p.38 抽出テキスト

Circulation. 2018;138:e272–e391. DOI: 10.1161/CIR.0000000000000549 September 25, 2018 e309 CLINICAL STATEMENTS AND GUIDELINES

2.	 Small observational studies demonstrated high mortality and frequent appropriate ICD shocks in patients with syncope and NICM.S7.2.1-7–S7.2.1-9 The assumption that malignant VAs are the likely cause of syncope and that the ICD would be protective has recently been challenged. In a subgroup analysis of SCD-HeFT that included 472 patients, the ICD did not reduce either recurrent syncope or the increased risk of mortality associated with syncope.S7.2.1-10 A subgroup analysis of the MADIT-RIT (Multicenter Automatic Defibrillator Implantation Trial-Reduce Inappropriate Therapy) trial found syncope to be arrhythmic only in 39% of patients.S7.2.1-11 These studies suggest that syncope in some HF patients may be an indicator of an end-stage cardiomyopathy associated with a poor prognosis.S7.2.1-11 In a substudy of DEFINITE, inducible sustained VT/VF was found in a minority of patients, but it was associated with appropriate ICD therapy.S7.2.1-14 Another study of electrophysiological testing in NICM found inducible VT/VF in 27.8% of patients, which was associated with future ICD events.S7.2.1-15 In a study of patients with NICM, cardiac mortality correlated with LVEF but not with inducibility on electrophysiological study.S7.2.1-16 Based on these data, many experts are uncomfortable withholding an ICD from patients with NICM who experience syncope potentially due to a VA even if the electrophysiological study shows no inducible sustained VT. 3.	 Access to ICDs may be limited by financial, medical, or personal considerations. In addition, not all patients at high risk of SCD meet ICD indications, such as those with class IV HF without CRT possibility or with a life expectancy  <1 year. A meta-analysis of RCTs, which examined the use of amiodarone for the prevention of SCD, included 15 studies with 8522 patients assigned to amiodarone or placebo/control.S7.2.1-12 Amiodarone reduced the risk of SCD by 29%; however, it did not reduce all-cause mortality and was associated with an increased risk of pulmonary and thyroid toxicity. In a subgroup analysis, the benefit of amiodarone appeared similar in patients with ischemic cardiomyopathy and those with NICM.S7.2.1-12 In a separate meta-analysis,S7.2.1-13 the evidence was insufficent to support amiodarone’s efficacy for reduction of SCD and all-cause mortality in  survivors of cardiac arrest or those with syncope due to VA. A subgroup analysis of the VALIANT (Valsartan in Acute Myocardial Infarction) trial found that amiodarone was associated with increased mortality in patients with NYHA class III HF.S7.2.1-17 These data call for a careful and nuanced approach to using amiodarone for the secondary prevention of SCD in patients with NICM.

7.2.2 Primary Prevention of SCD in Patients With NICM

Recommendations for Primary Prevention of SCD in Patients With NICM References that support the recommendations are summarized in Online Data Supplements 27 and 28.

COR LOE Recommendations

I A 1. In patients with NICM, HF with NYHA class II–III symptoms and an LVEF of 35% or less, despite GDMT, an ICD is recommended if meaningful survival of greater than 1 year is expected.S7.2.2-1–S7.2.2-6

IIa B-NR 2. In patients with NICM due to a Lamin A/C mutation who have 2 or more risk factors (NSVT, LVEF <45%, nonmissense mutation, and male sex), an ICD can be beneficial if meaningful survival of greater than 1 year is expected.S7.2.2-7–S7.2.2-10

IIb B-R 3. In patients with NICM, HF with NYHA class I symptoms and an LVEF of 35% or less, despite GDMT, an ICD may be considered if meaningful survival of greater than 1 year is expected.S7.2.2-5

III: No Benefit C-EO 4. In patients with medication-refractory NYHA class IV HF who are not also candidates for cardiac transplantation, an LVAD, or a CRT defibrillator that incorporates both pacing and defibrillation capabilities, an ICD should not be implanted.

Recommendation-Specific Supportive Text

1.	 For all patients with NICM, it is imperative that patients be on GDMT for HF for at least 3 months before a primary prevention ICD is offered. Four prospective RCTsS7.2.2-1,S7.2.2-2,S7.2.2-5,S7.2.2-6 initially evaluated ICDs for primary prevention of SCD in patients with NICM. TwoS7.2.2-2,S7.2.2-6 were small studies that were terminated early due to a low event rate. In DEFINITE,S7.2.2-5 an ICD reduced the risk of SCD, with a trend toward reduced allcause mortality. SCD-HeFT included 792 NICM patients.S7.2.2-1 Total mortality at 5 years was 27% in the placebo group and 21% in the ICD group (P =0.06). A pooled analysis of these studies demonstrated a significant 31% reduction in all-cause mortality for ICD relative to medical therapy.S7.2.2-4 The DANISH (Danish Study to Assess the Efficacy of ICDs in Patients with Non-ischemic Systolic Heart Failure on Mortality) trialS7.2.2-11 raised questions about the role of primary pevention ICDs in patients with NICM. This trial randomized 1116 patients with NICM LVEF <35% and class II, III, or IV (if CRT was planned) HF to an ICD or no ICD. CRT (either ICD or pacemaker) was present in 58% of patients in the ICD and medical

Downloaded from http://ahajournals.org by on October 7, 2026