The cryoballoon catheter ablation (CB) for atrial fibrillation (AF) has evolved over the last several decades, and has become a reasonable alternative to radiofrequency (RF) pulmonary vein isolation (PVI).1 The efficacy of CB procedure in paroxysmal AF has been proven in the recent FIRE and ICE trial, where CB was non-inferior to RF ablation with respect to efficacy on AF free survival, but with no significant difference in overall safety.2 Shorter procedure/fluoroscopy time for contiguous transmural lesion, shorter learning curve, and higher quality of life would be major benefits of CB over RF ablation.2-4 However, this efficacy and safety of CB procedure has been proven only in patients with paroxysmal AF, and the role of CB in persistent AF remains largely unknown. 
Persistent AF has been known to be fundamentally different from paroxysmal AF.  Paroxysmal AF occurs primarily from the PV triggers while persistent AF is perpetuated by focal/reentrant mechanism in the process of atrial structural remodeling.5 Therefore, application of CB, due to its main effect on the PVs, has been regarded to provide only a limited effect in patients with persistent AF . However, the STAR-AF II trial demonstrated that substrate modification in addition to PVI either via linear ablation or via complex fractionated electrogram ablation do not result in reduction of AF recurrences in patients with persistent AF.6 This landmark study has provided a renewed insight into the role of PV in AF, and possible clinical application of CB in persistent AF patients. Furthermore, improved outcomes with extended CB procedures (CB plus) such as wide antral ablation, posterior wall isolation, or roof/perimitral linear lesionss have also suggested that CB could be applied in substrate modification in more advanced forms of AF.7-9 The use of CB in patients with persistent AF is rapidly growing in the current electrophysiology practice, and the new concept of extended CB ablation (CB plus) is being actively tested in selected large volume centers. In spite of the widespread use of CB, a direct comparison of RF vs CB in persistent AF in a prospective randomized way has been performed in one published10 and two ongoing studies (Cryoballon Versus Radiofrequency Ablation for Persistent Atrial Fibrillation: IRON-ICE, NCT03365700, and FIRE and ICE 2 trial pilot, NCT03706677). Comparison of CB vs CB plus has been performed in only two retrospective observational studies.8, 9 The review of previous observational data showed that CB or CB plus are potentially associated with better effectiveness than RF. Especially CB plus could be a better treatment option than CB in terms of prevention of recurrent AF/AFL. Besides, considering general advantage of CB or CB plus such as shorter procedure time and higher safety, CB+ could be a better treatment option than traditional RF. To better understand the role of CB in the management of persistent AF, and to measure the additional effect of extended CB ablation as opposed to CB alone, a trial comparing all the three arms is necessary. From this trial we could demonstrate the higher efficacy of CB plus compared to RF and optimal extent of CB ablation in persistent AF. To the best of our knowledge, clinical trials that directly compare the efficacy and safety of the three treatment options (RFCA, CB, and CB plus) at the same time have not been reported. In the current study, we hypothesized that the CB plus procedure would be more effective than CB or RF in the management of patients with persistent AF.

Treatment

- Randomization is carried out through an internet-based randomization system, available 24 hours per day, 7 days per week, 365 days per year
- All procedures will be performed via transseptal access to the Left Atrium (LA)
- After transseptal access, patients should be anticoagulated with intravenous heparin to maintain an ACT of > 300 sec
- Pre-ablation CT should be performed in all patients 
- In all patients pre and post pulmonary vein potentials should be recorded to confirm the isolation of pulmonary vein, with verification of entrance or exit block. 
-With any techniques, PV isolation should be performed as the gold standard. Additional line ablation or complex atrial fractionated electrogram (CAFÉ) ablation in RFCA group or posterior wall isolation in CB group can be done by the physician’s discretion, but strongly discouraged unless incessant LA tachycardia or recurrent IRAF is documented. 
- In CB plus group roof line ablation or posterior isolation should be performed. 
- Cavo-tricuspid isthmus ablation using irrigated radiofrequency catheter can be performed in all groups when CTI-dependent AFL was documented or induced. 
- If the patient is in atrial arrhythmia after completion of study procedure, sinus rhythm should be restored with DCCV by protocol
- inducibility test will be performed using isoproterenol (up to 10 mcg/min) with or without rapid atrial pacing (up to 180ms) to reveal the non-PV trigger as well as to assess the post-ablation AF inducibility. 19-22 The protocol for the inducibility testing may change depending upon the patient’s condition or attending physician’s decision,and should be recorded in the CRF. AF will be considered inducible if it persists for ≥1 minutes
- Ablation of non-PV trigger will be allowed if the repetitive triggers are noted during isoproterenol infusion.

Arm Label

Target Number of Participant

Arm Type

Arm Description

	The method of ablation, use of contact-force sensing catheter or automatic tagging modules will be directed by the participating center’s policy or physician’s discretion. Following ablation strategies are recommended for all participants. Linear ablation or CFAE ablations are strongly discouraged by the protocol   
	When CARTO system (Biosense-Webster, USA) was used: AI guided ablation akin to the “CLOSE protocol” is recommended.23 Using Navistar Smart-touch surround-flow catheter; delivering 25-40 watts of energy; 40 seconds time limit; 10-30 gram of contact force; 40 seconds of time limit; targeting AI value of 550 in anterior wall and 400 in the posterior wall; interlesion distance ≤6mm; using automatic lesion annotation (Visitag™) – type 3 tagging, 3mm stability for 5 seconds, force >5g over 50% of time; target AI value could be reduced (450 in anterior wall and 350 in posterior wall) according to the participating center’s policy.24-26 
	When Ensite NavX system (Abott, USA) was used: Using TactiCath™ Quartz ablation catheter; delivering 25-35 watts of energy; 540 seconds time limit; 10-30 gram of contact force; using automatic lesion annotation (Automark Module), target lesion index value of 5-5.5 in the posterior wall and 5.5-6 in the anterior wall.

