Introduction. CLINICAL RESEARCH Pacing and Cardiac Resynchronization Therapy

Transcription

1 Europace (2009) 11, doi: /europace/eun288 CLINICAL RESEARCH Pacing and Cardiac Resynchronization Therapy Relationship between New York Heart Association class change and ventricular tachyarrhythmia occurrence in patients treated with cardiac resynchronization plus defibrillator Antoine Lepillier 1,2 *, Olivier Piot 1, Bart Gerritse 3, Xavier Copie 1, Thomas Lavergne 2, Olivier Paziaud 1, Gilles Lascault 1, Xavier Waintraub 2, Akli Otmani 2, and Jean-Yves Le Heuzey 2 1 Centre Cardiologique du Nord, 26 rue des Moulins Gémeaux Saint-Denis, France; 2 Service de cardiologie, Hopital Europeen Georges Pompidou, 20 rue Leblanc Paris, France; and 3 Medtronic Bakken Research Center, Maastricht, Netherlands Received 30 May 2008; accepted 29 October 2008; online publish-ahead-of-print 24 October 2008 Aims In patients with advanced heart failure (HF) and prolonged QRS interval, cardiac resynchronization therapy (CRT) reduces symptoms and risk of death. The added benefit of an implantable cardioverter defibrillator (ICD) remains questionable in some patients.... Methods and In 332 HF patients treated with CRT-D (CRT with ICD) [ years, 86% men, 23% New York Heart Association results (NYHA) class II, 65% class III, and 11% class IV, 70% primary prevention, 55% ischaemic cardiomyopathy, left ventricular ejection fraction %, and QRS width ms], we evaluated the relationship between functional status change, death at 6-month follow-up (FU), and the occurrence of ventricular tachyarrhythmia/ventricular fibrillation (VT/VF). A total of 68 patients (20.5%) experienced 1266 spontaneous episodes of VT/VF during FU. There was no difference in baseline characteristics between patients with or without VT/VF, except for ICD indication (primary or secondary prevention). Improvement in NYHA class was significantly associated with a decreased occurrence of VT/VF (P ¼ 0.004). Sixteen patients who died had significantly more often VT/VF than the survivors (50 vs. 19%, P ¼ 0.007).... Conclusion Within the initial 6-month post-crt therapy, 20% of patients received an appropriate ICD therapy. Patients improving on NYHA class (responders to CRT) have less VT/VF episodes than non-responders. Discriminant criteria for CRT response are awaited to optimize the choice of the device (CRT alone, defibrillator alone, or CRT-D) Keywords Defibrillator Resynchronization Ventricular arrhythmias Introduction Benefits of implantable cardioverter defibrillators (ICDs) on total mortality have been demonstrated in primary prevention of heart failure (HF) patients (SCD-Heft 1 and COMPANION studies 2 ). In HF patients with prolonged QRS duration, cardiac resynchronization therapy (CRT) reduces symptoms and the risk of death (CARE-HF study 3 ). In the COMPANION study, CRT with ICD (CRT-D) or without ICD therapy demonstrated the benefit on mortality. The extension phase of the CARE-HF study supports the benefit of resynchronization therapy alone on sudden cardiac death. Patients treated with CRT had a decreased incidence of sudden death and death due to worsening HF, both of which reflecting an improvement in cardiac function. Delaying the progression of cardiac dysfunction may prevent malignant arrhythmias. There is no pre-implantation criterion to discriminate HF patients who could take more benefit from CRT-D than from CRT. * Corresponding author. Tel: þ address: alepillier@hotmail.com Published on behalf of the European Society of Cardiology. All rights reserved. & The Author For permissions please journals.permissions@oxfordjournals.org

2 Relationship between NYHA class change and VT 81 On a large population of HF patients implanted with CRT-D, we retrospectively analysed the correlation between the functional improvement and the occurrence of ventricular tachyarrhythmia (VT). Methods Study design Patients were enrolled between 1999 and 2003 in three European studies concerning efficacy and safety of implantable cardiac defibrillator devices with CRT capability. These studies had similar design and similar inclusion criteria: InSync ICD ending in September 2001, InSync Marquis in March 2004, and InSync III Marquis in August The devices were Medtronic InSync ICD 7272, InSync Marquis 7277, and InSync III Marquis We pooled these three studies and retrospectively analysed the relationship between New York Heart