ABSTRACT
Objectives: Considering the risk of aerosolization during the COVID-19 pandemic associated with transesophageal echocardiography (TEE), we evaluated the diagnostic performance of cardiac computed tomography (CCT) before pulmonary vein isolation (PVI) in comparison to semi-invasive TEE in excluding left atrial (LA)/LA appendage (LAA) thrombus, limiting the need for TEE to only patients with possible thrombus on CCT. Material(s) and Method(s): We included a total of 145 consecutive patients with atrial fibrillation (AF) (age 52.4 +/- 10.3 years;63% males;89 paroxysmal AF) referred for radiofrequency ablation in National Cardiovascular Center Harapan Kita, Jakarta, Indonesia. All patients underwent preprocedural single-phase 128-slice multidetector CT and subsequent TEE as the reference standard with a mean time interval of 6.5 +/- 5.3 days between the two procedures. Result(s): CCT identified 30 patients (20.7%) with a filling defect in the LA/LAA, 8 of which were confirmed by TEE as thrombi (22 false positives and 8 true positives), whereas 9 true thrombi (6.2%) were detected by TEE (1 false negative by CCT). The sensitivity and specificity of CCT were 88.9% and 83.8%, respectively, with a positive predictive value of 26.8% and a negative predictive value of 99.1%. The overall accuracy was 84.1%. Conclusion(s): Apart from being a planned preparation modality before PVI, CCT is sufficient and could be used as an initial step to exclude the presence of LA/LAA thrombus, limiting the invasive TEE only for confirmation of the thrombus if detected by CCT.