To investigate whether acute-phase beta-blocker therapy has a harmful effect on left atrial appendage (LAA) function in patients with chronic nonvalvular atrial fibrillation by transesophageal echocardiography (TEE), we evaluated 21 patients with normal left ventricular systolic function and a poorly controlled ventricular rate, despite the use of digoxin. Baseline parameters that were obtained included heart rate, blood pressure, LAA emptying velocities, and left atrial spontaneous echo contrast intensity. Then, each patient was given a bolus dose of 5 mg metoprolol. Ten minutes later, a second set of assessments was performed. After the first TEE studies, each patient began treatment with metoprolol (50 ms orally twice daily for 1 week). A second TEE study was performed after 1 week of continuous oral metoprolol therapy at maintenance dose, and values were again determined. The average resting apical heart rate was 91 +/- 7 bpm. As expected, beta-blocker therapy showed a marked decrease in heart rate at 10 minutes (79 +/- 6 bpm P < .001) and at 1 week (71 +/- 4 bpm, P < .001). Beta-blocker therapy caused a significant reduction in systolic and diastolic blood pressures (144 +/- 16 / 93 +/- 6 mm Hg at baseline, 137 +/- 16 / 87 +/- 9 mm Hg at 10 minutes, and 135 +/- 12 / 86 +/- 8 mm Hg at 1 week, P < .001). With the beta-blocker therapy, the baseline transesophageal Doppler parameter of LAA emptying velocities (at baseline 24 +/- 7 cm/s) fell significantly at 10 minutes (19 +/- 7 cm/s, P < .001) and at 1 week (17 +/- 6 cm/s, P < .001) after initiation of beta-blocker therapy. After a bolus of metoprolol, spontaneous echo contrast intensity did not change in any patients, but 1 week later, it increased in 1 patient. In 2 patients who had not been found to have an LAA thrombus at baseline TEE study, the second TEE examination demonstrated new thrombi in the LAA. In conclusion, our endings suggest that in patients with chronic nonvalvular atrial fibrillation who have normal left ventricular systolic function and a poorly controlled ventricular rate despite the use of digoxin, acute-phase beta blockade may have a harmful effect on LAA function.