Rethinking Risk Stratification for Nonischemic HF: What Do the Latest Data Suggest?

Panelists

Program Goals

Introduction

• Considerable recent data showing reduced need for an implantable cardioverter-defibrillator (ICD) with improved medical therapy
  • Introduction of mineralocorticoids as a major therapy for heart failure (HF)
  • Introduction of neprilysin inhibitors
  • Improved use of guideline-directed medical therapies, such as β blockers, has reduced the likelihood of sudden cardiac death (SCD)
• Using left ventricular ejection fraction (LVEF) as the only stratifier for deciding whether to use an ICD is becoming obsolete
• Other risk stratifiers are needed

The SCD Conundrum^[1]

• Because LVEF <35% is the marker used for risk of SCD in clinical trials, it is now widely used in clinical practice
  • Somewhat arbitrary
  • Reflects a point of benefit for high-risk patients in early clinical trials
• HF is a syndrome with various underlying pathologies
  • Even the terms ischemic and nonischemic can be complicated
• Traditionally, obstructive coronary artery disease (CAD) is used as a marker for ischemic cardiomyopathy
  • Yet some patients have a myocardial infarction (MI) but no obstructive CAD
    • Some patients have obstructive CAD but lack the classical right coronary lesion with dilated cardiomyopathy
    • Underlying substrate may be a mixture
    • Most trials divide patients into nonischemic dilated cardiomyopathy (DCM), ischemic, or post-MI HF, thus the indication for an ICD in nonischemic DCM is perceived as worse than for patients with ischemic heart disease

SCD-HeFT Trial^[2]

• SCD-HeFT trial
  • Reference for much primary prevention data
  • Included patients with ischemic and nonischemic cardiomyopathy
  • Showed benefit for the entire study population; no heterogeneity as to ischemic vs nonischemic disease
• Clinical practice guidelines recommend ICD for primary prevention for patients with ischemic and nonischemic cardiomyopathy
  • More patients with ischemic than nonischemic cardiomyopathy in SCD-HeFT

DANISH Trial^[3]

• Patients received high quality medical therapy and HF care
• Maximized therapy not always achievable in practice, but is always a goal
• Decision to use cardiac resynchronization therapy (CRT) device was made before the trial
• DANISH was well conducted, but some issues around the study will raise questions
• Evidence base for nonischemic cardiomyopathy may not be as great as previously thought

Limitations of DANISH^[3-5]

• Leaves the clinician and patients at a point of therapeutic equipoise
• Many interpret DANISH as a reason not to use ICDs
• DANISH trial may not represent patients typically seen in practice
  • New-onset HF
  • Higher risk
  • Less stable medical regimens
• High target medication doses defined as optimal for treatment of HF
  • Metoprolol 200 mg once daily
  • Losartan 100 mg once daily
  • Captopril 50 three times daily in a combination regimen
• Danish clinical trial populations are typically highly disciplined, and many patients enroll

Reducing Residual Risk With ICD^[3]

• Risk of all-cause mortality and SCD substantially reduced on high-dose regimens
• In reality, a well-intentioned patient trying to get to 25 mg twice daily of carvedilol may only reach 6.25 mg
• DANISH provided very high quality evidence-based therapy for HF over 2 years, which is hard to do in real-world practice
  • Medical therapy for HF has changed since the first trials were conducted
  • Many patients seen in practice have newly diagnosed HF
    • On medications for some period of time
    • Referred following a catheterization or stress test
• DANISH results: first indication of how much can be done medically for patients with HF
• Difficult to conduct a trial like DANISH that runs against Class 1 guideline recommendations
• Prerandomization screening data not published
  • Total number screened to randomize 1106 patients is unknown
  • Potential for selection bias
• Patients not chosen to enroll in DANISH are more representative of real-world practice

Impact of DANISH Trial^[3,6]

• The misconception that ICD is not useful for people with nonischemic dilated cardiomyopathy (DCM) and EF of 35% is a real problem
• Clinicians must consider ICD in terms of value proposition
  • Primary function of ICD is to reduce SCD
  • Nearly every drug trial, regardless of agent, showed a reduction in all-cause mortality
  • Half of all-cause mortality is due to reduction in SCD

RALES Trial^[7]

