New England Journal of Medicine February 01, 2018

Catheter Ablation for Atrial Fibrillation with Heart Failure (CASTLE-AF)

Nassir F. Marrouche, Johannes Brachmann, Dietrich Andresen, et al.

Bottom Line

In patients with atrial fibrillation and heart failure with reduced ejection fraction, catheter ablation significantly reduced the composite endpoint of all-cause mortality and hospitalization for worsening heart failure compared to medical therapy.

Key Findings

1. At a median follow-up of 37.8 months, the primary composite endpoint (all-cause mortality or hospitalization for worsening heart failure) occurred in 28.5% of patients in the ablation group versus 44.6% in the medical-therapy group (Hazard Ratio [HR] 0.62; 95% CI, 0.43 to 0.87; P=0.007).

2. All-cause mortality was significantly lower in the ablation arm (13.4%) compared to the medical therapy arm (25.0%) (HR 0.53; 95% CI, 0.32 to 0.86; P=0.01).

3. Hospitalization for worsening heart failure occurred in 20.7% of ablation patients versus 35.9% of medical therapy patients (HR 0.56; 95% CI, 0.37 to 0.83; P=0.004).

4. Cardiovascular mortality was also significantly reduced in the ablation group (11.2% vs. 22.3%; HR 0.49; 95% CI, 0.29 to 0.84; P=0.009).

Study Design

Design

Randomized Controlled Trial

Open-Label

Sample

363

Patients

Duration

37.8 mo

Median

Setting

Multicenter, multinational

Population Adults with symptomatic paroxysmal or persistent atrial fibrillation refractory or intolerant to antiarrhythmic drugs, New York Heart Association (NYHA) class II-IV heart failure, LVEF ≤35%, and an implanted cardioverter-defibrillator (ICD) or cardiac resynchronization therapy defibrillator (CRT-D).

Intervention Catheter ablation (pulmonary-vein isolation) for atrial fibrillation in addition to guideline-based medical therapy for heart failure.

Comparator Conventional medical therapy (rate or rhythm control) in addition to guideline-based medical therapy for heart failure.

Outcome A composite of death from any cause or hospitalization for worsening heart failure.

Study Limitations

• The open-label design introduced the potential for performance and detection bias in patient management, though the primary endpoints of death and hospitalization were relatively objective.

• The study enrolled a highly selected cohort; only 363 patients were randomized across 33 centers over an 8-year period, representing a very small fraction of real-world patients with AF and heart failure.

• All participants were required to have an implanted defibrillator (ICD or CRT-D) capable of continuous rhythm monitoring, which limits the generalizability of the findings to patients without such devices.

• The extended enrollment period of 8 years meant that both ablation technologies and guideline-directed medical therapies for heart failure evolved considerably during the course of the trial.

Clinical Significance

CASTLE-AF was a paradigm-shifting trial demonstrating a robust survival and morbidity benefit for catheter ablation over conventional medical therapy in patients with atrial fibrillation and heart failure with reduced ejection fraction (HFrEF). By showing a 47% relative reduction in mortality, it elevated catheter ablation from a purely symptom-relieving procedure to a life-prolonging intervention in highly selected, device-monitored patients, substantially influencing contemporary electrophysiology and heart failure guidelines.

Historical Context

Historically, the primary goal of atrial fibrillation ablation was symptom control and quality-of-life improvement. Earlier trials comparing rhythm control (mostly via antiarrhythmic drugs) to rate control in heart failure patients, such as the AF-CHF trial, failed to show a survival benefit, largely due to the systemic toxicities and modest efficacy of drugs like amiodarone. While smaller preliminary studies had suggested that catheter ablation could improve left ventricular ejection fraction in heart failure, CASTLE-AF was the first landmark randomized controlled trial to definitively prove that restoring and maintaining sinus rhythm via ablation could significantly alter the natural history of HFrEF and reduce hard clinical endpoints like mortality.

Guided Discussion

High-yield insights from every perspective

Med Student

Medical Student

Why might restoring and maintaining sinus rhythm via catheter ablation provide a greater physiological and prognostic benefit in patients with heart failure with reduced ejection fraction (HFrEF) compared to those with a normal ejection fraction?

Key Response

In HFrEF, the loss of the 'atrial kick' significantly compromises ventricular filling and cardiac output. Additionally, irregular R-R intervals and rapid ventricular rates can lead to tachycardia-induced cardiomyopathy, further depressing the already reduced ejection fraction. Restoring sinus rhythm reverses these specific pathophysiological detriments, offering a mortality benefit not typically seen in patients with structurally normal hearts.

Resident

A 65-year-old patient with NYHA class III HFrEF (LVEF 30%) continues to experience symptomatic AF despite optimal guideline-directed medical therapy and amiodarone. Based on the CASTLE-AF trial, how does referring this patient for catheter ablation compare therapeutically to continuing pharmacological rhythm control?

