Three-dimensional analysis of the left atrial appendage for detecting paroxysmal atrial fibrillation in acute ischemic stroke.

BACKGROUND

Atrial fibrillation impairs left atrial appendage function and the thrombus formation in the left atrial appendage is a major cause of cardioembolic stroke.

AIMS: To evaluate the association between the volume of the left atrial appendage measured by real-time three-dimensional transesophageal echocardiography and presence of paroxysmal atrial fibrillation in patients with cerebral infarction or transient ischemic attack.

METHODS

Real-time three-dimensional transesophageal echocardiography was performed to measure left atrial appendage end-diastolic and end-systolic volumes to calculate left atrial appendage ejection fraction. Patients with normal sinus rhythm at the time of real-time three-dimensional transesophageal echocardiography were divided into groups with and without paroxysmal atrial fibrillation. Volumetric data were corrected with the body surface area.

RESULTS

Of 146 patients registered, 102 (29 women, 72·2 ± 10·7 years) were normal sinus rhythm at the examination. In 23 patients with paroxysmal atrial fibrillation, left atrial appendage end-diastolic volume (4·78 ± 3·00 ml/m(2) vs. 3·14 ± 2·04 ml/m(2), P = 0·003) and end-systolic volume (3·10 ± 2·47 ml/m(2) vs. 1·39 ± 1·56 ml/m(2), P < 0·001) were larger and left atrial appendage ejection fraction (37·3 ± 19·1% vs. 57·1 ± 17·5%, P < 0·001) was lower than in the other 79 patients without paroxysmal atrial fibrillation. The optimal cutoff for left atrial appendage peak flow velocity to predict paroxysmal atrial fibrillation was 39·0 cm/s (sensitivity, 54·6%; specificity, 89·7%; c-statistic, 0·762). The cutoffs for left atrial appendage end-diastolic volume, end-systolic volume, and ejection fraction were 4·52 ml/m(2) (sensitivity, 47·8%; specificity, 82·3%; c-statistic, 0·694), 1·26 ml/m(2) (sensitivity, 91·3%; specificity, 60·3%; c-statistic, 0·806), and 47·9% (sensitivity, 78·3%; specificity, 74·7%; c-statistic, 0·774), respectively. In multivariate analysis, all these parameters were independently associated with paroxysmal atrial fibrillation after adjusting for sex, age, diabetes mellitus, and previous stroke.

CONCLUSIONS

Left atrial appendage volumetric analysis by real-time three-dimensional transesophageal echocardiography is a promising method for detecting paroxysmal atrial fibrillation in acute cerebral infarction or transient ischemic attack.