott VL, Schauble and a measure of individual clinical judgment and experi- JF, et al. Termination of malignant ventricular arrhythmias with an ence of the managing physician with direct knowledge of the implanted automatic defibrillator in human beings. N Engl J Med patient’s clinical profile and desires is necessary (Figure 10). 1980; 303: 322 – 324. 19. Epstein AE. Benefits of the implantable cardioverter-defibrillator. Indeed, transparency, full disclosure and informed consent J Am Coll Cardiol 2008; 52: 1122 – 1127. in association with autonomous input concerning the desires 20. Melacini P, Maron BJ, Bobbo F, Basso C, Tokajuk B, Zucchetto of the fully informed patient and family is necessary for M, et al. Evidence that pharmacological strategies lack efficacy for resolving otherwise ambiguous decisions for prophylactic the prevention of sudden death in hypertrophic cardiomyopathy. Heart 2007; 93: 708 – 710. ICDs in which there are gaps in knowledge or absence of 21. Maron BJ, Casey SA, Poliac LC, Gohman TE, Almquist AK, Aeppli sufficient evidence-based data. DM. Clinical course of hypertrophic cardiomyopathy in a regional United States cohort. JAMA 1999; 281: 650 – 655 (correction: JAMA 1999; 281: 2288). Acknowledgment 22. Olivotto I, Maron MS, Adabag AS, Casey SA, Vargiu D, Link MS, Funded support from the Hearst Foundations, New York, NY, USA. et al. Gender-related differences in the clinical presentation and outcome of hypertrophic cardiomyopathy. J Am Coll Cardiol 2005; 46: 480 – 487. References 23. Spirito P, Bellone P, Harris KM, Bernabo P, Bruzzi P, Maron BJ. 1. Teare D. Asymmetrical hypertrophy of the heart in young adults. Magnitude of left ventricular hypertrophy predicts the risk of Br Heart J 1958; 20: 1 – 8. sudden death in hypertrophic cardiomyopathy. N Engl J Med 2000; 2. Braunwald E, Lambrew C, Rockoff D, Ross J Jr, Morrow AG. Idio- 342: 1778 – 1785. pathic hypertrophic subaortic stenosis. I: Description of the disease 24. Monserrat L, Elliott PM, Gimeno JR, Sharma S, Penas-Lado M, based on the analysis of 64 patients. Circulation 1964; 30(Suppl McKenna WJ. Non-sustained ventricular tachycardia in hypertro- IV): 1 – 217. phic cardiomyopathy: An independent marker of sudden death risk 3. Maron BJ. Hypertrophic cardiomyopathy: A systematic review. in young patients. J Am Coll Cardiol 2003; 42: 873 – 879. JAMA 2002; 287: 1308 – 1320. 25. Spirito P, Autore C, Rapezzi C, Bernabo P, Badagliacca R, Maron 4. Maron BJ, McKenna WJ, Danielson GK, Kappenberger LJ, Kuhn MS, et al. Syncope and risk of sudden death in hypertrophic cardio- HJ, Seidman CE, et al. ACC/ESC Clinical expert consensus docu- myopathy. Circulation 2009; 119: 1703 – 1710. ment on hypertrophic cardiomyopathy. J Am Coll Cardiol 2003; 26. Puntmann VO, Yap YG, McKenna W, Camm AJ. Significance of 42: 1687. maximal and regional left ventricular wall thickness in association 5. Elliott PM, Poloniecki J, Dickie S, Sharma S, Monserrat L, with arrhythmic events in patients with hypertrophic cardiomyopa- Varnava A, et al. Sudden death in hypertrophic cardiomyopathy: thy. Circ J 2010; 74: 531 – 537. Identification of high risk patients. J Am Coll Cardiol 2000; 36: 27. Maron MS, Finley JJ, Bos JM, Hauser RH, Manning WJ, Haas TS,