Key Response

The CASTLE-AF trial demonstrated that catheter ablation significantly reduces the composite of all-cause mortality and worsening heart failure hospitalizations compared to medical therapy. Historically, trials like AF-CHF showed that pharmacological rhythm control does not improve survival in HFrEF. CASTLE-AF establishes procedural rhythm control as superior to medical rhythm control for improving hard clinical outcomes in appropriately selected HFrEF patients.

Fellow

The Kaplan-Meier curves for the primary endpoint in the CASTLE-AF trial did not significantly diverge until approximately 3 years post-randomization. What procedural and pathophysiological factors explain this delayed benefit, and how should this temporal delay influence patient selection for AF ablation in advanced heart failure?

Key Response

The delayed separation of the curves reflects the immediate peri-procedural risks of ablation and the time required for structural reverse remodeling and recovery of left ventricular function once sinus rhythm is restored. For fellows, this emphasizes that ablation is most appropriate for HFrEF patients with a reasonable life expectancy (greater than 3 years); those with very advanced or end-stage heart failure may not survive long enough to derive the mortality benefit.

Attending

CASTLE-AF showed a remarkable 38% relative risk reduction in its primary endpoint, yet the investigators screened over 3,000 patients to randomize just 363. How does this highly selective enrollment impact your shared decision-making process when considering ablation for a typical HFrEF patient in your clinical practice?

Key Response

The highly selected trial population (all patients had an implanted ICD/CRT-D, specific AF burdens, and no recent irreversible causes) means the Number Needed to Treat (NNT) and striking benefits may not perfectly translate to the broader, less selected real-world clinic population. Attendings must carefully match their patient's phenotype to the trial's inclusion criteria to realistically set expectations regarding survival and symptom improvement during shared decision-making.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD

Given the unblinded nature of the CASTLE-AF trial, how might knowledge of the assigned treatment arm introduce bias into the composite primary endpoint of all-cause mortality and heart failure hospitalization, and what specific statistical sensitivity analyses would you expect to see to validate the robustness of the hospitalization component?

Key Response

Unblinded trials are highly susceptible to ascertainment and admission biases, where clinicians might have a lower threshold to admit a medical-therapy patient for heart failure than one who recently underwent ablation. A methodologist would look for competing risk analyses (like the Fine-Gray model) and evaluate the hard endpoint (all-cause mortality) independently from the softer endpoint (hospitalization) to ensure the composite result is not entirely driven by subjective admission thresholds.

Journal Editor

CASTLE-AF experienced notable rates of patient crossover and loss to follow-up. As a journal editor evaluating the manuscript, how do you assess the impact of crossover on the intention-to-treat (ITT) effect size, and what supplementary analyses would you mandate to ensure the findings are not statistically fragile?

Key Response

Crossover in trials evaluating interventions often biases the intention-to-treat (ITT) analysis toward the null. A rigorous editor would demand 'As-Treated' or 'Per-Protocol' sensitivity analyses to define the true biological efficacy of the ablation procedure, while also requiring an assessment of missingness (e.g., multiple imputation) to ensure that dropouts did not disproportionately affect one arm, thereby threatening the internal validity of the landmark findings.

Guideline Committee

Prior to CASTLE-AF, catheter ablation in patients with HFrEF was generally viewed as a Class IIb recommendation primarily for symptom control. How should the mortality and hospitalization benefits demonstrated in CASTLE-AF alter the class of recommendation and level of evidence for AF ablation in this population in updated clinical practice guidelines?

Key Response

The results of CASTLE-AF provide randomized, controlled data proving a mortality and morbidity benefit, transforming AF ablation from a purely symptomatic treatment into a disease-modifying therapy in HFrEF. Consequently, contemporary guidelines (such as the 2023 ACC/AHA/ACCP/HRS guidelines) have upgraded AF ablation in appropriately selected HFrEF patients to a Class 1 recommendation to improve survival and reduce heart failure hospitalizations.

Clinical Landscape

Noteworthy Related Trials

2016

AATAC Trial

n = 203 · JACC

Tested

Catheter ablation

Population

Patients with persistent atrial fibrillation, heart failure, and an implantable device

Comparator

Amiodarone

Endpoint

Freedom from atrial fibrillation recurrence

Key result: Catheter ablation was superior to amiodarone in maintaining sinus rhythm, reducing hospitalizations, and lowering mortality.

2019

CABANA Trial

n = 2,204 · JAMA

Tested

Catheter ablation

Population

Patients with new-onset or untreated atrial fibrillation

Comparator

Medical therapy (rate or rhythm control)

Endpoint

Composite of death, disabling stroke, serious bleeding, or cardiac arrest

Key result: Catheter ablation did not significantly reduce the primary composite endpoint compared to medical therapy, though it significantly reduced atrial fibrillation recurrence.

2021

RAFT-AF Trial

n = 411 · Circulation

Tested

Ablation-based rhythm control

Population

Patients with high-burden atrial fibrillation and heart failure

Comparator

Medical rate control

Endpoint

Composite of all-cause mortality and heart failure events

Key result: There was no statistically significant difference in the primary endpoint, though ablation showed trends toward improved outcomes and significant improvements in quality of life.

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