ABSTRACT
Background
Current recommendations for a prophylactic (primary prevention) implantable cardioverter defibrillator (ICD) in patients with both ischemic and nonischemic heart failure with reduced ejection fraction (HFrEF) originate from clinical trials conducted in selected patients over 20 years ago that showed an overall statistically significant survival benefit associated with a primary prevention ICD in the range of 23%-34%. The recent introduction of angiotensin receptor–neprilysin inhibitors [ARNI] and sodium glucose co-transporter 2 inhibitors [SGLT2i]) was shown to further reduce the risk of sudden cardiac death (SCD) in patients with HFrEF. Thus, there is an unmet need appropriately designed comparative effectiveness clinical trials aimed to reassess the survival benefit of a primary prevention ICD in contemporary patients with HFrEF.
Methods
The comparative effectiveness of ICD vs non-ICD therapy in contemporary heart failure patients at a low risk for arrhythmic death (CONTEMP-ICD) trial is a prospective, multicenter, open-label, randomized-controlled trial; enrolling 3,290 participants with HFrEF who are treated with optimal stable GDMT and are eligible for a primary prevention ICD, but have a lower predicted risk of life-threatening ventricular tachyarrhythmia (VTA) than nonarrhythmic mortality. Enrolled participants will be randomized to non-ICD vs ICD treatment arms and will be followed over an average period of 3.5 years. The specific aims of the proposed clinical trial are to: (1) Compare the risk of all-cause mortality of non-ICD vs ICD in HFrEF patients who have a lower predicted risk of VTA than nonarrhythmic mortality per the MADIT-ICD Benefit Score; (2) Evaluate whether non-ICD vs ICD is associated with improved survival free of major CV events in patients with HFrEF who are at a lower predicted arrhythmic risk; (3) Assess healthcare utilization and quality of life implications of non-ICD vs ICD management approaches in HFrEF patients who are at a lower predicted arrhythmic risk; and (4) Determine the effect of non-ICD vs ICD management on all-cause mortality in prespecified subgroups.
Conclusions
We hypothesize that, in patients with HFrEF who are at a lower predicted arrhythmic risk, non-ICD vs ICD is noninferior with respect to the primary endpoint of all-cause mortality and superior with respect to the secondary endpoint of survival free of major CV events.
NCT06543446; https://contemp-icd.org .
Background
The main function of an implantable cardioverter defibrillator (ICD) is to protect the patient against sudden cardiac death (SCD) due to ventricular tachyarrhythmia (VTA), by delivering a cardioverting or defibrillating shock and/or anti-tachycardia-pacing (ATP), based on device programming. ,,, During its early implementation, the ICD was used for patients who had survived sudden cardiac arrest or experienced life-threatening VTA. ,, In 1996, our group published the first randomized controlled trial investigating the efficacy of prophylactic (primary prevention) ICD implantation in patients at high risk of VTA but without prior arrhythmic events. Since then, 3 primary prevention clinical trials published in 2002, 2004, and 2005 (MADIT-II, conducted by our group, DEFINITE [survival benefit of ICD in this trial was not statistically significant] and SCD-HeFT ) indicated a mortality reduction in the range of 23%-31% with prophylactic ICD placement in patients with heart failure with reduced ejection fraction (HFrEF), ,, that was similar between patients with ischemic and nonischemic cardiomyopathy (ICM/NICM) (HR = 0.79 and 0.73, respectively). These 3 studies provide the supporting evidence for the current recommendation for a prophylactic ICD in patients with HFrEF who did not experience a prior VTA. ,,,, However, these clinical trials were conducted more than 2 decades ago and improved medical therapy for HFrEF has improved, as well as device programming, leading to very low rates of appropriate ICD therapy for life-threatening VTA in contemporary patients who receive a primary prevention ICD. Furthermore, the most recent clinical trial (DANISH, published in 2016) did not show a survival benefit associated with primary ICD implantation in patients with NICM. Thus, currently there is equipoise on the need to implant a prophylactic ICD (that is associated with short- and long-term complications and risk of inappropriate shocks) in contemporary patients with HFrEF who have a low predicted arrhythmic