In this study 404 patients with heart failure and an ejection fraction of 0.40 or less, with paroxysmal atrial fibrillation, will be randomly assigned to standard treatment or standard treatment plus a session of cryoballoon ablation (left atrial balloon cryoablation for pulmonary vein isolation). All patients with either have an ICD or CRT-D/P device implanted or an implantable electrocardiographic monitoring device. The primary study endpoint will be the time to AF burden exceeding 1% over any 30-day period (calculated as the ratio of time spent in AF over total time).16 This AF burden corresponds to 7.2 hours per month.

A powered secondary endpoint will be the time to the composite of all-cause mortality and unplanned hospitalization for heart failure.

Inclusion criteria:

• A minimum of at least two separate AF episodes recorded within the last 12 months, either on 12-lead rest ECG or on ambulatory ECG recordings (at least 5 minutes of AF on Holter recordings will be required),
• LVEF <40% on sinus rhythm,
• symptoms consistent with heart failure (New York Heart Association II or higher) despite rate control treatment,
• age >21 years old.

Exclusion criteria:

• previous left atrial ablation,
• left atrial diameter >28 mm/m2 BSA on TTE (parasternal long axis view),
• strong clinical suspicion that reduced LVEF is primarily due to tachycardiomyopathy,
• pre-existing device (ICD or CRT) without AF detection algorithms and/or (for CRT devices) diagnostics of effective biventricular pacing,
• known primary electrical heart disease (e.g. Brugada syndrome),
• presence of thrombus in a heart chamber,
• presence of prosthetic valve at any position,
• moderate/severe valvular heart disease,
• active infectious disease or malignancy,
• moderate or severe hepatic impairment (Child-Pugh class B or C),
• severe renal failure (estimated glomerular filtration rate <20 ml/min/1.73 m2),
• inability or unwillingness to adhere to standard treatment or to provide consent.