ABSTRACT
BACKGROUND AND PURPOSE: The antithrombotic management of atrial fibrillation (AF) is currently based on clinical scores (CHADS(2) or CHA(2)DS(2)VASc). The prevalence of left atrium (LA) thrombi in effectively anticoagulated AF patients has been reported as being up to 7.7 %. We tried to correlate LA/LA appendage (LAA) thrombus detection with possible clinical predictors in warfarin-treated patients. METHODS: We performed trans-esophageal echocardiography on 430 patients (mean age, 60.3 ± 9.8 years) receiving oral anticoagulant (OAC) therapy and undergoing pulmonary vein isolation. In 10/430 (2.3 %), an LA thrombus was found despite therapeutic OAC (mean INR 2.6 ± 0.6; range, 2.0-3.8) over the previous 4 weeks. RESULTS: Two study groups were identified: 1. T-positive group = with LAA thrombus (10 patients) 2. T-negative group = without LAA thrombus (420 patients) The T-positive patients had a higher CHADS(2) score (1.5 ± 0.7 versus 0.7 ± 0.8; p = 0.004), a lower LVEF (54.7 ± 9.5 % versus 60.2 ± 7.4; p = 0.02), and a larger LA size (LA diameter, 56 ± 12.2 mm versus 46 ± 6.5 mm; p < 0.001and normalized LA volume: 140.2 ± 66 ml/m² vs. 67 ± 39 ml/m²; p < 0.05). On multivariate analysis, a larger LA diameter and normalized LA volume (OR, 1.14; 95 % C.I., 1.04-1.26; p = 0.006 and OR, 1.02; 95 % C.I., 1.01-1.03; p = 0.001, respectively) and a higher CHA(2)DS(2)VASc score (OR, 2.4; 95 % C.I., 1.4-4.2; p = 0.001) predicted left atrium appendage (LAA) thrombus. In another 42/430 (9.8 %) patients, an LA spontaneous echo-contrast (SEC) was detected. Thus, cumulatively, 52/430 (12.1 %) patients had either LAA thrombi (10 patients) or SEC (42 patients). LA diameter continued to predict the presence of either thrombi or SEC (OR, 1.14; 95 % C.I., 1.07-1.2; p < 0.05). CONCLUSIONS: We found a 2.3 % prevalence of LA thrombus (12.1 % when SEC was also considered). The thrombus was present despite on-target warfarin prevention. In addition to a higher CHA(2)DS(2)VASc score, a larger LA size was a strong predictor of clot detection.
