120 consecutive patients with paroxysmal AF, slated for PV isolation, will be randomly assigned to RF or cryoballoon ablation (in a 1:2 allocation scheme). Real-time 3D echocardiography (RT3DE) will be performed before the ablation procedure and 1 month post-ablation. The RT3DE datasets will be stored digitally and quantitative analyses will be performed off-line at a core echo laboratory. Quantification of left atrial volumes will be performed and left atrial functional quantification will be performed by calculating the following indices: Left atrial ejection fraction, Active atrial emptying fraction, Passive atrial emptying fraction, Atrial expansion index. All patients will be hospitalized for 48 hours post-ablation. Blood samples will be obtained at 12-hour intervals to measure troponin T with a high-sensitivity assay, as well as other biochemical parameters.

The study will involve 2 study interventions, cryoballoon and RF ablation, to which patients will be randomly assigned (in a 2:1 allocation scheme). Randomization will be performed centrally using a random number generator - the following short script will be used in R language to generate a random sequence with a 2:1 probability of assignment to cryoballoon versus RF.

Each patient's randomization will be known only to the operator of the ablation procedure (and of course to the investigator who will attend the procedure to record procedural parameters, complications, outcome etc.). All personnel involved in patient follow-up and echo data analysis will be blinded to patient randomization (patient records will be accompanied by a randomization number corresponding to a masked randomization list kept in digital form at the coordinating center, which will be unlocked after completion of the last patient's follow-up). The Ablation Procedure Data Sheet will be filed separately from other CRF documents to maintain blinding.

1. RF ablation Antral PV isolation will be performed in all patients randomized to RF ablation, without additional ablation of extrapulmonary sites, with the exception of individuals with documented typical atrial flutter.

   An irrigated radiofrequency ablation catheter will be used to perform ablation, with the aid of electroanatomical mapping (Carto 3, Biosense Webster), following a single transeptal puncture. Pulmonary vein potentials will be recorded with a circular mapping catheter (Lasso® NAV eco, Biosense Webster) before, during and following antral ablation (the default strategy will be to create two ablation lines, around the two ipsilateral pulmonary veins, unless left atrial anatomy dictates another approach). The endpoint of the ablation procedure will be entrance and exit block in all pulmonary veins. A waiting time of 20-30 minutes will be observed after initial pulmonary vein isolation and further ablation will be performed in case of reconduction between the veins and the atrium.

   Patients with a history of documented sustained typical atrial flutter before the ablation procedure will additionally undergo cavotricuspid isthmus ablation. Use of additional modalities, including intracardiac ultrasound and integration of computed tomography or magnetic resonance imaging anatomical data of the left atrium, will be left to the discretion of the operators.

   Repeat ablation will not be allowed over the 3-month follow-up of the study.

2. Cryoablation A single big cryoballoon approach, using a 28-mm cryoballoon (Arctic Front Advance™ Cardiac CryoAblation Catheter, Medtronic, Minneapolis, MN) will be employed. The cryoballoon catheter will be introduced into the left atrium, following a single transeptal puncture, through a 12F FlexCath steerable sheath (Medtronic), constantly flushed with heparinized saline. A circular mapping catheter (Achieve, Medtronic) will be advanced through the cryoballoon to the PV orifice and positioned as proximally as possible inside the vessel to record the PV potentials at baseline and monitor the isolation procedure in real time.

The cryoballoon will be inflated and advanced to the ostium of each PV. The quality of vascular occlusion will be ascertained by injection of diluted contrast material into the PV. Once the best occlusion is obtained, cryothermal energy will be applied for 240 seconds, as recommended by the manufacturer. An additional application of energy will be systematically delivered after PV isolation.

Before ablation of the right-sided PVs, a quadripolar electrode catheter will be positioned in the superior vena cava to constantly pace the right phrenic nerve at a 2-sec cycle length during freezing. In case of cessation or weakening of diaphragmatic contraction, freezing will be immediately discontinued.

iv) Real-time 3D echocardiographic evaluation of left atrial dimensions and function Real-time 3D echocardiography (RT3DE) will be performed before the ablation procedure and 1 month post-ablation, on iE33 machines (Philips Medical Systems, Bothell, Washington) equipped with X3, a fully sampled matrix transducer. Apical full-volume data sets will be obtained during end-expiratory apnea within 1 breath hold. The RT3DE datasets will be stored digitally and quantitative analyses will be performed off-line at a core echo laboratory (at a different site than the recruiting center) by personnel blinded as to each patient's treatment allocation.

Quantification of left atrial volumes will be performed using the semiautomated contour tracing algorithm, marking 5 atrial reference points: 4 at the anterior, inferior, lateral, and septal parts of the atrial dome and 1 at the level of the mitral annulus. Volumes will be measured at three time points during the cardiac cycle: (a) LAmax at end-systole, just before mitral valve opening, (b) LAmin at end-diastole, just before mitral valve closure; and (c) LApreA obtained at the time of the P wave on the surface electrocardiogram. Indexing to body surface area will be applied, as recommended.11

Left atrial functional quantification will be performed by calculating the following indices:

• Left atrial ejection fraction: LAEF = [(LAmax - LAmin)/LAmax] * 100%

• Active atrial emptying fraction: LAactive = [(LApreA - LAmin)/LApreA] * 100%

• Passive atrial emptying fraction: LApassive = [(LAmax - LApreA)/LAmax] * 100%

• Atrial expansion index: LAreservoir = [(LAmax - LAmin)/LAmin] * 100%. Echocardiographic evaluation will also include all standard 2D and Doppler measurements. Measurements of particular interest will be: left ventricular ejection fraction (modified Simpson's rule), mitral E/e' velocity ratio (lateral aspect of the mitral annulus), semi-qualitative assessment of mitral regurgitation (grade 0 to 3: no regurgitation, mild, moderate, severe), right ventricular systolic pressure.

  v) Other study procedures Before the index procedure, in a run-in visit, eligibility criteria will be reviewed and a 24-hour Holter recording will be undertaken.

All patients will be hospitalized for 48 hours post-ablation. Blood samples will be obtained as 12-hour intervals to measure troponin T with a high-sensitivity assay, as well as other biochemical parameters (blood samples will be centrifuged and plasma aliquots will be kept at -80oC until measurement).

Patients will return for 2 visits after the ablation procedure: a 1-month visit for RT3DE, symptom review, assessment for potential complications, clinical and electrocardiographic evaluation, and a 24-h Holter recording, and a 3-month visit for symptom review, assessment for potential complications, clinical and electrocardiographic evaluation, and a 24-h Holter recording.

Inclusion Criteria:

• At least two AF episodes within the last 12 months (either self-terminating within 7 days or cardioverted, medically or electrically, in less than 48 hours),
• AF episodes should have been symptomatic on at least 2 occasions within the last 12 months,
• Failure of at least one class I or III antiarrhythmic to prevent AF paroxysms,
• Age 40-80 years old.

Exclusion Criteria:

• Previous left atrial ablation procedure,
• Left atrial diameter >50 mm on TTE (parasternal long axis view),
• Known primary electrical heart disease (e.g. Brugada syndrome),
• Presence of atrial thrombus,
• Prosthetic valve at any position,
• Moderate/severe mitral stenosis,
• Severe mitral regurgitation,
• Active infectious disease or malignancy,
• Moderate or severe hepatic impairment (Child-Pugh class B or C),
• Severe renal failure (estimated glomerular filtration rate <20 ml/min/1.73 m2),
• Participation in a different research protocol (current or within 1 year),
• Inability or unwillingness to adhere to standard treatment or to provide consent.