To examine changes in left atrial appendage flow velocity patterns in relation to left atrial pressures during sinus rhythm, transesophageal echocardiography and cardiac catheterization were performed in 31 patients with myocardial diseases in sinus rhythm and 20 control subjects without cardiovascular disease. The 31 patients were divided into two groups according to mean pulmonary capillary wedge pressure: the group with high wedge pressure (19.9 +/- 5.8 mmHg) and the group with low wedge pressure (8.6 +/- 2.9 mmHg). The left atrial appendage peak early emptying velocity was decreased significantly in the groups with both high and low wedge pressure compared with the control group. The left atrial appendage peak late emptying velocity was significantly greater in the group with low wedge pressure compared with the control group, whereas it was decreased significantly in the group, with high wedge pressure compared with the control group. The left atrial appendage peak late emptying velocity had a significant negative correlation with wedge pressure. The maximum left atrial appendage area at end systole in the group with high wedge pressure was significantly greater than that in both the group with low wedge pressure and the control group. There was a significant positive correlation between the maximum left atrial appendage area and the wedge pressure, as well as a significant negative correlation between the left atrial appendage ejection fraction during atrial contraction and the wedge pressure. In the group with high wedge pressure, one patient had evidence of left atrial appendage thrombi and two had spontaneous echo contrast. These results suggest that even in patients in sinus rhythm, a marked elevation in the left atrial pressure is likely to reduce the left atrial appendage peak early and late emptying velocities. These changes may be accompanied by an increased incidence of thrombus formation in the left atrial appendage compared with individuals with normal or only slightly elevated left atrial pressures.