• Clinicians need to reframe their thinking: taking appropriate medical therapy for HF is "like swallowing your ICD"
• ICD is an expensive alternative to medical therapy
  • Getting patients on the DANISH regimen, including the very high doses, is imperative for the clinician
  • Whether it can be done in real-world practice is hard to know
• It is important to appreciate that the results of DANISH were unexpected among electrophysiologists

Guidelines for ICD in DCM^[8,9]

• Guidelines based on older trials
• US and European guidelines for use of ICD in nonischemic DCM are fairly consistent
• In Europe, other indications for familial conditions are included
  • In DCM, lamin A/C (LMNA) mutation in patients with positive family history
  • Other indications not well developed

Role of the WCD^[3]

• The DANISH study population was very stable with respect to sudden death
  • Mean time to enrollment 1.7 years
  • SCD event rate: 1.6 per 100 patients per year
• In such a stable population, ICD might not be of benefit
• About 45% of patients have improvement in LVEF with medication
• A wearable cardioverter-defibrillator (WCD) could be used during the initial period on medication, giving time to evaluate the patient's residual risk

Methods of Risk Stratification^[3,6,10-12]

• There has always been interest in better aligning patients to therapies
  • Many patients who have SCD do not have a low ejection fraction (EF)
  • Many patients with low EF never had the device go off
• Signal-averaged electrocardiogram (ECG): issues with reproducibility
• Biomarkers have been evaluated in some HF trials
  • Elevated N-terminal pro-B-type natriuretic peptide (NT-proBNP) an inclusion criterion of the DANISH trial
• Cardiac magnetic resonance imaging (MRI) may assist in risk stratification
  • If half of the deaths are SCD, what are the other half?
  • Gadolinium is an ethylenediaminetetraacetic acid (EDTA) molecule, thus stays outside of the cells
    • Visualized on MRI as a voxel, or mass of tissue
  • Has been evaluated in sarcoidosis, idiopathic cardiomyopathy, ischemic MI, and other conditions

Cardiac MRI^[13]

• Evidence suggests that patients with nonischemic cardiomyopathies are not all the same
• Sarcoidosis is a relatively uncommon condition with a specific pattern on cardiac MRI
  • More prevalent in the southeastern United States
  • More prevalent in women than in men

Cardiac MRI (cont)^{[3, 12]}

• Cardiac MRI useful in looking for patterns
  • Subendocardial infarction in the left anterior descending (LAD) distribution indicates LAD infarct
  • 10% of myocardium may indicate risk of defibrillation
  • Amyloid also has a subendocardial pattern
• Using cardiac MRI before defibrillator implantation is an area of study
• Cardiac MRI is an expensive test, used in conjunction with clinical therapy and biomarkers
• Has been used to predict HF but not to determine the use of ICDs

Biomarkers in DANISH^[3]

• How HF is managed will predict future events, but it is not an easy task
• There is need for very objective tests

GUIDed Trial^[14]

• Cardiac MRI with late gadolinium enhancement (LGE) is not a standard test, but reproducibility has improved over the years
• For patients in the moderate group with LVEF between 35% and 50%, there is a gray zone of not knowing how to manage them
  • Imaging with MRI and then randomly assigning patients to monitoring or defibrillation would be valuable
• GUIDed trial, ongoing in Australia, is using cardiac MRI
• Fits current understanding that scar leads to heterogeneity of the cell population that becomes the substrate for arrhythmias
• Emerging data suggest that without this LGE finding, risk of arrhythmia is extremely low
• It is not always clear why people die suddenly, thus GUIDed is a powerful trial because it will reveal exactly what happened, particularly in the control group

Implications for Clinical Care^[15,16]

• For ischemic patients,clinicians are comfortable in management strategies
• For nonischemic, most patients still candidates for ICD if LVEF is still <35% after treatment with reasonably good medical therapy
• Consider a hypothetical patient who wore the life vest after 30 days, received an optimal, high-dose DANISH regimen, but still has LVEF of 30%
  • Applying the Seattle HF model to DANISH data shows that there is still very significant all-cause mortality reduction associated with implanting an ICD, even above this optimized, high-dose regimen

Summary

• Absolute rate of sudden death among patients with intermediate LVEF is small and cannot be used as an indication for ICD
• WCD could be used during the initial period on medication, giving time to evaluate the patient's residual risk

Thank you