Study design:

Randomized prospective comparison of pulmonary vein isolation alone versus a strategy aiming for atrial fibrillation termination by using a stepwise approach consisting of pulmonary vein isolation, biatrial defragmentation aiming for fractionated electrograms and linear ablation if required.

Study protocol:

In patients randomized to PVI alone, the index procedure consists of wide circumferential pulmonary vein isolation (1 cm distal of the PV ostium) including optional right atrial isthmus ablation. Endpoint for PVI is entrance block as evaluated by a circumferential mapping catheter during sinus rhythm at the end of the procedure. In all patients the NavX® three-dimensional mapping system should be used. The St. Jude Medical Coolflex® or Coolpath® catheters may be used at the operator's discretion.

In patients randomized to the strategy aiming for atrial fibrillation termination PVI is also the initial step of the procedure followed by atrial defragmentation and eventual linear ablation in case of atrial tachycardia aiming for acute restoration of sinus rhythm. The procedural steps with regard to mapping and ablation of atrial fibrillation and consecutive atrial tachycardia follow the previously described methods (8,9). After initial ablation the previous antiarrhythmic drug treatment can be continued but no de novo antiarrhythmic drug regiment should be applied. The primary endpoint of this study is recurrence of atrial arrhythmias between 6 and 12 months after the initial ablation procedure.

In both study arms, the same CE certified catheters will be used. Usually, the setting consists of on decapolar catheter positioned in the coronary sinus, one circumferential mapping catheters for mapping of the pulmonary veins, one quadripolar catheter for mapping the right atrium. In both study groups a externally irrigated catheter for mapping and ablation will be used (CoolFlex/ Coolpath as mentioned above).

Sample size calculation:

According to the assumption of a difference of 30 % regarding the primary endpoint (power 80, alpha=0.05) 70 patients have to be included. Considering a potential drop out of 10 % 160 patients should be included. However, an adaptive statistical plan is applied (Prof. Wegscheider, University Hospital Hamburg).

Follow up and repeat procedures:

Crossover between the two treatment options should be avoided. Previous antiarrhythmic drug treatment can be continued until month 6 after the index procedure is completed. Repeat procedures between month 0 and 3 should be avoided. Repeat procedures should be planned between month 3 and 6 after the index procedure. The initial approach of the repeat procedure is the same as during the index procedure (Figure 1). Thus, in patients randomized for PVI alone, PVI will be repeated in case of reconnection of the veins. In the setting of complete block of all pulmonary veins also patients randomized to PVI alone will receive a substrate modification as in the other group. Patients randomized to the ablation procedure aiming for AF termination will again receive a 'complete' procedure including assessment of PVI and ablation of atrial tachycardias and atrial fibrillation.

Follow-up:

All patients will be seen in the outpatient clinic every 3 month during the 12 month study period. The visit will include ECG and Holter ECG recordings. Additionally, echocardiography and interrogation of pacemaker will be performed.

Inclusion Criteria:

• Age > 18
• Persistent atrial fibrillation > 7 days or previous cardioversion
• Obtained written consent

Exclusion Criteria:

• Age less than 18 years
• Previous operative or interventional treatment of atrial fibrillation
• Paroxysmal atrial fibrillation
• Pregnancy
• Contraindication for oral anticoagulation
• Contraindication for heparin administration
• Transient factors causing atrial fibrillation (e.g. Hyperthyroidism)
• Drug and alcohol abuse
• Severe underlying heart disease with left atrial enlargement > 60 mm