Circulation Journal Vol.74, November 2010 2282 MARON BJ

et al. Prevalence, clinical significance and natural history of left ven- for severely symptomatic patients with obstructive hypertrophic tricular apical aneurysms in hypertrophic cardiomyopathy. Circula- cardiomyopathy [Controversies in cardiovascular medicine]. Circu- tion 2008; 118: 1541 – 1549. lation 2007; 116: 196 – 206. 28. Fattori R, Biagina E, Lorenzini M, Buttazzi K, Lovato L, Rapezzi 48. McLeod CJ, Ommen SR, Ackerman MJ, Weivoda PL, Shen WK, C. Significance of magnetic resonance imaging in apical hypertro- Dearani JA, et al. Surgical septal myectomy decreases the risk for phic cardiomyopathy. Am J Cardiol 2010; 105: 1592 – 1596. appropriate implantable cardioverter defibrillator discharge in ob- 29. Lim KK, Maron BJ, Knight BP. Successful catheter ablation of structive hypertrophic cardiomyopathy. Eur Heart J 2007; 28: hemodynamically unstable monomorphic ventricular tachycardia in 2583 – 2588. a patient with hypertrophic cardiomyopathy and apical aneurysm. 49. Olivotto I, Cecchi F, Casey SA, Dolara A, Traverse JH, Maron BJ. J Cardiovasc Electrophysiol 2009; 20: 445 – 447. Impact of atrial fibrillation on the clinical course of hypertrophic 30. Harris KM, Spirito P, Maron MS, Zenovich AG, Formisano F, cardiomyopathy. Circulation 2001; 104: 2517 – 2524. Lesser JR, et al. Prevalence, clinical profile, and significance of left 50. Kubo T, Kitaoka H, Okawa M, Hirota T, Hayato K, Yamasaki N, ventricular remodeling in the end-stage phase of hypertrophic car- et al. Clinical impact of atrial fibrillation in patients with hypertro- diomyopathy. Circulation 2006; 114: 216 – 225. phic cardiomyopathy: Results from Kochi RYOMA Study. Circ J 31. Maron MS, Olivotto I, Betocchi S, Casey SA, Lesser JR, Losi MA, 2009; 73: 1599 – 1605. et al. Effect of left ventricular outflow tract obstruction on clinical 51. Alcalai R, Seidman JG, Seidman CE. Genetic basis of hypertrophic outcome in hypertrophic cardiomyopathy. N Engl J Med 2003; 348: cardiomyopathy: From bench to the clinics. J Cardiovasc Electro- 295 – 303. physiol 2008; 19: 104 – 110. 32. Elliott PM, Gimeno JR, Tomé MT, Shah J, Ward D, Thaman R, et 52. Ackerman MJ, VanDriest SL, Ommen SR, Will ML, Nishimura al. Left ventricular outflow tract obstruction and sudden death risk RA, Tajik AJ, et al. Prevalence and age-dependence of malignant in patients with hypertrophic cardiomyopathy. Eur Heart J 2006; mutations in the beta-myosin heavy chain and troponin T genes in 27: 1933 – 1941. hypertrophic cardiomyopathy: A comprehensive outpatient perspec- 33. Efthimiadis GK, Parcharidou DG, Giannakoulas G, Pagourelias ED, tive. J Am Coll Cardiol 2002; 39: 2042 – 2048. Charalampidis P, Savvopoulos G, et al. Left ventricular outflow 53. Maron BJ, Roberts WC, Arad M, Haas TS, Spirito P, Wright GB, tract obstruction as a risk factor for sudden cardiac death in hyper- et al. Clinical outcome and phenotypic expression in LAMP2 car- trophic cardiomyopathy. Am J Cardiol 2009; 104: 695 – 699. diomyopathy. JAMA 2009; 301: 1253 – 1259. 