risk. We have recently provided a comprehensive review on data underlying the current clinical equipoise on the need for routine implantation of a primary prevention ICD in patients with HFrEF and the rationale for conducting clinical trials that aim to reassess the role of the ICD in this population. Two trials are currently reassessing the role of primary prevention ICDs in the setting of contemporary guideline-directed medical therapy (GDMT). The PROFID EHRA trial in Europe aims to evaluate the benefit or harm of routine ICD implantation for primary prevention of SCD with contemporary GDMT in 3595 postmyocardial infarction (MI) patients with LVEF ≤35%. The comparative effectiveness of ICD vs non-ICD therapy in contemporary heart failure patients at a low risk for arrhythmic death (CONTEMP-ICD) trial is a prospective, multicenter, open-label, randomized-controlled trial; enrolling 3,290 participants with HFrEF, on stable and optimal guideline-directed medical therapy (GDMT), who are eligible for a primary prevention ICD, but have a low predicted arrhythmic risk (NCT06543446; https://contemp-icd.org ). Enrolled participants will be randomized in a 1:1 ratio to non-ICD vs ICD treatment arms. We hypothesize that, in patients with HFrEF who have a low predicted arrhythmic risk, Non-ICD vs ICD is noninferior with respect to the primary endpoint of all-cause mortality and superior with respect to survival free of major cardiovascular (CV) events.
Herein we outline the design of the trial. The study is funded by the Patient Centered Outcome Research Institute (PCORI) and has engaged representatives from the Heart Rhythm Society (HRS), the Heart Failure Society of America (HFSA), the American College of Cardiology (ACC), the American Heart Association (AHA), the Association of Black Cardiologist (ABC) and Women Heart. The trial is further endorsed globally by the Canadian Heart Rhythm Society (CHRS), the Canadian Heart Failure Society (CHFS), the European Heart Rhythm Association (EHRA) and the Heart Failure Association of the ESC (HFA).
Objectives
Primary endpoint
The main function of the ICD is to protect the patient against SCD from VTA, by delivering a cardioverting or defibrillating shock and/or ATP based on device programming. Thus, it is imperative to show to patients, providers and stakeholders, that by not implanting an ICD for primary prevention in patients with HFrEF who at a lower predicted arrhythmic risk there will be no corresponding increase in the risk of death associated with untreated VTA events. Accordingly, this study is designed as a noninferiority (NI) comparative effectiveness clinical trial for the primary endpoint of all-cause mortality. We justify the proposed approach and choice of margin below.
Secondary endpoints
survival free of major CV events (defined as first occurrence of admission for heart failure (HF), stroke, myocardial infarction (MI), device-related complications, or ICD shocks requiring admission), healthcare utilization (HCU) and quality of life (QoL). We hypothesize that non-ICD management is associated with improved survival free of major CV events requiring hospitalization compared with prophylactic ICD implant, driven by a reduction in the risk of ICD-related major complications, inappropriate ICD shocks requiring admissions, and increased risk of HF admissions following ICD shocks. We also hypothesize that the reduction in major CV events associated with non-ICD management approach will translate into lower HCU and improved QoL (assessed through the Kansas City Cardiomyopathy Questionnaire [KCCQ], EuroQol-5 Dimension (EQ-5D), Patient-reported Outcomes [PRO]). ICD therapies (comprising either ATP or shocks) and SCD events will be uniformly adjudicated, and will be assessed as a tertiary endpoints.
Subgroup analysis
We hypothesize that, in HFrEF patients with a lower predicted arrhythmic risk, non-ICD vs ICD is consistently noninferior with respect to the primary endpoint of all-cause mortality in prespecified subgroups (including ICM/NICM status [primary subgroup of interest], age, sex, race/ethnicity, Charlson comorbidity index, diabetes, the presence of chronic kidney disease, NYHA Class, and the use of novel GDMT (ie, ARNI and/or SGLT2i). Imaging data, including cardiac magnetic resonance (CMR, when performed as standard of care) and echocardiograms, will be collected for planned substudies that will evaluate clinical and arrhythmic outcomes by CMR and echocardiographic variables.
