Transesophageal echocardiograpy in patients with persistent atrial fibrillation undergoing electrical cardioversion on new oral anticoagulants: A multi center registry

Increased risk of thrombo-embolism following cardioversion foratrial fibrillation (AF) is well recognized. Therefore, anticoagulation isconsidered mandatory before elective cardioversion for AF of N48 h orAF of unknown duration. Vitamin K antagonists (VKAs) (INR 2.0 to3.0) are recommended for at least 3 weeks prior to and 4 weeks aftercardioversion, regardless of the CHA2DS2-VASc score and the method(electrical or pharmacological) used to restore sinus rhythm [1,2].Therisk of thromboembolic events is high (up to 7%) if anticoagulation isinadequate, and is reduced to less than 1% if adequate anticoagulationis achieved [3]. Recently, new oral anticoagulants (NOACs) have beenintroduced intheclinical practiceandsuggestedalso inpatientsunder-going electrical cardioversion for AF. In this context compliance andadherence to treatment are crucial, as these drugs have a relativelyshort half-life. In fact patients missing more than one drug assumptioncannotbeconsideredunderanadequatetherapeuticregimen.Moreover,differently from VKA treatment, that can be followed through INRmeasurements,there are no effective hemoreological tests to assessther-apeutic anticoagulation. One way to assess the ef ficacy of anticoagulationtherapy is the direct search for left atrial thrombus by means oftransesophageal echocardiography (TEE). The aim of our study was toevaluate the incidence of left atrial (LA) thrombus in patients referredfor electrical cardioversion, while on treatment with NOACs, using apre-procedural TEE.Between January 2014 and December 2014 we collected the TEEdata of 219 consecutive patients with persistent AF referred to sevenItalian centers for elective electrical cardioversion, while on NOACs forat least 3 weeks before electrical cardioversion. This study followedtheprinciplesoutlinedinthelatestupdateoftheDeclarationofHelsinkiand all patients signed informed consent. This study was approved byour Institutional Review Boards. Main exclusion criteria were a historyof thromboembolic events, a history of LA thrombus diagnosed byTEE, a TEE procedure performed in the 21 days before starting NOACtherapy, and the need for anticoagulation other than AF. TEE wasperformed with a 5–7-MHz multiplane transducer connected to anultrasoundsystem.Multiplestandardtomographicplaneswereimaged.TheLAappendagepeakflowvelocity,presenceofthrombiintheLA,andseverityofspontaneousechocontrastintheLAweredetermined.Atrialcavityorappendagethrombiwereconsideredtobepresentwhenwell-circumscribed, echodense, intracavitary masses that were acousticallydistinct from the underlying endocardium and pectinate muscles wereidentified.The clinical characteristics of the study population are summarizedin Table 1. The mean duration of AF was 6.2 ± 4.3 months, and themean duration of NOAC therapy was 4.1 ± 3 months. NOACs usedwere dabigatran (86 patients, 39%), rivaroxaban (61 patients, 28%),and apixaban (73 patients, 33%). The daily dabigratan dose was220mgin21patients,and300mgin65patients;thedailyrivaroxabandose was 15 mg in 2 patients, and 20 mg in 59 patients; the dailyapixaban dose was 5 mg in 1 patient, and 10 mg in 72 patients. The