Decision to discontinue in light of new evidence demonstrating that the intervention in question is not effective.

RABLAP-AF will compare pulmonary vein isolation (PVI) in combination with posterior wall isolation (PWI) for patients with persistent atrial fibrillation (AF).

Ablation for AF has a high success rate for patients with paroxysmal AF, but success rates for persistent AF remain significantly lower. These patients frequently have more advanced disease in their left atrium than those with paroxysmal AF.

A common practice is to perform left atrial PWI in addition to PVI in order to reduce AF recurrence. This approach has conflicting evidence behind it and a large scale clinical trial in those with advanced disease is required. The investigators hope to fill this gap with RABLAP-AF.

The investigators will randomise patients with advanced atrial disease 1:1 to PVI+PWI or PVI alone. All patients will receive implantable loop recorders two months prior to the ablation procedure in order to accurately assess the effect of PWI and compare arrhythmia burden before and after ablation.

Inclusion criteria:

• Age 18 to 80 (inclusive)
• Patients with symptomatic advanced persistent AF, defined as:
• AF which does not spontaneously terminate and requires medical intervention (antiarrhythmics or DCCV) to terminate, AND
• Recurs following cardioversion, AND
• Is associated with evidence of left atrial cardiopathy, defined as at least mild left atrial enlargement on imaging (echo, CT or MRI) as defined by standard values (for example British Society of Echocardiography or European equivalent) - in the event of uncertainty, Principal Investigator discretion is allowed.

Exclusion Criteria

• Any previous catheter (or surgical) ablation for AF (previous ablation for atrial flutter or focal atrial tachycardias are allowable)
• An indwelling atrial septal defect occluder device or anatomical structure that pre-vents free access to the left atrium
• Severe left ventricular systolic dysfunction (ejection fraction less than 35%)
• Recent stroke/transient ischaemic attack within 3 months
• Inability, unwillingness or absolute contraindication to taking an oral anticoagulant medication
• Severe kidney function impairment (eGFR less than 30ml/min/1.73m2)
• Morbid obesity with a body mass index ≥40
• Extreme frailty (a score of 7,8 or 9 on the Rockwood Clinical Frailty Scale)
• Severe valvular heart disease of any kind as assessed by the investigator, with or without a prosthetic valve in place
• Pregnancy

Patients and operators will be aware of the treatment arm. Researchers interpreting and assessing outcome data will be blinded to treatment received.