Association (NYHA) class improvement and the occurrence of VT. Recruitment The inclusion criteria were similar for the three pooled studies: all patients had to be at least 18 years of age and have stable cardiac HF due to ischaemic or non-ischaemic cardiomyopathy with HF history, a left ventricular ejection fraction (LVEF) of,35%, and a QRS interval of at least 130 ms on the electrocardiogram (ECG). Patients had had an episode of asymptomatic, unsustained ventricular tachycardia (a run of 3 30 ventricular ectopic beats at a rate of 120 bpm) or a cardiac arrest due to ventricular arrhythmia or sustained VT. Patients were excluded from consideration for enrolment if one or more of the following conditions were present: ventricular arrhythmia due to a reversible cause, the presence of a mechanical right heart valve, patients with any condition other than cardiac disease that was associated with a reduced likelihood of survival for the duration of the trial, expected poor compliance with the protocol, and patients who were participating in other clinical trials. Implantable cardioverter defibrillator therapy Standard techniques were used to implant CRT-D devices: a standard right atrial pacing lead, a standard right ventricular pacing/defibrillation lead, and a choice of several left ventricular (LV) transvenous leads positioned in a coronary sinus tributary. All devices deliver atria-synchronized biventricular pacing for cardiac resynchronization, anti-tachycardia pacing through right ventricle (RV) or RV and LV leads, and cardioversion and defibrillation to treat VT delivered through the RV lead only. Defibrillation testing was carried out during the implantation procedure. Every effort was made to achieve defibrillation, with a 10 J safety margin. Devices were implanted in prepectoral or retropectoral area. Programming the defibrillator and prescribing medications were left to the discretion of the patients physicians. The appropriate use of b-blockers, angiotensin-converting enzyme inhibitors, and lipid-lowering drugs was encouraged. Data collection Patients meeting the criteria for entry underwent the following evaluations at baseline: age, primary or secondary indication for ICD, atrial fibrillation (AF) history, ischaemic or non-ischaemic aetiology, twodimensional Doppler-flow echocardiography to assess the LVEF, NYHA class, 6 min walking test, the QRS duration (from a 12-lead ECG), and medications. At 6-month follow-up (FU), the following data were collected: NYHA class and functional status change, QRS interval after resynchronization, left ventricular end-systolic and end-diastolic diameters, and the occurrence of ventricular tachyarrhythmia/ventricular fibrillation (VT/VF). Ventricular events were collected and classified for rhythm by the investigators who analysed different devices at 6 months. The analysis included all device-detected arrhythmias that were spontaneous, appropriate (per investigator), and which received a therapy. Recording and classification of ventricular arrhythmia episodes depended on the programmed-detection windows for VT and VF, the number of intervals to detect, and on the programmed therapies, as episodes that terminated spontaneously during charge have not been considered in the analysis. It should be noted that device programming was left at investigators discretion. Statistical analysis We retrospectively analysed different data by comparing two groups: a VT/VF group including all patients who experienced VT during FU and a no VT/VF group including patients free of ventricular event. Continuous values were presented as mean and standard deviation. These values were analysed using Student s t-test. Where appropriate, a paired analysis was used. Dichotomous results were analysed with Fisher s exact test. New York Heart Association class and NYHA class change is compared between patient groups using a Cochran Mantel Haenszel test for ordinal variables. A P-value of less than 0.05 was used to assess statistical significance. Results Study population Three hundred and thirty-two patients were included in the study. Baseline characteristics are displayed in Table 1: most of the patients were male, years, a history of AF was present in 31.9%, NYHA class II was observed in 23%, class III in 65%, and class IV in 11%, primary prevention concerned 70%, 55% were ischaemic cardiomyopathy, LVEF was %, and QRS width was ms. At 