Study design and methods
Study population and research setting
The CONTEMP-ICD is a prospective, multicenter, open-label, randomized-controlled clinical trial planning to enroll 3,290 participants from 115 sites in the U.S. and Canada with ICM/NICM who are eligible for a primary prevention ICD per current guidelines, but have a lower likelihood of risk of VTA than nonarrhythmic mortality based on the MADIT-ICD Benefit Score and are treated with optimal GDMT. Study participants will be identified from HF, electrophysiology (EP) and general cardiology practices. Enrolled participants will be randomized in a 1:1 ratio to non-ICD vs ICD treatment arms, with randomization stratified by ICM/NICM. The independent Data Coordinating Center (DCC) personnel will be blinded to participant allocation. All clinical and arrhythmic events will be blindly adjudicated by the study’s event committees. Eligibility criteria are provided in Table 1 and the schedule of activities and the design of the study are presented in Table 2 and Figure 1 , respectively.
Table 1
Inclusion and exclusion criteria.
| Inclusion criteria: | Exclusion criteria: |
| Age ≥ 18 years | Existing ICD/CRT-D |
| Class I or IIa indication for a primary prevention ICD | Class I or IIa indication for CRT |
| Echocardiogram documenting LVEF ≤ 35% after being stable for 1 month on optimal GDMT | Acute MI within the past 3 months |
| Optimal GDMT is prespecified as: either receiving all 4 pillar therapies (beta-blockers, ARNI, MRA and SGLT2i) or a GDMT Score 58 ≥ 6 (reason for not receiving all 4 therapies will be documented) | Chronic renal failure requiring hemodialysis |
| MADIT-ICD Benefit Score < 50 | Coronary revascularization within the past 3 months |
| Ischemic or nonischemic cardiomyopathy | History of sustained VT or VF |
| Life expectancy < 1 year | |
| Inability to consent |
Table 2
Schedule of activities ( N = 3,290).
| Visits | Enrollment | Follow-up visits |
|---|---|---|
| Timing | 3, 6, 12, 18, 24, 30, 36 months (±1 month) until the end of follow-up | |
| Informed consent | ✓ | |
| Medical history * |
✓
✓ ✓ ✓ ✓ ✓ |
|
| Pregnancy test † | ✓ | |
| Physical exam * |
✓
✓ ✓ ✓ ✓ ✓ |
|
| NYHA class * | ✓ | |
| Medications/GDMT adherence | ✓ | ✓ |
| ECG * |
✓
✓ ✓ ✓ ✓ ✓ |
|
| Echo * , § |
✓
✓ ✓ ✓ ✓ ✓ |
✓ |
| Quality of life (KCCQ/EQ-5D) ‡ |
✓
✓ ✓ ✓ ✓ |
✓ ¶ |
| PRO ‡ | ✓ | ✓ ¶ |
| Randomization/ICD implant | ✓ | |
| Blood pressure, laboratory parameters (creatinine, BUN, BNP. electrolytes) ║ | ✓ | ✓ |
| Adverse events and terminal event | ✓ | ✓ |
| Healthcare utilization ‡ | ✓ | |
| ICD interrogation # | ✓ |
Design of the CONTEMP-ICD trial.
Eligibility
Candidates eligible for participation in this study include patients 18 years of age or older (no upper age cutoff) with either ICM or NICM who are eligible for prophylactic (primary) ICD implantation (Class I or IIa per 2022 AHA/ACC/HFSA Guideline for the Management of Heart Failure) and have a MADIT-ICD Benefit Score of < 50 ( Table 1 ). For both ICM/NICM etiology, eligibility requires LVEF ≤ 35% documented by a standard of care echocardiogram after being stable for 1 month on optimal GDMT and NYHA II, III or ambulatory IV (NYHA Class I is also allowed if LVEF is ≤ 30%) per current guidelines. Potential candidates who meet elgibility criteria with an echocardiogram showing LVEF ≤ 35%, but have a GDMT score of <6 and proceed to further optimization, will require another echocardiogram demonstrating LVEF ≤ 35% being stable on optimal GDMT to meet eligibility for enrollment. Potential candidates will be excluded if they have an existing ICD/CRT-D device or are eligble for a CRT. Additional exclusion criteria include a Class I or IIa indication for CRT (ie, the presence of LBBB and/or QRS ≥150 msec or an existing pacemaker that will require an upgrade to CRT) or history of sustained VT or VF ( Table 1 ). Patients who are eligible for a CRT-D are excluded since the device was shown to be associated with a reduction of ventricular arrhythmic events, independently of GDMT.
