P1895Predictors of left atrial thrombosis in patients undergoing planned electrical cardioversion of atrial fibrillation treated with direct anticoagulant

Abstract

Abstract Background Scheduled electrical cardioversion (CV) of persistent atrial fibrillation (AF) can be performed after an adequate period of oral anticoagulation (OAC) therapy to reduce thromboembolic complications. Exclusion of left atrial appendage (LAA) thrombi can be achieved by transoesophagel echocardiography (TOE), that is indicated if early CV is desired in patients (pts) with ≥48 hours-lasting arrhythmia, while is not recommended after at least 3 weeks of OAC. Purpose We investigated with TOE the presence of LAA thrombi in a consecutive series of ambulatory pts scheduled for electrical CV of AF and treated with DOAC for at least 3 weeks. Methods We evaluated 93 pts referred to our Clinic for CV of persistent AF (October 2017–December 2018), all treated with DOAC. We collected clinical history, physical examination, 12-lead ECG, lab tests and transthoracic echocardiography measures. Right before cardioversion we systematically performed TOE to exclude LAA thrombi. If absent, we proceeded to CV, while in case of grade III echo-contrast and/or thrombotic stratification we didn't perform the procedure. All values are reported, as appropriate, as mean±SD or number and percentage. All results were considered statistically significant when p<0.05. The entire analysis was performed using the Statistical Package for Social Sciences package, version 19.0 (SPSS, Chicago, Illinois). Results 76 out of 93 pts (82% – group A) were free from intracavitary thrombi and underwent successful electrical CV; the other 17 pts (18% – group B) had thrombi in LAA that contraindicated electrical CV. Group B pts (74±10 years old, 65% male sex) had higher CHA2DS2-VASc score (4,1±2 vs 3,1±1,4, p=0,014), they were significantly more affected by chronic kidney disease (CKD, 59% vs 32%, p=0.035), peripheral artery disease (PAD, 35% vs 12%, p=0.017), they had larger LA dimension (48±5 vs 45±5 mm, p=0,09 - non significant) and as expected they had reduced LAA emptying velocity (76% vs 30%, p<0.001) compared to group A pts (73±8 years old, 70% male sex). There was no significant difference neither in duration (median 7 weeks) nor in type of DOAC therapy between the two groups. Conclusions Current Guidelines recommend at least three weeks of OAC before electrical CV of persistent AF. TOE is recommended only if early cardioversion is needed. We observed that in a significant part of pts treated with DOAC for more than 3 weeks TOE still identified LAA thrombi, especially in pts with multiple comorbidities. Even if we currently don't have data on clinical endpoints in this population, CV of patients with LAA thrombi despite adequate DOAC therapy may lead to an increased risk of thromboembolic events. We think it's necessary to identify more accurate predictors (i.e. CKD) to stratify the thromboembolic risk. Further randomized studies are needed to identify whether there are patients in which CV shouldn't be performed without TOE, irrespective of anticoagulation duration.