Insertable cardiac monitor detection of silent atrial fibrillation in candidates for percutaneous patent foramen ovale closure.

Aim An underlying atrial vulnerability or a preexisting misdiagnosed atrial fibrillation in some patients who are candidates for patent foramen ovale (PFO) closure may lead to an unnecessary percutaneous intervention. The aim of this work was to define paroxysmal atrial fibrillation rate, through a 6-month insertable loop-recorder monitoring, in patients over 55 years old with cryptogenic stroke and PFO. Methods PFO closure criteria: significant right-to-left shunt and at least one high-risk feature (permanent right-to-left shunt, atrial septal aneurysm, prominent Eustachian valve, recurrent brain ischemia, previous deep vein thrombosis, thrombophilia). Insertable cardiac monitoring criteria: previous cryptogenic stroke, more than 55 years and at least one atrial fibrillation risk factor (heart failure, hypertension, age older than 65 years, diabetes, atrial runs, left atrium dilatation, left ventricle hypertrophy, pulmonary disease, thyroid disease, obesity). Atrial fibrillation detection threshold: arrhythmia duration longer than 5 min. Results From January 2008 to March 2017, 195 patients underwent to loop-recorder monitoring. A total of 70 (36%) patients were candidates for PFO closure. The 6-month silent atrial fibrillation rate was 11.4%. In the arrhythmia-free cohort, 28 patients (45.2%) underwent percutaneous foramen ovale closure (group A) and 34 (54.8%) were medically treated (group B). Atrial fibrillation detection rate was 14.3% in group A and 0% in group B. The 36-month atrial fibrillation-free survival was 76%. Conclusion An occult preexisting atrial fibrillation may lead to unnecessary percutaneous foramen ovale closure in a significant proportion of patients. A 6-month loop-recorder monitoring may improve the patient oriented decisionmaking.