Who Benefits From a Defibrillator—Balancing the Risk of Sudden Versus Non-sudden Death

Current Heart Failure Reports (2018) 15:376–389 https://doi.org/10.1007/s11897-018-0416-6 DEVICES (C VELTMANN, SECTION EDITOR) Who Benefits From a Defibrillator—Balancing the Risk of Sudden Versus Non-sudden Death Simon A. S. Beggs 1 & Roy S. Gardner 2 & John J. V. McMurray 1 Published online: 10 November 2018 # The Author(s) 2018 Abstract Purpose of Review Treatment with a defibrillator can reduce the risk of sudden death by terminating ventricular arrhythmias. The identification of patient groups in whom this function reduces overall mortality is challenging. In this review, we summarise the evidence for who benefits from a defibrillator. Recent Findings Recent evidence suggests that contemporary pharmacologic and non-defibrillator device therapies are altering the potential risks and benefits of a defibrillator. Summary Who benefits from a defibrillator is determined by both the risk of sudden death and the competing risk of other, nonsudden causes of death. The balance of these risks is changing, which calls into question whether historic evidence for the use of defibrillators remains robust in the modern era. Keywords Defibrillator . ICD . Sudden death . Competing risk Introduction In 1980, Mirowski et al. reported the first three patients treated with an automatic implantable defibrillator (AID) [1]. This early AID was a ‘shock box’, delivering only high-voltage defibrillation therapy. Later devices evolved to include a pacing function and are now usually described as implantable cardioverter defibrillators (ICDs). The randomised controlled trials (RCTs) that followed examined the efficacy of ICD therapy in populations targeted for their high incidence of sudden cardiac death (SCD). Specifically, these were patients with prior cardiac arrest, ventricular arrhythmias (VA) or myocardial infarction (MI), or the presence of coronary artery disease (CAD), left ventricular systolic dysfunction (LVSD), heart failure or some combination of these problems. Incrementally, these RCTs established our current understanding of which patients are more, or less, likely to benefit from an ICD. The importance of several major themes that This article is part of the Topical Collection on Devices * John J. V. McMurray 1 British Heart Foundation Cardiovascular Research Centre, University of Glasgow, Glasgow, Scotland 2 Golden Jubilee National Hospital, Clydebank, Scotland progressively emerged from these trials must be recognised if a thorough examination of this evidence is undertaken. These include the role of competing risks of death, appropriate trial design—including length of follow-up and the selection of a suitable primary endpoint—the technical evolution of the defibrillator, and incremental advances in prognostic pharmacologic and additional device therapy. Who Benefits From a Secondary Prevention Defibrillator? The benefit of an ICD as a secondary prevention treatment was examined in three major prospective, multi-centre RCTs. Summary characteristics of these studies are presented in Table 1. The first published and largest study was the Antiarrhythmics Versus Implantable Defibrillators (AVID) trial [2•]. Patient eligibility in AVID required one of the following: (1) resuscitation from ventricular fibrillation (VF), (2) syncope with sustained ventricular tachycardia (VT) or (3) sustained VT with significant associated symptoms and left ventricular ejection fraction (LVEF) ≤ 40%. AVID enrolled 1016 patients, randomising them to treatment with an ICD or antiarrhythmic drugs (AADs), primarily amiodarone. Over a mean follow-up of 18 months, there were 80 and 122 deaths from any cause in the ICD and AAD arms, respectively N/A N/A 288** 220 57 (34) 56 80 (230) 16 (47) NR 46 (18) 45 (125) 33 (96) NR 0 ICD vs amio. vs BB ACM 44/56 100 5.8 (11) 6.1 (6) N/A N/A 1016 193 (19.0) 18.2 (12.2) 65 (10.5) 80 (808) 9.5 (97) NR 31 (13) 68.5 (680) 29.4 (292) NR 0 ICD vs antiarrhythmic ACM 93/5 100 18.9 (96)† 25.7 (130) LVEF cut-off NYHA criteria No. randomised No. (%) with NICM Follow-up, mean (SD), month Demographics Age, mean (SD), year Male, % (No.) NYHA class III/IV, % (No.) Duration of CHF, mean LVEF, mean (SD), % Medications at baseline, % (No.) ACE inhibitor/ARB BB MRA CRT Design Primary endpoint Defibrillator Transvenous/epicardial, % Internal validity Follow-up, % Crossovers to ICD, % (No.) Crossovers to control % (No.) 100 15.7 (52) 28.1 (92) 84/10 NR 27.4 (181) NR 0 ICD vs amio. ACM 64 (9.6) 84.5 (557) 10.8 (71) NR 34 (14) i) documented VF; OR ii) CA requiring defibrillation iii) susVT + syncope; OR iv) susVT + angina or presyncope if LVEF ≤ 35%; OR v) unmonitored syncope + subsequent spontaneous VT or PVS-induced susVT N/A N/A 659 63 (9.6) 35 CA due to VA i) Resuscitated VF; OR ii) susVT + syncope; OR iii) symptomatic susVT + LVEF ≤ 40% Inclusion criteria CIDS [4] 2000 1990–1997 CASH [3] 2000 1987–1996 AVID [2•] 1997 1993–1997 Secondary prevention trials Outline characteristics of randomised controlled trials utilising defibrilators as primary or secondary prevention therapy Trial name Year of publication Years of enrollment Table 1 ≤ 35% I to IV 900 0 32 (16) ≤ 35% I to III 196 0 27 99 11 (11) 5 (5) 53 / 47 58 (107) 17 (31) NR 0 ICD vs CMT ACM 100 4 (18) 12 (52) 0/100 53 (473) 20 (183) NR 0 ICD vs CMT ACM 64 (9) 16 (759) NR n/a 27 (6) Planned CABG + SAE positive Previous MI (> 3 weeks) + asymptomatic nsVT, EPS+ despite procainamide suppression 63 92 (172) NR NR 26 (7) CABG-Patch [6] 1997 1990 - NR MADIT [5] 1996 1990 - NR Primary prevention trials–ischaemic aetiology Curr Heart Fail Rep (2018) 15:376–389 377 MADIT II [8•] 2002 ≤ 30% I to III 1232 0 20 64 (10) 84 (1040) 29 (355) NR 23 (5) 70 (858) 70 (862) NR 0 ICD vs CMT (3:2) ACM 100/0 NR 4.5 (22) 6 (44) ≤ 40% I to III 704 0 39 67 90 24 NR 30 75 (525) 40 (282) NR 0 EPS+ guided assignment to ICD or AAD, or standard care if negative EPS CA or arrhythmic death NR NR 12 (n/a) n/a LVEF cut-off NYHA criteria No. randomised No. (%) with NICM Follow-up, mean (SD), month Demographics Age, mean (SD), year Male, % (No.) NYHA class III/IV, % (No.) Duration of CHF, mean LVEF, mean (SD), % Medications at baseline, % (No.) ACE inhibitor/ARB BB MRA CRT Design Primary endpoint Defibrillator Transvenous/epicardial, % Internal validity Follow-up, % Crossovers to ICD, % (No.) Crossovers to control % (No.) Previous MI (> 1 month) CAD + PVS-induced susVT Inclusion criteria MADIT II [8•] 2002 1997–2001 MUSTT [7] 1999 1990–1997 1990–1997 1997–2001 Primary prevention trials–ischaemic aetiology MUSTT [7] 1999 Primary prevention trials–ischaemic aetiology Trial name Year of publication Years of enrollment Trial name Year of publication Table 1 (continued) Years of enrollment Table 1 (continued) 99% 0 6 (20) 100/0 ACM 95 (638) 87 (585) NR 0 ICD vs CMT 62 (11) 76 (514) 13 (89) NR 28 (5) ≤ 35% I to III 674 0 30 (13) Recent MI (6–40 days) + CAF+ DINAMIT [9•] 2004 1998–2002 1998–20 (...truncated)