6-month FU, a total of 68 patients (20.5%) experienced 1266 spontaneous episodes of VT/VF. Ventricular arrhythmias were classified into three groups based on programmed VT/VF intervals: 1025 episodes of ventricular tachycardia, 140 episodes of fast VT, and 101 episodes of VF. Baseline characteristics of the VT/VF (n ¼ 68) and no VT/VF groups (n ¼ 264) are shown in Table 2. There was no statistical difference in baseline characteristics between patients with or without VT/VF, except for ICD indication (primary or secondary prevention). The number of patients with ischaemic aetiology (52.94% in the VT/VF group vs. 55.7%, P ¼ 0.68) or a history of myocardial infarction (50.0% in the VT/VF group vs %, P ¼ 0.59) was similar in the two groups. There was no statistical difference in AF history (29.44% in the VT/VF group vs. 32.6%, P ¼ 0.66). There was no statistical difference in LVEF ( % in the VT/VF group vs %, P ¼ 0.35) and in QRS duration at baseline ( ms in the VT/VF group vs ms, P ¼ 0.055). The distribution in NYHA class was similar

3 82 A. Lepillier et al. Table 1 Baseline characteristics of the patients Characteristics Total (n 5 332)... Male sex (n, %) 287 (86.4%) Age (years, mean + SD) Primary prevention (n, %) 232 (69.9%) Ischaemic aetiology (n, %) 183 (55.1%) Prior myocardial infarction (n, %) 178 (53.64%) History of atrial fibrillation 106 (31.9%) Left ventricular ejection fraction (%, mean + SD) NYHA class (n,%) a I 3 (0.9%) II 77 (23.3%) III 216 (65.3%) IV 35 (10.6%) QRS duration (ms, mean + SD) Six-minute walking test (m, mean + SD b Medication at baseline (n,%) Diuretics 302 (91.0%) b-blockers 203 (61.1%) Angiotensin-converting enzyme inhibior/ 281 (84.6%) Angiotensin receptor blocker Anti-arrhythmic 157 (47.3%) Anticoagulant/antiplatelet 221 (66.67%) Digoxin 145 (43.74%) Lipid lowering 104 (31.3%) a Baseline New York Heart Association missing for one patient. b Only collected in 68 patients. in the two groups (P ¼ 0.63). There was no significant difference in medication, especially in the prescription of b-blockers. Endpoint analysis Follow-up NYHA class is available from 292 patients. Improvement in NYHA class at last visit was observed in 159 patients (54.5%). The mean functional class improvement was for all patients. Improvement in NYHA class during FU was significantly associated with a decreased occurrence of VT/VF (P ¼ 0.004): patients with more improvement have fewer episodes (Figure 1). In the 30 patients who improved their NYHA class by 2 or 3 classes, only 2 patients experienced ventricular arrhythmia events (6.7%). In the 129 patients who improved in only 1 class, 23 have events (17.8%). In the 117 patients who did not improve their NYHA class, ventricular arrhythmias occurred in 28 (23.9%). In the 16 patients who decreased their functional status of 1 to 2 classes, 7 of them (43.8%) experienced VT. Additional analysis Six-minute walking test With available data for 76 patients, the mean distance walked at 6 months was m in the VT/VF group and m in the no VT/VF group (P ¼ 0.11). The mean improvement at 6-month FU of the 6 min walking test was similar in the two groups ( m vs m, P ¼ 0.33). QRS duration In the analysis of QRS width for 79 patients, the mean QRS duration after resynchronization was ms in the VT/VF group and ms in the no VT/VF group (P ¼ 0.67). When compared with the VT/VF group, there was a trend towards more reduction in QRS width in the patients who did not have any VT episodes ( vs ms, P ¼ 0.07) (Figure 2). Echocardiographic data Echocardiographic data available for 67 patients were similar in the two groups. The mean reduction of the end-diastolic left ventricle diameter (LVED) was mm after CRT. We did not find any relationship between LVED and the occurrence of VT: a reduction in end-diastolic diameter was mm in the VT/VF group and mm in the no VT/VF group (P ¼ 0.53), and the reduction in LVED was similar (P ¼ 0.81). Deaths At 6-month FU, 16 patients died (5%), mainly due to HF. Of these 16 patients, 8 of them had experienced ICD therapy for VT (50%). In contrast, of the 256 survivors, only 60 experienced ventricular events with appropriate ICD therapy (19%). Sixteen patients who died had significantly more often VT/VF than the survivors (P ¼ 0.007). Mortality is strongly associated with the occurrence of appropriate ICD therapy in this population (Figure 3). Patients with the most severe HF, especially those who do not benefit from CRT therapy, have a high risk of episodes and a high risk of death. Discussion Cardiac resynchronization has recently become an established therapy for patients with NYHA class III/IV systolic HF and prolonged QRS interval. 