Medications Are Important for Sudden Cardiac Death Prevention But So Is the Implantable Cardioverter-Defibrillator^∗Free Access

Introduction

Patients with heart failure with reduced ejection fraction (HFrEF) are at an increased risk of sudden cardiac death (SCD) (1). As shown by several randomized clinical trials, this risk can be significantly reduced with an implantable cardioverter-defibrillator (ICD) (1). However, because of the protective effect several HF medications have against SCD and their potential to improve the EF, clinicians should optimize HF medical therapy before an ICD is considered. Indeed, the 2017 American Heart Association/American College of Cardiology/Heart Rhythm Society guideline for the prevention of SCD highlights the important role of HF medications in reducing SCD through the following Class I recommendation: “In patients with HFrEF ([left ventricular ejection fraction] LVEF ≤40%), treatment with a beta blocker, a mineralocorticoid receptor antagonist and either an angiotensin-converting enzyme inhibitor, an angiotensin-receptor blocker, or an angiotensin receptor-neprilysin inhibitor is recommended to reduce SCD and all-cause mortality.” In this and other professional guidelines, recommendations related to primary prevention ICDs are predicated on the persistence of an LVEF ≤35% and New York Heart Association functional class II or III HF symptoms, despite the use of guideline-directed medical therapy (1).

Notwithstanding evidence supporting the survival benefit of primary prevention ICDs in patients with HFrEF and professional guidelines that recommend their use, the role of these devices in such patients has been questioned lately due partly to ongoing improvements in HF medical therapy. It is clear that guideline-directed medical therapy in patients with HFrEF decreases the risk of SCD and all-cause mortality. In fact, sacubitril/valsartan therapy appears to reduce episodes of ventricular arrhythmia and appropriate ICD shocks likely through its role in reversing remodeling and attenuating myocardial fibrosis (2,3). Importantly, however, no study has ever shown that ICD implantation in patients with HFrEF could be forgone in the setting of optimal medical therapy even in contemporary clinical practice.

In this issue of JACC: Heart Failure, Rohde et al. (4) performed a secondary analysis of the PARADIGM-HF (Prospective Comparison of ARNI with ACEI to Determine Impact on Global Mortality and Morbidity in Heart Failure) trial to determine rates of ICD use by geographic region and to examine the association between ICD and SCD and between sacubitril/valsartan and SCD by ICD use and eligibility and cause of HF. Of the 8,399 patients studied, 7,145 patients (85%) were deemed eligible for a primary prevention ICD. Of these ICD-eligible patients, only 1,156 (16%) had an ICD. Use of ICD was highest in North America (56%) and lowest in Asia-Pacific (1.8%) regions. After propensity score adjustment, use of ICD was associated with a 56% relative risk reduction in SCD among ICD eligible patients; this association seemed to hold true for patients with ischemic cardiomyopathy as well as those with nonischemic cardiomyopathy. An inverse relationship between use of ICD use and SCD was observed, with North America having the lowest SCD rates and Asia-Pacific having the highest SCD rates. Interestingly and not surprisingly, use of ICD was associated with altering the mode of death from SCD to HF death, regardless of cause of HF. Finally, sacubitril/valsartan therapy was found to be associated with a 51% relative risk reduction in SCD in patients with an ICD and a 19% relative risk reduction in SCD in patients eligible for ICD who did not have an ICD (4).

The authors are to be commended for this well-performed and informative analysis. Strengths of the study include robust data collection; central adjudication of SCD, for example, borderline SCD cases such as patients with presumed sudden death were excluded from the analysis; and rigorous data analyses. However, several issues should be emphasized. First, despite the authors’ assertions that the association between sacubitril/valsartan therapy and reduced risk of SCD was true for both patients with nonischemic and for patients with ischemic cardiomyopathy, their data do not fully support these statements, as all p values for comparisons between sacubitril/valsartan and enalapril in patients with ischemic cardiomyopathy were >0.30. However, this may be the case given the result of interaction analyses showing no interaction by cause of HF. Second, the authors proposed a potential framework to try to explain differences in their results between patients with ischemic cardiomyopathy and those with nonischemic cardiomyopathy. They stated that the antiremodeling effect of sacubitril/valsartan may be especially beneficial in patients with nonischemic cardiomyopathy who tend to have “mechanically-mediated events” such as episodes of bradyarrhythmia whereas “electrically-mediated events” that tend to be more prevalent in patients with ischemic cardiomyopathy can best be treated with an ICD. However, as the authors caution, these views are mere speculations that may not be true, especially given the prior analyses of the PARADIGM-HF trial that found no differences between the effects of sacubitril/valsartan on clinical outcomes by cause of HF. Third, although we laud the authors for attempting to find an explanation for the association between sacubitril/valsartan and a 51% relative risk reduction in SCD in patients with an ICD versus a 19% relative risk reduction in SCD in ICD-eligible patients who did not have an ICD, data are needed on if and how sacubitril/valsartan may affect mechanisms of SCD other than ventricular arrhythmia. Research studies to generate such data should be proposed. Fourth, data were not provided for cardiac resynchronization therapy, which can be effective at reversing remodeling in selected patients and has been shown to reduce the risk of SCD. Other issues include references to causality when the authors could only report on associations; lack of information for reasons for ICD noneligibility; and absence of data for who received an ICD during follow-up.

An important issue in the paper by Rohde et al. (4) is their recommendation to allow a longer waiting period for optimization of HF medications for SCD risk reduction before implantation of an ICD. They argued this approach may be particularly appropriate in patients with nonischemic cardiomyopathy, especially in light of the results of the DANISH (Danish Study to Assess the Efficacy of ICDs in Patients with Non-ischemic Systolic Heart Failure on Mortality) trial. However, their data do not support this recommendation as sacubitril/valsartan therapy was associated with a greater reduction in SCD in patients with an ICD than in those with no ICD. It should also be pointed out that the rate of use of HF medications and doses achieved in clinical practice are substantially lower than that in clinical trials. Of 2,588 patients with chronic HFrEF in the CHAMP-HF (Change the Management of Patients with HF) registry, <1% of patients were able to tolerate target doses of beta-blockers, renin-angiotensin system inhibitors, and mineralocorticoid receptor antagonists (5). Although many of these patients are receiving maximally tolerated doses of medical therapy, they are not receiving the target doses that have been proven to be beneficial in clinical trials. Therefore, these observations underscore the importance of the ICD in eligible patients.

Despite the limitations of the study by Rohde et al. (4), their paper is important, as it is a good reminder of the pivotal role of HF medications in the prevention of SCD. However, it is also a reminder of the frighteningly low rate of the use of ICDs in patients with HFrEF and the need to develop interventions that effectively increase the use of ICDs in eligible patients globally. Such efforts could be informed by data for factors associated with no use of ICD provided by their study such as female sex and nonischemic cardiomyopathy. Although more data are needed for the role of primary prevention ICDs in contemporaneous patients with HFrEF treated with optimal medical therapy, it is very unlikely that medications alone will be enough, and a combination of best medical therapy with optimal device therapy will continue to be needed.

Footnotes