Clinical features of patients with left atrial thrombus undergoing anticoagulant therapy

Left atrial (LA) thrombus was sometimes found by transesophageal echocardiography (TEE) in non-valvular atrial fibrillation (AF), even in the setting of continuous warfarin therapy. A D-dimer cutoff level of 0.50 μg/mL appears to be a useful marker to rule out venous vein thrombosis. This study analyzed the clinical features of patients with LA thrombi who received anticoagulant therapy and whether the D-dimer test is useful to exclude LA thrombus.

Two hundred twenty-five consecutive patients with AF (149 with paroxysmal and 76 with persistent) were enrolled. All patients received continuous warfarin therapy with the prothrombin time-international normalized ratio (PT-INR) between 1.6 and 3.0 for more than 3 months and TEE was performed.

LA thrombi were present in 23 and absent in 202 patients. Age, gender, and PT-INR (1.96 ± 0.59 vs. 1.98 ± 0.53) were not different between the two groups. Persistent AF (65 vs. 30%; p < 0.005), LA diameter (44 ± 5 vs. 40 ± 7 mm; p < 0.005), ejection fraction (57 ± 13 vs. 65 ± 9%; p < 0.005), brain natriuretic peptide levels (203 ± 209 vs. 105 ± 166 pg/mL; p < 0.05), D-dimer (0.55 ± 0.70 vs. 0.16 ± 0.20 μg/mL; p < 0.001), LA appendage flow velocity (33 ± 15 vs. 54 ± 19 cm/s; p < 0.001), CHADS(2) scores (2 ± 1 vs. 1 ± 1; p < 0.005), and CHA(2)DS(2)-VASc scores (3 ± 2 vs. 2 ± 1; p < 0.005) were significantly different in both groups. Although multivariate analysis showed that D-dimer and LA appendage flow velocity were significant independent predictors of LA thrombus, D-dimer levels below 0.5 μg/mL were found in 19 of 23 patients with LA thrombi.

D-dimer levels below 0.5 μg/mL is not enough to rule out LA thrombus in AF patients with well-controlled anticoagulation.