The broad, long-term objective of this project is to evaluate the therapeutic value of vein of Marshall (VOM) ethanol infusion when added to catheter ablation of atrial fibrillation (AF). AF is the most common sustained arrhythmia in adults, and it is a leading cause of stroke, disability and increased mortality. Catheter ablation - pulmonary vein (PV) antral isolation (PVAI)- can lead to cure, but is best suited for paroxysmal AF, in which ectopic beats arising from the pulmonary veins were shown to initiate AF. PVAI success is lower in persistent AF, in which the role of the cardiac autonomic system, particularly the intrinsic cardiac ganglia, is being increasingly recognized. Expanding the ablation lesions to include greater areas the left atrial (LA) anatomy marginally improves outcomes, but also leads to increases in procedural complexity and duration, need of repeat procedures, and complications such as atrial flutters, particularly perimitral flutter (PMF). The investigators have developed a technique to perform rapid ablation of atrial tissues in AF using ethanol infusion in the vein of Marshall (VOM), and have shown: 1) Effective, rapid and safe tissue ablation of LA tissue neighboring the LA ridge and left inferior PV; 2) Regional LA vagal denervation by reaching the intrinsic cardiac ganglia; and 3) Facilitation of cure of PMF by ablating most of the mitral isthmus.

The investigators propose to evaluate outcomes differences yielded by VOM ethanol when added to conventional PVAI. The specific aims are: #1.To assesses the impact of VOM ethanol infusion in procedure success when added to de novo catheter ablation of persistent AF. The investigators will randomize patients with persistent AF undergoing a first AF ablation to standard PVAI vs. a combined VOM ethanol infusion plus PVAI (VOM-PV) #2. To assess the impact of VOM ethanol infusion added to repeat catheter ablation of recurrent AF after a failed ablation. Patients undergoing a repeat procedure for persistent AF after a failed PVAI will be randomized to either PVAI or VOM-PV as their repeat procedure. End points will include freedom from symptomatic or electrocardiographic AF after 12-15 months.

Although the risk of stroke is comparable in persistent and paroxysmal AF, the prevalence of persistent AF increases dramatically with increasing age, and thus is an overall more significant cause of morbidity and mortality. In the United States, there are currently an estimated 3.0 million adults with AF, and this number is expected to double in the next 25 years. Hospitalizations with a primary diagnosis of AF are close to half a million per year, which generates a tremendous economic burden on the health care system. When compared to health care costs of non-AF control subjects, patients with AF have greater annual healthcare costs (up to $8705 total annual incremental cost). On the basis of current prevalence data, it is estimated that AF leads to a national incremental health care cost of up to $26 billion. Inadequacy of pharmacological treatment options for persistent AF Management strategies are directed at heart rate control and stroke prevention -mere palliation- or at rhythm control. It has been shown that rhythm control strategies using antiarrhythmic drugs offer no benefit in elderly patients or patients with heart failure. Most of the lack of benefit of such rhythm control strategy is thought to be due to the adverse effects and suboptimal efficacy of antiarrhythmic drugs that can potentially augment mortality. Indeed, preservation of normal sinus rhythm is associated with decreased mortality. Dronedarone, the only antiarrhythmic drug shown to improve outcomes in nonpermanent AF compared to placebo, has been shown to double mortality, stroke and hospitalization for heart failure in the PALLAS study in patients with permanent. Thus, antiarrhythmic drugs remain suboptimal at best for the treatment of AF.

Shortcomings of catheter ablation of persistent AF Weak mechanistic rationale: Isolation of the pulmonary vein (PVs2) and adjacent LA (PV antrum) is the accepted procedural endpoint, based on the mechanistic concept that atrial extrasystoles arising from the PVs initiate paroxysmal AF. Other, non-PV triggers have been demonstrated.36 The link between PV extra systoles and AF is clear in paroxysmal AF, but not in persistent AF, in which the mechanisms of AF seem to be related more to a chronic atrial substrate than to acute triggers.4 Indeed, intramural reentry in the posterior LA seems to be particularly relevant in chronic models of AF. In persistent AF, the procedure has evolved, rather simplistically, to include additional lesions -besides isolation of the PVs, variably placed in the posterior wall, LA roof, and towards the mitral annulus, the superior vena cava,44 left atrial appendage, and other areas where complex fractionated atrial electrograms (CFAE) may be mapped. This brute force approach of simply destroying more tissue has yielded additional success, but new procedural targets with solid mechanistic bases are needed.

