The New England Journal of Medicine DECEMBER 16, 2010

Cardiac-Resynchronization Therapy for Mild-to-Moderate Heart Failure (RAFT)

Anthony S.L. Tang, George A. Wells, Mario Talajic, Malcolm O. Arnold, Robert Sheldon, Stuart Connolly, Stefan H. Hohnloser, Graham Nichol, David H. Birnie, John L. Sapp, Raymond Yee, Jeffrey S. Healey, Jean L. Rouleau, and the Resynchronization-Defibrillation for Ambulatory Heart Failure Trial Investigators

Bottom Line

The RAFT trial demonstrated that adding cardiac resynchronization therapy (CRT) to an implantable cardioverter-defibrillator (ICD) reduces all-cause mortality and heart failure hospitalizations in patients with mild-to-moderate heart failure, left ventricular dysfunction, and a wide QRS complex.

Key Findings

1. The primary outcome (composite of death from any cause or hospitalization for heart failure) occurred in 33.2% (297/894) of the CRT-D group compared to 40.3% (364/904) in the ICD-only group (hazard ratio 0.75; 95% CI, 0.64 to 0.87; P<0.001).
2. All-cause mortality was significantly lower in the CRT-D group compared to the ICD-only group (18.6% vs. 23.3%; hazard ratio 0.75; 95% CI, 0.62 to 0.91; P=0.003).
3. CRT-D implantation was associated with a higher rate of procedure-related adverse events compared to ICD-only (124 events vs. 58 events; P<0.001).
4. The observed survival benefit in the CRT-D group did not become statistically evident until approximately 2 years after therapy initiation.

Study Design

Design
RCT
Double-Blind
Sample
1,798
Patients
Duration
40 mo
Median
Setting
Multicenter, international
Population Patients with NYHA class II or III heart failure, left ventricular ejection fraction of 30% or less, and intrinsic QRS duration of 120 ms or more (or paced QRS of 200 ms or more)
Intervention Implantable cardioverter-defibrillator (ICD) plus cardiac resynchronization therapy (CRT)
Comparator Implantable cardioverter-defibrillator (ICD) alone
Outcome Composite of death from any cause or hospitalization for heart failure

Study Limitations

Increased risk of early device- or procedure-related complications in the CRT-D arm.
The study was not specifically powered to identify the optimal patient subset for therapy, although subsequent analyses have explored this.
Higher device-related hospitalization rates partially offset the reductions in heart failure-related hospitalizations.
Performance of CRT was suboptimally delivered in a portion of the experimental group, potentially attenuating the observed clinical effect.

Clinical Significance

The RAFT trial established the standard of care for using CRT-D in patients with NYHA class II or III symptoms who have a wide QRS complex and reduced left ventricular ejection fraction, providing evidence for mortality reduction in this less symptomatic population compared to ICD therapy alone.

Historical Context

Prior to RAFT, CRT was primarily indicated for patients with severe (NYHA class III/IV) heart failure. RAFT expanded the evidence base to include the more ambulatory, mildly symptomatic population, influencing international guidelines and solidifying the role of resynchronization therapy in earlier stages of heart failure.

Guided Discussion

High-yield insights from every perspective

Med Student
Medical Student

What is the physiologic rationale for using a biventricular pacemaker (CRT) in a patient with a wide QRS complex and heart failure, and how does it improve cardiac output?

Key Response

A wide QRS complex, particularly with Left Bundle Branch Block (LBBB), indicates electrical dyssynchrony where the left ventricular lateral wall contracts later than the septum. This reduces the efficiency of the pump. CRT provides biventricular pacing to resynchronize the contraction, decreasing functional mitral regurgitation, improving ventricular filling time, and increasing the left ventricular ejection fraction (LVEF) through more coordinated mechanical activity.

Resident
Resident

Based on the RAFT trial, which specific NYHA classes and QRS thresholds define the population that benefits from CRT-D, and what was the impact on all-cause mortality?

Key Response

The RAFT trial enrolled patients with NYHA Class II or III heart failure, an LVEF of 30% or less, and a QRS duration of 120 ms or more (or a paced QRS of 150 ms or more). Unlike some previous trials that focused primarily on hospitalizations, RAFT demonstrated that adding CRT to an ICD significantly reduced all-cause mortality (hazard ratio 0.75) in addition to reducing heart failure hospitalizations.

Fellow
Fellow

How did the RAFT trial's findings regarding QRS morphology and duration influence our understanding of 'non-responders,' and how do these results compare to the MADIT-CRT subgroup analyses?

