2.1

Intracardiac echocardiography protocol

A 9F 9-MHz UltraICE catheter (EP Technologies, Boston Scientific, San Jose, CA, USA) was used to perform a complete intracardiac study as previously described .

The intracardiac echocardiography protocol (ICE) study was conducted by performing a manual pullback from the superior vena cava to the inferior vena cava through five sectional planes and measurement of diameters of the fossa ovalis; the entire atrial septum length and rims were obtained with electronic caliper edge-to-edge on the aortic valve plane and the four-chamber plane. PFO tunnel length was carefully measured. Atrial septal aneurysm has been classified by intracardiac echocardiography following the classification of Olivares et al. . PFO channel was distinguished in short (<9 mm) and long (>10 mm) channel. Intracardiac echocardiographic monitoring of the implantation procedure was conducted on the four-chamber plane. However, decision about device to be used was made after intracardiac study.

2.2

Closure protocol

On the basis of ICE findings the operators selected the Amplatzer Occluder family (PFO Occluder, Cribriform Occluder, AGA Medical, Golden Valley, MN, USA) or the Premere Closure System (St. Jude Medical GLMT). The Amplatzer PFO occluder has been selected when ASA was bidirectional but moderate (3RL or 3LR ASA), whereas the Premere occlusion system has been chosen when ASA was absent or in case of motionless or unidirectional ASA (1R, 2L ASA), and when PFO tunnel length was >10 mm. The Amplatzer Cribriform Occluder has been selected in case of multiperforated ASD, keeping attention to cross with the guidewire the most central hole in the fossa ovalis under ICE guidance as previously described . This kind of device was selected also when ASA was huge and bidirectional (5RL ASA) in order to obtain a complete coverage of the fossa ovalis at both sides of the atrial septum.

2.3

Follow-up protocol

In patients who underwent closure, transesophageal echocardiography was scheduled at one month and repeated at 6 months after closure if there was no more than a trivial shunt. Patients underwent transcranial Doppler echocardiography and magnetic resonance imaging of the brain 1 month after the procedure. Transthoracic echocardiography was scheduled at 6 months and 1 year after the transcatheter closure. Any residual shunt was graded as trivial, small, moderate and severe, as previously described . Clinical cardiological and neurological visits were scheduled at 1, 6, and 12 months.

2.4

End points definition

Procedural success was defined as ability to release the device in a stable position on fluoroscopy and ICE guidance with no more than a trivial shunt. Complications included groin haematomas of any grade, atrial wall perforation, pericardial effusion, entrapment of device or sheath or ICE equipment through venous valves or embryonic remnants, and air embolism. Follow-up failure was defined as device removal or residual shunt >small.

Migraine with aura improving was set as decrease in MIDAS score of at least 50% and decrease in antiheadache drugs number of at least 50%. Migraine with aura abolition was defined as abolition of migraine attacks with no other medications than Aspirin 100 mg/day.

2.5

Statistical analysis

chi-square, corrected chi-square, and Student’s t tests were used to compare frequencies and continuous variables: level of significance was set to P value <.05.