Determinants of left atrial thrombus or spontaneous echo contrast in nonvalvular atrial fibrillation

Abstract

BackgroundThe CHADS2 and CHA2DS2-VASc scores are well-established clinical scales to estimate the risk of stroke in patients with atrial fibrillation (AF). However, the predictive power of the two scales concerning left atrial thrombus (LAT) or spontaneous echo contrast (SEC) has not been well investigated. Therefore, we investigated the predict power of CHADS2 and CHA2DS2-VASc scores concerning LAT/SEC; identified clinical, echocardiographic and laboratory predictors of LAT/SEC in addition to the CHADS2 and CHA2DS2-VASc scores; and derived a new scale to predict LAT/SEC accurately, it might improve thromboembolic risk stratification in patients with nonvalvular atrial fibrillation. MethodsWe identified 1102 consecutive AF patients who underwent transesophageal echocardiography (TEE) for the purpose of the exclusion of LAT before catheter ablation, cardioversion or left atrial appendage occlusion. The clinical, echocardiographic and laboratory characteristics of patients were collected from the electronic medical record system. ResultsIn the study, the prevalence of LAT/SEC was only 4.36%. In the multivariate logistic analysis, hypertension, left atrial enlargement, prior stroke/TIA, left ventricular dysfunction, and renal dysfunction were predictors of LAT/SEC. Receiver operating characteristic curve analysis showed that c-statistics of the CHADS2 and CHA2DS2-VASc scores concerning LAT/SEC were 0.673 and 0.643, respectively. We derived a new scale composed of variables from the multivariate logistic analysis that showed a higher c-statistic value (0.761) than the CHADS2 and CHA2DS2-VASc scores for the prediction of LAT/SEC. ConclusionIn our cohort, we found two variables not included in the CHA2DS2-VASc score (renal dysfunction, left atrial enlargement) were independent predictors of LAT/SEC. A new scale combining clinical, echocardiographic and laboratory predictors might improve thromboembolic risk stratification.And there is a great need to carry out a new prospective and multicenter study, with a population more homogenous and including all the determinants for LAT/SEC to establish the independent degree of each variable and the applicability in clinical practice, facilitating the emergence of a new score of thromboembolic risk in patients with nonvalvular atrial fibrillation.