Atrial fibrillation (AF) is a common heart rhythm disorder that causes an irregular heart beat and is a cause of heart failure (HF). Treatments include drugs to slow the heart rate, anti-arrhythmic drugs or ablation of the heart to help preserve normal rhythm. A number of trials have suggested that ablation may be superior to drug treatment to reduce hospitalisations or prevent early death. However, these studies have been small and the results not applicable to the general population with AF and heart failure in the UK. This international study will compare catheter ablation and optimal medical therapy versus optimal medical therapy alone to see if catheter ablation reduces unplanned heart failure hospitalisations and death rates and improves quality of life.

Atrial fibrillation (AF) increases the severity of, and death from, heart failure (HF). Several small studies have demonstrated that restoration of sinus rhythm by catheter ablation in patients with HF improves left ventricular (LV) function and exercise tolerance. What is unknown is whether or not AF ablation reduces all-cause death and urgent CV hospitalisations in populations with HF. The current trial will answer this outstanding question, which is faced by HF clinicians and electrophysiologists on a daily basis. AF ablation can be performed very effectively and efficiently using a cryo-balloon or radio-frequency ablation PVI technique. These techniques have evolved slowly and are unlikely to change substantially over the course of this trial. One small trial (n=363) in implantable cardioverter-defibrillator and CRT defibrillator recipients (CASTLE-AF) reported a death benefit of AF ablation but the patients were highly selected and the death reduction was far higher than real world expected differences. Recent studies have noted that the population randomised in CASTLE-AF was not representative of the general HF population with only 7% of patients in the "real world" setting meeting the trial entry criteria. CASTLE-AF is therefore provocative but inconclusive; it has made little change to clinical practice. As no studies have investigated the death benefit in a general HF population, the proposed trial is necessary and warranted. This study is designed as a randomised, open label multicentre clinical trial in which catheter ablation and medical therapy is compared to medical therapy alone in patients with HF with reduced ejection fraction (<50%) and paroxysmal or persistent AF to determine if this reduces all-cause death and urgent CV hospitalisations as well as improving QoL. By utilising the clinical and research networks of the British Heart Failure Society and British Heart Rhythm Society (BHRS) we will recruit 1200 patients. The current trial will be almost three times the size of the only previous inconclusive trial which was reported in the New England Journal of Medicine.

Inclusion criteria:

1. Patients aged ≥18 years.

2. Patient is willing and able to give informed consent for participation.

3. Able and willing to comply with all study requirements, including ability to participate in study for 12 months.

4. Willing to allow their General Practitioner (GP) to be notified of participation in the study.

5. Patient with one of the following AF categories and at least one episode of AF documented (by any means eg ECG, Holter, Cardiac Implantable Electronic Device (CIED) interrogation or any other means):

   • Paroxysmal AF defined as spontaneous self-terminating AF lasted > 6 hours and <7 days.
   • Persistent AF as defined by at least one episode of AF >7 days but not >3 years (since 1st documentation)

6. Optimal tolerated medical therapy for HF (including ACE-I (or ARB or ARNi), beta-blocker, SGLT2 inhibitor and mineralocorticoid receptor antagonist (MRA) and cardiac resynchronisation therapy (CRT) where indicated & tolerated) for at least 6 weeks (according to the most contemporary European Society of Cardiology (ESC) HF guidelines). Maximal doses of these drugs are not mandated.

7. New York Heart Association Classification (NYHA) class II to III

8. LVEF <50% (Cardiac imaging report of LVEF<50% within 1 year (by echocardiography, cardiac magnetic resonance imaging or nuclear cardiology assessment)) AND after optimisation of medical therapy (see previous definition). Note - a LVEF of <50% must be documented by any cardiac imaging performed after optimisation of medical therapy. Documentation of other baseline echocardiographic parameters (eg LA volume, E/E' etc can be obtained from any echocardiogram within 2.5 years). This allows a handheld or echocardiogram focused on LVEF assessment.

   1. For those with LVEF 41-49% and without ongoing atrial fibrillation/flutter, N-terminal pro B-type natriuretic peptide (NT-proBNP) of ≥300pg/mL is required within 12 months prior to randomisation.
   2. For those with LVEF 41-49% and with ongoing atrial fibrillation/flutter, NTproBNP of ≥600pg/mL is required within 12 months prior randomisation.
   3. For those with LVEF ≤40%, NTproBNP is not required

Exclusion criteria:

1. Long standing (>3 year) persistent or permanent AF.
2. Previous atrioventricular (AV) nodal ablation.
3. Previous pulmonary vein isolation (PVI) or surgical ablation.
4. Recent (<90 days) (type 1 spontaneous) myocardial infarction (type 2 myocardial infarctions are not an exclusion criterion), percutaneous coronary intervention, coronary artery bypass grafting, cardiac resynchronisation therapy or stroke.
5. Severe aortic or pulmonary valve disease.
6. Severe primary or secondary mitral valve regurgitation.
7. Active illness (other than HF) likely to result in death within 2 years.
8. People who are pregnant or planning to become pregnant during the trial.
9. People who are breastfeeding.
10. Known allergy to contrast.
11. Contraindication for PVI.
12. Other conditions that may prevent subjects from adhering to the trial protocol, in the opinion of the investigator.
13. Currently participating in another randomised controlled trial of another drug or medical device.