34. Maron MS, Olivotto I, Zenovich AG, Link MS, Pandian NG, 54. Ingles J, Doolan A, Chiu C, Seidman J, Seidman C, Semsarian C. Kuvin JT, et al. Hypertrophic cardiomyopathy is predominantly a Compound and double mutations in patients with hypertrophic car- disease of left ventricular outflow tract obstruction. Circulation diomyopathy: Implications for genetic testing and counseling. J Med 2006; 114: 2232 – 2239. Genet 2005; 42: e59. 35. Maron MS, Appelbaum E, Harrigan C, Buros J, Gibson M, Hanna 55. Maron BJ, Zipes DP. 36th Bethesda Conference: Eligibility recom- CA, et al. Clinical profile and significance of delayed enhancement mendations for competitive athletes with cardiovascular abnormal- in hypertrophic cardiomyopathy. Circ Heart Fail 2008; 1: 184 – 191. ities. J Am Coll Cardiol 2005; 45: 1312 – 1375. 36. Adabag AS, Maron BJ, Appelbaum E, Harrigan CJ, Buros JL, 56. Maron BJ. Sudden death in hypertrophic cardiomyopathy. J Car- Gibson CM, et al. Occurrence and frequency of arrhythmias in diovasc Trans Res 2009; 2: 368 – 380. hypertrophic cardiomyopathy in relation to delayed enhancement 57. Maron BJ, Semsarian C, Shen WK, Link MS, Epstein AE, Estes on cardiovascular magnetic resonance. J Am Coll Cardiol 2008; NAM III, et al. Circadian patterns in the occurrence of malignant 51: 1369 – 1374. ventricular tachyarrhythmias triggering defibrillator interventions 37. Bruder O, Wagner A, Jensen CJ, Schneider S, Ong P, Kispert EM, in patients with hypertrophic cardiomyopathy. Heart Rhythm 2009; et al. Myocardial scar visualized by cardiovascular magnetic reso- 6: 599 – 602. nance imaging predicts major adverse events in patients with hyper- 58. Przybylski A, Małecka L, Pytkowski M, Chojnowska L, trophic cardiomyopathy. J Am Coll Cardiol 2010 [E-pub ahead of Lewandowski M, Sterlin´ski M, et al. Implantable cardioverter-defi- print]. brillators in patients with hypertrophic cardiomyopathy: Dilemmas 38. O’Hanlon R, Grasso A, Roughton M, Moon JC, Clark S, Wage R, and difficulties.Kardiol Pol 2005; 63: 391 – 397. et al. Prognostic significance of myocardial fibrosis in hypertrophic 59. Eicken A, Kolb C, Lange S, Brodherr-Heberlein S, Zrenner B, cardiomyopathy. J Am Coll Cardiol 2010 [Epub ahead of print]. Schreiber C, et al. Implantable cardioverter defibrillator (ICD) in 39. Motoyasu M, Kurita T, Onishi K, Uemura S, Tanigawa T, Okinaka children. Int J Cardiol 2006; 107: 30 – 35. T, et al. Correlation between late gadolinium enhancement and 60. Vieira AP, Adraga˜o PP, Santos KR, Morgado FB, Cavaco DM, diastolic function in hypertrophic cardiomyopathy assessed by Rossi R, et al. Implantable cardioverter-defibrillators in hypertro- magnetic resonance imaging. Circ J 2008; 72: 378 – 383. phic cardiomypathy. Rev Port Cardiol 2005; 24: 407 – 415. 40. Noseworthy PA, Rosenberg MA, Fifer MA, Palacios IF, Lowry 61. Kaski JP, Esteban MTT, Lowe M, Sporton S, Rees P, Deanfield JE, PA, Ruskin JN, et al. Ventricular arrhythmia following alcohol et al. Outcomes after implantable cardioverter-defibrillator treat- ablation for obstructive hypertrophic cardiomyopathy. Am J Car- ment in children with hypertrophic cardiomyopathy. Heart 2007; diol 2009; 104: 128 – 132. 