he major complications of atrial fibrillation (AF) are hemodynamic compromise and thromboembolism; 6-24% of all ischemic strokes have been attributed to AF. 1,2 In these patients, the left atrium (LA), especially the left atrial appendage (LAA), is the presumed site of thrombus formation and a source of arterial thrombus. 3 Transesophageal echocardiography (TEE) is a potentially useful diagnostic imaging modality because it enables reliable assessment of LA thrombus and LAA flow velocity (LAAFV), 4 but the ability of LAAFV alone to predict LA and LAA thrombus in patients with AF remains controversial. [5][6][7][8] Although the risk of embolism is thought to be lower with atrial flutter (AFL), 9-11 recent TEE studies have demonstrated an unexpectedly high incidence of atrial thrombus formation. 12-15 LAA blood drainage is an important factor in the pathogenesis of local flow stasis and thrombus formation, and we hypothesized that in both AF and AFL the LAA blood drainage was reduced when the frequency of LAA movement increased. Thus, we examined whether or not LAA flow time (LAAFT), the new index of LAA movement and defined as the average duration of LAA flow with emptying and filling waves, is a useful predictor of thrombus formation.The aim of the present study was to evaluate the relationship between LAA function and the incidence of LA thrombus in patients with nonvalvular chronic AF and AFL.
Methods
Study PatientsFrom April 1995 to May 2000, we retrospectively reviewed patients with nonvalvular chronic AF and AFL who had undergone TEE. Patients with left ventricular (LV) systolic dysfunction (LV ejection fraction <0.5) and patients undergoing anticoagulant therapy were excluded. Nonvalvular chronic AF was defined by conventional electrocardiogram (ECG) on 2 occasions separated by at least 1 month and by the absence of rheumatic heart disease as determined by echocardiography. Lone AF was defined by excluding coronary artery disease (based on clinical or laboratory criteria), hyperthyroidism, valvular heart diseases, congestive heart failure, cardiomyopathy, chronic obstructive pulmonary disease, cardiomegaly on chest X-ray, history of hypertension, age over 60 years, insulin-dependent diabetes, AF only during trauma or surgery, and acute medical illness. 16 Sixty-two patients (53 male, 9 female; mean age, 60.0± 9.7 years) entered the study: 50 had chronic AF and 12 had AFL. The patients with AF were divided into 2 groups: lone AF (n=14) and non-lone AF (n=36). The relationship between the left atrial appendage (LAA) function, as assessed by transesophageal echocardiography, and the incidence of left atrial thrombus was evaluated in 62 patients with nonvalvular chronic atrial fibrillation (AF; n=50) and atrial flutter (AFL; n=12). It was hypothesized that in both AF and AFL not only the LAA flow velocity (LAAFV), but also the frequency of the LAA movement (the LAA flow time, LAAFT) is a major contributing factor to thrombus formation. LAAFT was defined as the average duration of LAA flow with emptying and fil...