As can be seen in the eligibility criteria, described above, in this trial will employ 2 scores in attempt to address ethical issues of randomizing to Non-ICD therapy despite guideline recommendations. The MADIT-ICD Benefit Score will exclude patients with a higher predicted risk of arrhytmic vs nonarrhythmic mortality (even though it is possible that even higher risk patients who are treated with contemporary GDMT may not derive benefit from a primary prevention ICD) and the GDMT score (that will standardize optimal GDMT among all study participants. These are described briefly below.
The MADIT-ICD Benefit Score
The benefit of prophylactic ICD placement in patients with HFrEF is not uniform due to differences in the risk of life-threatening VTA and nonarrhythmic mortality within this population. Thus, improved selection for primary prevention ICD therapy can be achieved by weighing the patient-specific risk of VTA (for which primary device implantation may be life-saving) against the competing risk of nonarrhythmic mortality (for which primary ICD implantation does not provide protection). We have recently developed and externally validated an ICD benefit prediction score that integrates these competing risks based on simple clinical parameters ( Table 3 ). The study population comprised all 4,531 patients enrolled in all MADIT trials (MADIT-II, MADIT-CRT, MADIT-RISK and MADIT-RIT), including 3,001 patients with ICM (66%) and 1,532 patients with NICM (34%), with a wide variability in arrhythmic and nonarrhythmic risk factors. We identified 8 simple clinical predictors of life-threatening VTA (VT ≥ 200 bpm or VF) and 7 predictors of nonarrhythmic mortality in 2 separate models ( Table 3 ). A personalized ICD Benefit Score was developed based on the integrated distribution of the 2 competing risks scores in the study population ( https://is.gd/madit ). The C indices for VTA and nonarrhythmic mortality scores were 0.83 and 0.81, respectively. Internal validation using bootstrapping showed good correlation between the original cohort and the derivate cohort. External validation in the more contemporary primary prevention RAID trial population ( N = 864) confirmed model stability with similar C indices for the VTA score (0.75) and for the nonarrhythmic mortality (0.72). The score was subsequently externally validated with high performance in additional independent external cohorts. ,,, In the CONTEMP-ICD trial we will enroll only patients with a MADIT-ICD Benefit Score of < 50, ie, patients who have a lower predicted risk of life-threatening VTA than nonarrhythmic mortality. The score is based on the distribution of its components within the internal and external validation cohorts. Thus, a score < 50 represents approximately 50% of HFrEF patients who are eligible for an ICD. Analysis of event rates from the MADIT Database shows that at 3 years patients with a score of < 50 have increased rates of nonarrhythmic and total mortality (29% and 35%, respectively), but lower rates of life-threatening VTA (6%). We believe that using the MADIT-ICD benefit in this comparative effectiveness clinical trial has important advantages: (1) it addresses ethical concerns of not implanting an ICD for primary prevention in patients with a higher predicted VTA vs nonarrhythmic mortality risk; (2) it is the only established score to date that takes into account the competing risk of life-threatening VT/VF vs the competing risk of nonarrhythmic mortality (whereas older scores, such as the Seattle Heart Failure Model , and our prior MADIT-II score, only stratified risk by of all-cause mortality); (3) it is based on simple, readily available, clinical factors that can easily be implemented into real world clinical practice; (4) high performance and external validation in contemporary cohorts; and (5) the score can be used for shared decsion-making on the potential life-saving benefit vs risks of the ICD (eg, unnecessary shocks and device-related complications).
