Predictive value of thromboembolic risk scores before an atrial fibrillation ablation procedure

It is not clear whether transesophageal echocardiography (TEE) should be performed prior to a planned atrial fibrillation (AF) ablation in all patients.

The objectives of this study were to determine in 681 consecutive patients: (i) the relationship between the CHADS2 and CHA2DS2-VASc scores, the presence of a thrombogenic milieu and left atrial (LA) volume; (ii) the need for TEE in patients with low and intermediate thromboembolic risk assessed; and (iii) the predictive accuracy of the these 2 scores for the presence of thrombi in the LA/LAA (LA appendage) before a planned AF ablation. The prevalence of thrombi was 1%. All patients with thrombi had LA dilatation, a CHADS2 score ≥ 1 and a CHA2DS2-VASc score ≥ 2. CHADS2 or CHA2DS2-VASc scores <2 had an almost maximal negative predictive capability of excluding the presence of a thrombus (99.8% and 100%, respectively; 95% CI: 99-100). A CHADS2 score ≥ 2 had a sensitivity and specificity of 86% (95% CI: 42-100) and 82% (95% CI: 79-85), respectively, to predict the presence of a thrombus in the LA/LAA, while a CHA2DS2-VASc score ≥ 2 had a sensitivity and specificity of 100% (95% CI: 59-100) and 67% (95% CI: 63-70). The area under the curve for CHADS2 and CHA2DS2-VASc scores ≥ 2 was 0.928 (95% CI: 0.906-0.946) and 0.933 (95% CI: 0.912-0.951), respectively.

Not all patients undergoing planned endocardial pulmonary vein isolation need preprocedural TEE. Both scores <2 had an almost maximal negative predictive capability of excluding the presence of a thrombus in the LA/LAA.