93: 372 – 374. 41. Sorajja P, Valeti U, Nishimura R, Ommen SR, Rihal CS, Gersh BJ, 62. Lin G, Nishimura RA, Gersh BJ, Ommen S, Ackerman M, Brady et al. Outcome of alcohol septal ablation for obstructive hyper- PA. Device complications and inappropriate implantable cardio- trophic cardiomyopathy. Circulation 2008; 118: 131 – 139. verter-defibrillator shocks in patients with hypertrophic cardiomy- 42. Cuoco FA, Spencer WH III, Fernandes VL, Nielsen CD, Nagueh S, opathy. Heart 2009; 95: 709 – 714. Sturdivant JL, et al. Implantable cardioverter-defibrillator therapy 63. Berul CI, Van Hare GF, Kertesz NJ, Dubin AM, Cecchin F, Collins for primary prevention of sudden death after alcohol septal abla- KK, et al. Results of a multicenter retrospective implantable car- tion of hypertrophic cardiomyopathy. J Am Coll Cardiol 2008; 52: dioverter-defibrillator registry of pediatric and congenital heart dis- 1718 – 1723. ease patients. J Am Coll Cardiol 2008; 51: 1685 – 1691. 43. tenCate F, Soliman, OII, Michels M, Theuns D, De Jong PL, 64. Maron MS, Maron BJ, Harrigan C, Buros J, Gibson CM, Olivotto Geleijnse MI, et al. Long-term outcomes of alcohol septal ablation I, et al. Hypertrophic cardiomyopathy phenotype revisited after 50 in patients with obstructive hypertrophic cardiomyopathy: A word years with cardiovascular magnetic resonance. J Am Coll Cardiol of caution. Circ Heart Fail 2010; 3: 362 – 369. 2009; 54: 220 – 228. 44. Nishimura RA, Ommen SR. Septal reduction therapy for obstruc- 65. Maron MS, Olivotto I, Maron BJ, Prasad SK, Cecchi F, Udelson tive hypertrophic cardiomyopathy and sudden death: What statistics JE, et al. The case for myocardial ischemia in hypertrophic cardio- cannot tell you. Circ Cardiovasc Int 2010; 3: 91 – 93. myopathy. J Am Coll Cardiol 2009; 54: 866 – 875. 45. Valeti US, Nishimura RA, Holmes DR, Araoz PA, Glockner JF, 66. Hauser RG, Maron BJ. Lessons from the failure and defibrillator. Breen JF, et al. Comparison of surgical septal myectomy and alco- Circulation 2005; 112: 2040 – 2042. hol septal ablation with cardiac magnetic resonance imaging in 67. Maron BJ, Haas TS, Shannon KM. Long-term survival after car- patients with hypertrophic obstructive cardiomyopathy. J Am Coll diac arrest in hypertrophic cardiomyopathy. Heart Rhythm 2009; Cardiol 2007; 49: 350 – 357. 6: 993 – 997. 46. van Dockum WG, ten Cate FJ, ten Berg JM, Beek AM, Twisk JW, 68. Tung R, Zimetbaum P, Josephson ME. A critical appraisal of im- Vos J, et al. Myocardial infarction after percutaneous transluminal plantable cardioverter-defibrillator therapy for the prevention of septal myocardial ablation in hypertrophic obstructive cardiomyop- sudden cardiac death. J Am Coll Cardiol 2008; 52: 1111 – 1121. athy: Evaluation by contrast-enhanced magnetic resonance imag- 69. Hauser RG, Kallinen LM, Almquist AK, Gornick CC, Katsiyiannis ing. J Am Coll Cardiol 2004; 43: 27 – 34. WT. Early failure of a small-diameter high-voltage implantable 47. Maron BJ. Surgical myectomy remains the primary treatment option cardioverter-defibrillator lead.Heart Rhythm 2007; 4: 892 – 896.

Circulation Journal Vol.74, November 2010