Table 3
The MADIT-ICD benefit score * , .
| MADIT-ICD Benefit Group | Lowest | Intermediate | Highest | |||
|---|---|---|---|---|---|---|
| VT/VF score | LVEF ≤25% | 1 | Low (<7) | Low (<7) | High (≥7) | High (≥7) |
| Atrial arrhythmia | ||||||
| Heart Rate > 75 bpm | ||||||
| SBP <140 mmHg | 2 | |||||
| Myocardial Infarction | ||||||
| Age < 75 yrs | ||||||
| Male | ||||||
| Prior NSVT | ||||||
| Nonarrhythmic mortality score | CRT-D | −1 | High (≥3) | Low (<3) | High (≥3) | Low (<3) |
| NYHA ≥II | 1 | |||||
| Diabetes | ||||||
| BMI <23 kg/m² | 2 | |||||
| Atrial arrhythmia | ||||||
| LVEF ≤25% | ||||||
| Age ≥ 75 years | ||||||
The GDMT Score
Guidelines strongly recommend that patients with HFrEF be treated with multiple medications proven to improve clinical outcomes, as tolerated. , Currently the “4 pillars” of GDMT for HFrEF consist of beta-blockers, ARNI (or ACEI/ARB as an alternative if ARNI not tolerated or not accessible), MRAs, and SGLT2i. However, implementation of this recommendation is variable and suboptimal in clinical practice. Furthermore, current guidelines for a primary prevention ICD states that consideration should be given for patients “on GDMT” but do not specify the type and dosages that comprise optimal GDMT required prior to device implantation or documentation of reasons for lack of optimal GDMT, ,,,, resulting in wide variability in the medical management of patients who receive an ICD. In the proposed study, we aim to standardize treatment with optimal GDMT for HFrEF at the time of enrollment. We plan to achieve this in 2 steps. First, patients will be required to be on all 4 therapies unless the site investigator documents a reason (e.g., intolerance, costs, contraindications, patient preference etc.). Second, to address variability in GDMT management for HFrEF in real world clinical practice, we will require that the GDMT Score ( Table 4 ) is used as part of standard qualification and the inclusion criteria of the trial. The score was developed in 2021 by the Heart Failure Collaboratory multistakeholder group (comprising clinical investigators, clinicians, patients, government representatives including U.S. Food and Drug Administration and National Institutes of Health participants, payers). Of note, Co-PI of this application, Dr. Butler, was a participant in this initiative. The GDMT Score aims to provide a more pragmatic approach to real world implementation of background drug therapy in clinical trials of patients with HFrEF, wherein patients with HFrEF can achieve a GDMT Score of ≥ 6 with novel (nongeneric) GDMT (ARNI/SGLT2i) regardless of dose or with target doses of generic GDMT (ACEI/ARB and beta-blockers with MRAs) if there is intolerability or contraindication to one or more novel GDMT ( Table 4 ). Thus, in CONTEMP-ICD candidates for participation who do not receive all “4 pillar” therapies (with a documented reason) will only be eligible for enrollment if LVEF ≤ 35% is documented after being stable on optimal GDMT with a Score of at least 6. This approach will standardize management to ensure that patients enrolled into the Non-ICD arm do not receive suboptimal therapy. Importantly, potential bias arising from nonstandardized medical management was not addressed in prior primary prevention ICD trials. ,,,, A GDMT score of ≥ 6, compensates for intolerability of some drugs, that may occur in otherwise eligible patients, by requiring an optimized regimen involving alternative therapies. This approach will facilitate implementation of the trial results reflective of a real-world practice setting and at the same time avoid enrolling patients who are on inadequate baseline medical therapy.
Table 4
The GDMT score .
| Medication class | Medication | GDMT Score | |
|---|---|---|---|
| Beta-blockers | None | 0 | |
| <50% max daily dose | 1 | ||
| ≥50% max daily dose | 2 | ||
| MRA | None | 0 | |
| Any dose | 2 | ||
| RAAS Inhibition OR ARNI (preferred) | None | 0 | |
| <50% max daily dose ACEI/ARB | 1 | ||
| ≥50% max daily dose ACEI/ARB | 2 | ||
| Sacubitril-valsartan | any dose | 3 | |
| SGLT2i/Other | SGLT2i | None | 0 |
| Any dose | 2 | ||
| Ivabradine | None | 0 | |
| Any dose | 1 | ||
| Vericiguat | None | 0 | |
| Any dose | 1 | ||
| Hydralazine/Nitrates | None | 0 | |
| Any dose | 1 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