Suboptimal success and need for repeat procedures. Despite the additional tissue destruction, ablation success in persistent AF is with much lower than in paroxysmal AF, with single procedure success reported as low as 27%, 36%, or 49%, but up to 61% or 67%, depending on study heterogeneities in: definitions of persistent AF and of recurrence of AF, the type of AF monitoring, and ablation technique and operator experience. In order to achieve overall acceptable success rates, (which can reach up to 79%-94%), there is a consistent need for repeat procedures (sometimes up to 4) and the concomitant use of antiarrhythmic drugs. The rate of repeat procedures in experienced centers can reach up to 70 to 80%.PMF after catheter ablation of persistent AF. Clinical failures of a first ablation procedure are caused, in a significant portion of patients, by atrial flutters, rather than recurrent AF, and recurrence as flutter portends a greater chance of success in a second procedure. Such atrial flutters may be caused by perimitral reentry in up to 33-60% of the patients. Catheter ablation of PMF involves the creation of a linear lesion from the mitral annulus to the left inferior PV (the so-called mitral isthmus).Achieving a complete ablation (defined by bidirectional conduction block across the ablation line) can be very difficult, with success rates reported as 32%, 64%, or 71%. It sometimes requires ablation inside the coronary sinus (CS), in close proximity to the circumflex coronary artery, which could be damaged of note, incomplete ablation of the mitral isthmus is proarrhythmogenic, increasing the risk of recurrent flutter by up to 4 times.

Inclusion Criteria:

1. Patients between the ages of 21 and 85 years
2. Diagnosed with symptomatic persistent AF Documentation of history of AF for at least 6 months AF not spontaneously converting to sinus rhythm, persisting for ≥7 days Sinus rhythm after cardioversion is NOT exclusion, provided that≥2 episodes of persistent AF occurred in the previous 6 months
3. Resistant or intolerant to at least one class I, II, or III anti arrhythmic drugs (AAD)
4. Patients deemed candidates for radio frequency(RF) ablation of AF
5. Able and willing to comply with pre-, post-, and follow-up requirements.

Exclusion Criteria:

1. Patients with previous PVAI procedure or left heart ablation procedure.

2. Left atrial thrombus.

3. LA diameter greater than 65 mm on long axis parasternal view, or left atrial volume more than 200 cc by MRI or CT.

4. Left ventricular ejection fraction < 30%.

5. Cardiac surgery within the previous 180 days.

6. Expecting cardiac transplantation or other cardiac surgery within 180 days.

7. Coronary percutaneous transluminal coronary angioplasty (PTCA)/stenting within the previous 90 days.

8. Documented history of a thrombi-embolic event within the previous 90 days.

9. Diagnosed atrial myxoma.

10. Significant restrictive, constrictive, or chronic obstructive pulmonary disease with chronic symptoms.

11. Significant congenital anomaly or medical problem that in the opinion of the investigator would preclude enrollment

12. Women who are pregnant.

13. Acute illness or active infection at time of index procedure documented by either pain, fever, drainage, positive culture and/or leukocytosis (WBC > 11. 000 mm3) for which antibiotics have been or will be prescribed.

14. Creatinine> 2. 5 mg/dl (or > 221 μmol/L, except for patients in dialysis).

15. Unstable angina.

16. Myocardial infarction within the previous 60 days.

17. History of blood clotting or bleeding abnormalities.

18. Contraindication to anticoagulation.

19. Contraindication to computed tomography or MRI procedures.

20. Life expectancy less than 1 year.

21. Uncontrolled heart failure.

22. Presence of an intramural thrombus, tumor, or other abnormality that precludes catheter introduction or positioning.

23. Presence of a condition that precludes vascular access.

24. Institute for Natural Resources (INR) greater than 3. 5 within 24 hours of procedure.

25. Cannot be removed from antiarrhythmic drugs for reasons other than AF.

26. Unwilling or unable to provide informed consent.

27. Current reported alcoholism.

    -

Individual participant data is not planned for sharing at this time.

Patients will be blinded to the randomization outcome