Key Response

Subgroup analysis in RAFT suggested that the benefit of CRT-D was most pronounced in patients with a QRS duration of 150 ms or greater. Similar to MADIT-CRT, subsequent interpretations of RAFT data emphasized that patients with LBBB morphology derive significantly more benefit than those with non-LBBB patterns (e.g., RBBB or IVCD). This has led to the current practice of prioritizing CRT in patients with LBBB and wider QRS durations to minimize the rate of clinical non-response.

Attending
Attending

The RAFT trial reported a higher rate of device-related complications in the CRT-D group compared to the ICD-only group. How should this data shape the shared decision-making process for a NYHA Class II patient with borderline QRS duration?

Key Response

Attending-level care requires balancing the 25% relative risk reduction in mortality against the increased risk of peri-procedural complications (14% in CRT-D vs 6.7% in ICD-only at 30 days). For a 'mild' HF patient with a QRS of 120-130ms, the clinician must discuss that while long-term survival is improved, the risk of lead displacement, infection, and procedural failure is significantly higher with the more complex CRT-D system.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD
PhD

RAFT utilized a time-to-first-event composite endpoint of death or HF hospitalization. From a trial design perspective, how might a recurrent-events analysis (e.g., Andersen-Gill model) have altered the perceived treatment effect compared to the traditional Cox proportional hazards model?

Key Response

Traditional time-to-first-event analyses ignore subsequent heart failure hospitalizations after the initial event, potentially underestimating the total disease burden and the full magnitude of the treatment effect. A recurrent-events model would capture the cumulative reduction in hospitalizations over the 40-month mean follow-up, likely providing a more robust estimate of the therapy's impact on healthcare utilization and chronic disease stability.

Journal Editor
Journal Editor

Given that the RAFT trial could not be double-blinded due to the nature of the device implantation, what measures were taken to protect the 'hospitalization for heart failure' endpoint from detection bias, and is the mortality benefit sufficient to overcome this potential weakness?

Key Response

The potential for bias in non-blinded trials lies in the subjective decision to hospitalize. RAFT mitigated this by using an independent, blinded adjudication committee for all clinical events. However, the fact that the mortality benefit (an objective, 'hard' endpoint) was statistically significant (p=0.003) serves as a critical internal validation, reassuring editors that the overall study conclusions are not merely artifacts of biased hospitalization decisions.

Guideline Committee
Guideline Committee

How did the RAFT trial change the strength of recommendation for CRT in NYHA Class II patients compared to previous iterations of the ACC/AHA/HFSA guidelines?

Key Response

Prior to trials like RAFT and MADIT-CRT, CRT was primarily a Class I recommendation for NYHA Class III/IV. RAFT provided the necessary evidence for 'mild' heart failure (Class II) to be upgraded. Current guidelines (e.g., 2022 AHA/ACC/HFSA) now give a Class 1 recommendation for CRT in patients with LVEF ≤35%, LBBB, and QRS ≥150ms who are in NYHA Class II, III, or ambulatory IV, specifically citing the mortality benefits seen in RAFT.

Clinical Landscape

Noteworthy Related Trials

2004

COMPANION Trial

n = 1,520 · NEJM

Tested

Cardiac resynchronization therapy (CRT) or CRT plus an implantable cardioverter-defibrillator (CRT-D)

Population

Patients with NYHA class III or IV heart failure and prolonged QRS interval

Comparator

Optimal pharmacological therapy alone

Endpoint

Composite of death from any cause or first hospitalization

Key result: CRT and CRT-D significantly reduced the risk of death or hospitalization compared to optimal medical therapy alone in advanced heart failure.
2005

CARE-HF Trial

n = 813 · NEJM

Tested

Cardiac resynchronization therapy (CRT)

Population

Patients with NYHA class III or IV heart failure, left ventricular systolic dysfunction, and cardiac dyssynchrony

Comparator

Optimal medical therapy

Endpoint

Composite of death from any cause or an unplanned hospitalization for major cardiovascular events

Key result: CRT significantly reduced the risk of death and cardiovascular hospitalizations in patients with moderate to severe heart failure.
2009

MADIT-CRT Trial

n = 1,820 · NEJM

Tested

Cardiac resynchronization therapy with an implantable cardioverter-defibrillator (CRT-D)

Population

Patients with asymptomatic or mildly symptomatic (NYHA class I or II) heart failure, reduced LVEF, and prolonged QRS

Comparator

Implantable cardioverter-defibrillator (ICD) alone

Endpoint

Death from any cause or a nonfatal heart-failure event

Key result: The addition of CRT to ICD therapy significantly reduced the risk of heart-failure events in patients with asymptomatic or mild heart failure.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis