In ablation strategy for persistent Atrial Fibrillation (PsAF), ablation limited to Pulmonary Vein (PV) isolation is the most straightforward approach but the result give only 50% of arrhythmia free follow-up. Substrate modification strategies have failed to demonstrate their superiority with variable reported success rate. The Marshall network is a highly arrhythmogenic structure that has not been incorporated in current ablation strategies. The investigators sought to investigate a new ablation strategy that target systematically the vein of Marshall by ethanol infusion. This step is integrated in a new ablation strategy consisting in a global anatomical substrate based ablation including PV isolation and left atrial linear ablation (Marshall-Plan).

The main objective of this study is to compare the 12 month freedom from any arrhythmia (Atrial Fibrillation (AF)/Atrial Tachycardia (AT)) between the Marshall-Plan approach and the PV isolation approach.

Ablation strategy for persistent AF besides pulmonary vein isolation remains controversial. Indeed, two approaches have prevailed over the past two decades "cox-maze" strategy and mapping of the left atrium but both methods have failed to decrease AF recurrences (as shown by the clinical trial STAR AF 2). Two studies have demonstrated that the ligament of Marshall (LOM) could be the source of focal activities, the substrate of reentries and a strong parasympathetic modulator. For these reasons, LOM may represent a major target in AF treatment besides PV isolation. Nevertheless, conventional ablation techniques do not ensure the complete destruction of Marshall's musculature and parasympathetic ganglia that surround it, largely isolated by a sheath of adipose tissue. To overcome this technical limitation, LOM elimination can be achieved by alcohol injection into the vein of Marshall.The investigator innovative approach called Marshall Plan will then consists in 3 steps: 1) the destruction of Marshall bundles by ethanol infusion followed by the ablation of the distal Coronary Sinus (CS) bundles, the ridge and the saddle; 2) the standard PV isolation; 3) and finally ablation of the mitral line, the roof and of the cavo-tricuspid isthmus, main causes of recurrence in atrial flutter.

Before ablation procedure patients will be randomized in 2 arms: Marshall Plan (treatment arm) or pulmonary vein isolation (control arm). Patients will be followed at 3, 6, 9 and 12 months in-office visits or hospitalization. Patients will have different tests: electrocardiogram (ECG), cardiac echography, stress test, 24hours Holter and transtelephonic ECG monitor with weekly transmitted ECG and at any time in case of symptoms.

Inclusion Criteria:

• Age > 18 years of both genders

• Suitable candidate for catheter non-emergent mapping and ablation of atrial fibrillation defined as:

  • History of symptomatic persistent atrial fibrillation in the past year documented by ECG AND
  • Treatment failure by at least one class of anti-arrhythmic medications (I-IV) (intolerance or recurrence of symptomatic AF)

• Patient affiliated or beneficiary of social security scheme

• Free, informed and written consent signed by the participant and the principal investigator (at least at the inclusion date and before all exams required for the clinical research)

• Effective contraception for women of childbearing potential

Exclusion Criteria:

• Minor
• Prior left atrial heart ablation procedure
• Documented left atrial thrombus or another abnormality which precludes catheter introduction
• Contraindication to anticoagulation therapy (heparin, warfarin, or Novel Oral Anticoagulant (NOAC)
• Contraindication to iodinated contrast product XENETIX® (iobitridol hypersensitivity or at one of these excipients, history of major immediate reaction or cutaneous reaction to XENETIX® infusion, thyrotoxicosis)
• Hypersensitivity to ethanol
• Unstable angina or ongoing myocardial ischemia
• Myocardial infarction within 3 months prior to inclusion
• Congenital heart disease, where the underlying abnormality increases the ablation risk
• Pulmonary hypertension (pulmonary arterial hypertension > 50 mmHg)
• Severe uncontrolled systemic hypertension with systolic blood pressure (SBP) > 200 mm Hg within 30 days prior to inclusion
• Severe bleeding, clotting or thrombotic disorder
• Left atrial diameter > 60 mm (parasternal view)
• Hypertrophic cardiomyopathy defined by a left ventricular septum thickness > 1.5 cm
• Pregnant, parturient or nursing women
• Unable or unwilling to provide written informed consent
• Patient detained by judicial or administrative order
• Patient under psychiatric care
• Patient admitted in a social or healthcare establishment for any purpose other than the research
• Subject to a legal protection order (guardianship, patient under legal protection)