4 In addition to improvement in exercise capacity and NYHA class symptoms, evidence of reversal of ventricular remodelling has been demonstrated. A recent meta-analysis 5 found that CRT reduces mortality from progressive HF and suggests a trend towards longer survival in patients with CRT. Total mortality was also reduced in recent trials. The COMPA- NION trial included patients treated by CRT alone and CRT-D. These two groups were compared with a control group with medical therapy alone, and not against each other. This left unanswered the issue of whether CRT-D is superior to a CRT. The CARE-HF study showed significant long-term benefits of CRT alone on mortality. In addition, the extension phase of CARE-HF showed a significant reduction of sudden death in CRT patients. This suggests an anti-arrhythmic effect of CRT. The results of our study are strongly in favour of this hypothesis: a significant relationship between NYHA class improvement by CRT and a decrease in the occurrence of VT is shown at 6-month FU. There are few experimental studies that examined direct antiarrhythmic effects of CRT. Restivo et al. 6 published an experimental study that demonstrated this anti-arrhythmic effect of applying dual site ventricular stimulation in the canine post-infarction model. In the last few years, several clinical reports have strongly

4 Relationship between NYHA class change and VT 83 Table 2 Comparison of baseline characteristics between ventricular tachyarrhythmias and no ventricular tachyarrhythmias groups Characteristics Ventricular No ventricular P value tachyarrhythmias tachyarrhythmias group n 5 68 group n Male sex (n, %) 61 (89.7%) 226 (85.6%) 0.43 Age (years, mean + SD) 65.5 (+10.6) 65.1 (+9.4) 0.77 Primary prevention (n, %) 59 (86.8%) 173 (65.5%),0.001 Ischaemic aetiology (n, %) 36 (52.94%) 147 (55.7%) 0.68 Prior myocardial infarction (n, %) 34 (50.0%) 144 (54.54%) 0.59 History of atrial fibrillation (n, %) 20 (29.4%) 86 (32.6%) 0.66 Left ventricular ejection fraction (%, mean + SD) 25.6 (+8.1) 24.6 (+7.4) 0.35 New York Heart Association class (n,%) I 0 (0.0%) 3 (1.1%) 0.63 II 19 (27.94%) 58 (22.00%) III 42 (61.8%) 174 (65.9%) IV 7 (10.3%) 28 (10.6%) QRS duration (ms, mean + SD) 161 (+28.3) 169 (+32.6) Medication (n) Diuretics 62 (91.2%) 240 (90.9%) 1.00 b-blockers 36 (52.94%) 167 (63.3%) 0.13 Angiotensin-converting enzymes inhibitors/angiotensin 61 (89.7%) 220 (83.3%) 0.26 receptor blocker Anti-arrhythmias 34 (50.0%) 123 (46.6%) 0.68 Anticoagulant/antiplatelet 48 (70.6%) 173 (65.5%) 0.47 Digoxine 25 (36.8%) 120 (45.5%) 0.22 Lipid lowering 23 (33.8%) 81 (30.7%) 0.66 Figure 1 Relationship between New York Heart Association class improvement and the occurrence of ventricular tachyarrhythmias. suggested that cardiac structural and contractile reverse remodelling following CRT could also result in a favourable anti-arrhythmic effect. The decrease in the incidence of VT after CRT was suggested in a combined analysis of MIRACLE ICD and CONTAK-CD. 7 Higgins et al. 8 showed that 32 patients with CRT-D received less ICD therapy during a 3-month period with Figure 2 QRS width shortening in heart failure patients treated with CRT-D and the occurrence of ventricular tachyarrhythmias. CRT on than with CRT off. In a study of Arya et al., 9 resynchronization therapy was associated with decreased incidence of ventricular arrhythmia and a more delayed occurrence of the first appropriate therapy. However, Lin et al. 10 showed that upgrading

5 84 A. Lepillier et al. observed a relationship between the occurrence of appropriate ICD therapy and the death due to HF. A recent analysis of the COMPANION trial 22 demonstrated that an appropriate shock therapy was associated with a worse outcome. An appropriate shock was experienced in 88 (15%) of the 595 CTR-D patients. The risk of the occurrence of ICD therapy increases with time after device implantation: 11.6% in the first year and 19.3% at 2 years. The occurrence of appropriate shock is a marker of death from any cause, HF, and sudden death. In the MADIT II trail, 23 first appropriate ICD therapy identifies more than a three-fold increase in mortality risk and a two-fold increase in the risk of HF exacerbation and hospitalization. Figure 3 Relationship between mortality and the occurrence of ventricular tachyarrhythmias. patients already implanted by ICD to CRT-D were not associated with the reduction of ventricular arrhythmia incidence at 6-month FU. The ability to induce VT was reduced in patients with significant LV reverse remodelling after 6 months of implanting a CRT device. 