Arm Label

Target Number of Participant

Arm Type

Arm Description

	The method of ablation, size of balloon, freezing time, and booster freezing will be made upon the participating center’s policy or physician’s discretion. Following ablation strategies are highly recommended for all participants. 27 Additional Linear ablation or CFAE ablations are strongly discouraged by the protocol
	Application of TTI + 2 minutes or 180 seconds single-freezing is recommended when time-to-isolation (TTI) <60 seconds. Additional 120 seconds of booster-freezing after TTI + 2 or 180 seconds initial freezing is recommended when TTI value >60 seconds, or PV potentials were not discernable in the Achieve multipolar catheter. If temperatures below <-40 ‘C cannot be achieved due to improper occlusion or TTI > 90 seconds, freezing should be discontinued and reposition of balloon is recommended. Segmental ablation is allowed when the proper occlusion of PV by balloon cannot be achieved. Continuous phrenic capture should be monitoring during procedure in the right PVs to avoid phrenic nerve injury. Stop ablation immediately if phrenic nerve damage is suspected. No further ablation would be performed when the phrenic nerve damage is suspected. Touch-up ablation using irrigated radiofrequency ablation catheter would be permitted with conventional mapping method.

Arm Label

Target Number of Participant

Arm Type

Arm Description

	The method of ablation, size of balloon, freezing time, and booster freezing will be decided by the participating center’s policy or physician’s discretion. Ablation strategies described above are highly recommended for all participants. 27 
	Additional posterior wall isolation should be performed by single 120 to 180 second cryo application delivered at each site. 
	Utmost debulking of posterior wall should be attempted although complete posterior wall isolation may not be successful, as we considered that a large scare created by the CB application would provide sufficient anti-arrhythmic effects in terms of suppressing AF. The number and duration of cryo application should be recorded in the case report form.  
	Successful posterior wall isolation should be determined by one of the following methods. The method and achievement of posterior wall isolation should be recorded in the case report form.
	Entrance and Exit block of posterior wall with high-voltage pacing from achieve catheter or
	Bidirectional block of roof line by differential pacing or
	Reduction of the electrical activity (<0.1mV) in the left atrial posterior wall in the 3D-mapping system

Gender

Age

Description

- Patients with age between 18-80 years
AND
- Patients with persistent (≥7 days) AF documented on 12-lead ECG or Holter monitoring. 
AND
- AF refractory/intolerant to class I or III antriarrhytmic drug	
AND
- Able and willing to provide written informed consent to CRYO or RFCA and participation in this investigation

- Patients refusing participation in the registry
- Age< 18 years old or > 80 years old
- Paroxysmal AF which lasting < 7 days
- Mitral stenosis or mechanical prosthetic valve
- Patients with left atrial size ≥ 55 mm (2D echocardiography, Anterior-posterior diameter in parasternal long axis view)
- Anatomy of pulmonary vein was not suitable for CB or CB+, including common ostium or ostium size > 26mm
- Pregnant woman of child-bearing age with a positive pregnancy test before treatment
- Presence of atrial septal defect or patent foramen ovale closure device
- Presence of intracardiac thrombus
- Contraindications to the systemic anticoagulation
- NYHA functional class IV heart failure
- Prior catheter or surgical ablation of AF
- Acute coronary syndrome within 3 months
- Planned open heart surgery within 3 months
- Prior open-heart surgery within 3 months
- End stage renal disease or chronic kidney disease (estimated glomerular filtration rate [MDRD method] < 30mL/min/1.73m)
- Life expectancy less than 1 year

Freedom from documented atrial arrhythmia episodes lasting > 30 seconds without antiarrhythmics medications (after 3 months of blanking period)

3 months after index procedure

-	Atrial arrhythmia recurrence during blanking period (3 months) after one or two procedures with/without antiarrhythmic medications -	Atrial tachycardia or flutter recurrence during long-term follow-up ) after one or two procedures with/without antiarrhythmic medications -	Atrial fibrillation recurrence during long-term follow-up) after one or two procedures with/without antiarrhythmic medications -	Incidence of peri-procedural complications (including stroke, cardiac tamponade, esophageal injury, phrenic nerve damage, and death)  -	Rate of first-pass isolation or first-freeze isolation (%) -	Procedure duration (minutes) -	Ablation time (minutes) -	Fluoroscopy time (minutes) -	Procedure time (minutes, skin to skin) -	LA dwelling time  -	Ablation time (seconds) -	Quality of life changes at 12 months compared to baseline  -	Number of repeat procedures -	AF burden at 12 months after ablation assessed by 2-week patch monitoring -	LA pressure, mmHg (max/min/mean) after septal puncture and after procedure  -	Role of 2-week patch electrode in the detection of post-procedure atrial tachyarrhythmia in comparison to conventional Holter monitoring

3 and 12 months after index procedure