Abstract 18895: Left Atrial Ejection Fraction (LAEF) Predicts Cardioembolic Stroke In Patients With Chronic Atrial Fibrillation
Background: Recently, it was reported that decreased left atrial ejection fraction (LAEF) is a marker of left atrial dysfunction. However, the feasibility of LAEF remains to be determined. We investigated whether LAEF predicts cardioembolic stroke in patients with chronic atrial fibrillation.
Methods: We performed transthoracic echocardiography (TTE) and transesophageal echocardiography in 79 consecutive patients with chronic atrial fibrillation (male/female = 49/30, 70 ± 12 years). There were 30 patients with the history of cardioembolic stroke. LAEF was calculated by the biplane modified Simpson method from apical 4- and 2-chamber views at end systolic and diastolic phase. We classified study patients into 2 groups based on the history of cardioembolic stroke.
Results: There were no differences in gender, heart rate, proportion of warfarin therapy, prothrombin time and international normalized ratio, left atrial dimension (LAD), and the parameters of left ventricular function between 2 groups. The patients with cardioembolic stroke were older and had higher prevalence of spontaneous echo contrast (SEC), higher left atrial volume index (LAVI), lower left atrial appendage peak flow velocity (LAAeV) and lower LAEF compared with those without it. LAEF was significantly decreased in patients with left atrial appendage thrombus compared with those without it. LAEF showed a good correlation with LAAeV (R = 0.602, p < 0.0001), whereas other parameters did not, such as LAD, LAVI, and transmitral flow velocity to mitral annular velocity ratio. Furthermore, LAEF was decreased with advancing stage of CHADS2 and grade of SEC. The logistic regression analysis revealed that only LAEF was an independent predictor of cardioembolic stroke in TTE parameters (per 1 standard deviation decrease, odds ratio 2.176; 95% confidence interval 1.213 – 3.911; P < 0.01). LAEF remained to be an independent risk factor for cardioembolic stroke after age adjustment (per 1 standard deviation decrease, odds ratio 1.825; 95% confidence interval 1.003 – 3.319; P < 0.05).
Conclusion: LAEF may be a reliable non-invasive echocardiographic parameter to predict cardioembolic stroke in patients with chronic atrial fibrillation.
- © 2010 by American Heart Association, Inc.