11 Patients with lost inducability had a significantly larger reduction in LV end-diastolic and end-systolic volume compared with those who remained inducable. The arrhythmogenic substrate in patients with depressed LV systolic function and HF included dispersion of repolarization, 12 calcium channel alterations, neurohumoural depression, myocardial ischaemia, and genetic predispositions. Berger et al. 13 show in 25 HF patients, using a high resolution surface ECG, that bi-ventricular pacing resulted in a reduction of ECG markers of ventricular dispersion of repolarization. In patients treated with CRT, ECG repolarization indices are related to pacing-induced activation sequences rather than transmural dispersion of repolarization. 14 A positive microvolt T-wave alternans is considered an indirect marker of increased dispersion of repolarization. 15 The value of microvolt T-wave alternans is controversial. The ABCD trial 16 realized in 566 patients treated with ICD showed that microvolt T-wave alternans was a strong predictor of malignant VT and sudden cardiac death. On the contrary, the MASTER trial 17 realized in 575 ICD patients did not find any statistical difference. In a study of Turitto et al., 18 the prevalence of T-wave alternans was investigated in a group of patients who received CRT. Cardiac resynchronization therapy was associated with amelioration of all T-wave alternans indices. Ehrlich et al. 19 recently showed that bi-ventricular pacing does not affect microvolt T-wave alternans in HF patients. Several hypotheses can account for anti-arrhythmic effects of CRT, mainly related to ventricular remodelling. The reduction of QRS width is probably a marker of ventricular remodelling. 20 Molhoek et al. 21 demonstrated in 61 patients that responders had a significant reduction of QRS width, and there is a correlation between the QRS shortening and the response to CRT. In our study, we found that the shortening of QRS width seems to be associated with a decrease of VT occurrence. Moreover, we Conclusion In our study, there is a significant decrease of VT/VF occurrence in HF patients implanted with a CRT-D with NYHA class improvement. This result suggests an anti-arrhythmic effect of CRT that could be related to reverse remodelling. In these HF patients with CRT-D, a higher rate of appropriate therapy was observed in patients who died due to HF during FU. No predictor of VT/ VF occurrence was found in our population to further help to indicate CRT alone or CRT-D in such patients. However, the higher morbidity and the higher cost of CRT-D compared with CRT should be taken into account. To confirm our results, a controlled randomized study should be designed. However, as mentioned in the CARE-HF study, assuming that the combination of CRT and ICD could prevent two-thirds of sudden deaths, a future study would require 1300 patients per group and an FU period of at least 2 years to have significant statistical power to detect a reduction in the risk of death. Acknowledgements We would like to thank the ADREC team (Association Dyonisienne de Recherche et d Enseignement Cardiologique) for its support in the preparation of the manuscript. Conflict of interest: none delcared. References 1. Bardy GH, Lee KL, Mark DB, Poole JE, Packer DL, Boineau R et al. For the Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT) investigators. Amiodarone or an implantable cardioverter-defibrillator for congestive heart failure. N Engl J Med 2005;352: Bristow MR, Saxon LA, Boehmer J, Krueger S, Kass DA, De Marco T et al. For the Comparison of Medical Therapy, Pacing, and Defibrillation in Heart Failure (COMPANION) investigators. Cardiac resynchronization therapy with or without an implantable defibrillator in advance chronic heart Failure. N Engl J Med 2004;350: Cleland JG, Daubert JC, Erdmann E, Freemantel N, Gras D, Kappenberger L et al. for the Cardiac Resynchronization-Heart Failure (CARE-HF) study investigators. The effect of cardiac resynchronization on morbidity and mortality in heart failure. N Engl J Med 2005;352: Aliot E, Chauvin M, Daubert JC, Franck R, Jondeau G, Leclercq JF et al. Indications du défibrillateur automatique implantable ventriculaire: Mise à jour de la version française. Arch Mal Coeur Vaiss 2006;99: Bradley DJ, Bradley EA, Baughman KL, Berger RD, Calkins H, Goodman SN et al. Cardiac resynchronization and death from progressive heart failure. A meta-analysis of randomized controlled trials. JAMA 2003;289:

