Risk of left atrial thrombus in patients undergoing catheter ablation for atrial fibrillation and atrial flutter

Abstract Background Catheter ablation (CA) for atrial fibrillation (AF) and atrial flutter (AFL) is contraindicated is left atrial thrombus (LAT) is found. The present registry aims to provide a contemporary estimate of LAT prevalence and to identify potential subsets of patients at higher risk. Methods All consecutive patients undergoing TOE before AF/AFL CA were enrolled between January 2019 and April 2024. Inclusion criteria were AF/AFL with less than 3 weeks optimal OAT, persistent or paroxysmal AF/AFL with CHA2DS2-VASc≥2. Risk factors associated with LAT were tested by regression analysis. Results 586 patients were enrolled (age 66±6 years; 72% males and 14 thrombi (2.4%), located in the left atrial appendage. Patients with LAT had a higher prevalence of coronary artery disease (CAD), heart failure (HF), and persistent AF (11 vs 3 patients, p = 0.048) as well as lower left ventricular ejection fraction (56 vs 49%, p = 0.01) and higher systolic pulmonary artery pressure (34 vs 29 mmHg, p = 0.028). No difference was found in patients according to adherence to oral anticoagulant therapy, left atrial echocardiographic characteristics or renal function. No LAT was found in patients with atrial flutter. According to univariate analysis (figure 1), CAD and HF were identified as predictors of LAT formation. LEVF was analyzed with a receiving operator characteristic curve (ROC) with an area under the curve (AUC) of 0.752 (95% CI 0.629-0.875; p=0.001) with the best cut-off at 50%. Moreover, LVEF less than 50% is a predictor of thrombi with an HR of 8.81 (95% CI; 2.89-26.87; p=0,001). Based on our data, we propose a protocol for TOE before CA in patients with AF as seen in Figure 2. Conclusions LAT is a rare occurrence in candidates for CA. Patients with persistent AF, history of CAD or HF present an elevated risk and should be considered for TOE prior to CA. Patients with AFL and no structural disease have an extremely low risk of LAT.Univariate table Proposed protocol