6 Relationship between NYHA class change and VT Restivo M, Gough WB, El Sherif N. Reentrant ventricular rhythms in the late myocardial infarction period: prevention of reentry by dual stimulation during basic rhythm. Circulation 1988;77: McSwain RL, Schwartz RA, DeLurgio DB, Mera FV, Langberg JJ, Leon AR. The impact of cardiac resynchronization therapy on ventricular tachycardia/fibrillation: an analysis from the combined Contak-CD and InSync-ICD studies. J Cardiovasc Electrophysiol 2005;16: Higgins SL, Yong P, Scheck D, Martin McDaniel M, Bollinger F, Vadecha M et al., for the Ventak CH investigators. Biventricular pacing diminishes the need for implantable cardioverter defibrillator therapy. J Am Coll Cardiol 2000;36: Arya A, Haghjoo M, Dehghani MR, Alasti M, Alizadeh H, Kazemi B et al. Effect of cardiac resynchronization therapy on the incidence of ventricular arrhythmias in patients with an implantable cardioverter-defibrillator. Heart Rhythm 2005;2: Lin G, Rea RF, Hammill SC, Hayes DL, Brady PA. Effects of cardiac resynchronisation therapy on occurrence of ventricular arrhythmia in patients with implantable cardioverter defibrillators undergoing upgrade to cardiac resynchronisation therapy devices. Heart 2008;94: Kiès P, Bax JJ, Molhoek SG, Bleeker GB, Zeppenfeld K, Bootsma M et al. Effect of cardiac resynchronization therapy on inductibility of ventricular tachyarrhythmias in cardiac arrest survivors with either ischemic or idiopathic dilated cardiomyopathy. Am J Cardiol 2005;95: Tomaselli GF, Zipes DP. What causes sudden death in heart failure? Circ Res 2004; 95: Berger T, Hanser F, Hintringer F, Poelzl G, Fischer G, Modre R et al. Effects of cardiac resynchronization therapy on ventricular repolarization in patients with congestive heart failure. J Cardiovasc Electrophysiol 2005;16: van Huysduynen BH, Swenne CA, Bax JJ, Bleeker GB, Draisma H van Erven L et al. Dispersion of repolarization in cardiac resynchronization therapy. Heart Rhythm 2005;2: El Sherif N, Turitto G, Pedalino R, Robotis D. T-wave alternans and arrhythmia risk stratification. Ann Noninvasive Electrocardiol 2001;6: Costantini O. The Alternans Before Cardioverter Defibrillator (ABCD) trial: a non-invasive strategy for primary prevention of sudden cardiac death using T wave alternans. Late breaking clinical trials III, American Heart Association 2006, Scientific Sessions, 15 November Chow T. Primary results from the microvolt T-wave alternans testing for risk stratification of post MI Patients (MASTER I) trial. Late-breaking clinical trials III, American Heart Association 2007, Scientific Sessions, 7 November Turitto G, Houy S, Gupta R, El Sherif N. Right ventricular but not biventricular pacing increases markers of ventricular electrical instability. J Am Coll Cardiol 2004;43:149A. 19. Ehrlich JR, Wegener FT, Anneken L, Duray G, Israel CW, Hohnloser SH. Biventricular pacing does not affect microvolt T-wave alternans in heart failure patients. Heart Rhythm 2008;5: Vogt J, Krahnefeld O, Lamp B, Hansky B, Kirkels H, Minami K et al. Electrocardiographic remodelling in patients paced for heart failure. Am J Cardiol 2000;86: Molhoek SG, Van Erven L, Bootsma M, Steendijk P, Van Der Wall EE, Schalij MJ. QRS duration and shortening to predict clinical response to cardiac resynchronization therapy in patients with end-stage heart failure. Pacing Clin Electrophysiol 2004;27: Saxon LA, Bristow MR, Boehmer J, Krueger S, Kass DA, De Marco T et al. Predictors of sudden cardiac death and appropriate shock in the Comparison of Medical Therapy, Pacing, and Defibrillation in Heart Failure (COMPANION) trial. Circulation 2006;114: Moss AJ, Greenberg H, Case RB, Zareba W, Hall WJ, Brown MW et al. Multicenter Automatic Defibrillator Implantation Trial-II (MADIT-II) research group. Longterm clinical course of patients after termination of ventricular tachyarrhythmia by an implanted defibrillator. Circulation 2004;110: