OBJECTIVE The aim of our study was to compare measurements of pulmonary venous flow velocity obtained either by transesophageal Doppler echocardiography or by intravascular catheter Doppler velocimetry. Furthermore, the relation among pulmonary venous flow velocity, left atrial compliance and left atrial pressure was evaluated. BACKGROUND Data about the relation between left atrial pressure and pulmonary venous flow velocity are controversial. METHODS A total of 32 patients undergoing elective open heart surgery for coronary artery bypass grafting were included prospectively in the study. Pulmonary venous flow velocity (Doppler catheter) and left atrial pressure (microtip pressure transducer) were recorded simultaneously with recordings of pulmonary venous flow velocity obtained by transesophageal Doppler echocardiography. RESULTS Agreement between Doppler catheter and Doppler echocardiographic measurements of pulmonary venous flow velocity (n = 18 patients) was analyzed using the Bland-Altmann technique. The 95% limits of agreement were -0.16 to +0.11 m/s for systolic peak velocity, -0.14 to +0.09 m/s for diastolic peak velocity and -0.12 to +0.10 m/s for atrial peak velocity. The closest agreement between both methods was found for the ratio of systolic to diastolic peak velocity, the ratio of systolic to diastolic flow duration and the time from Q deflection on the electrocardiogram to maximal flow velocity. Mean left atrial pressure was strongly correlated with the ratio of systolic to diastolic peak velocity (r = -0.829), systolic velocity-time integral (r = -0.653), time to maximal flow velocity (r = 0.844) and the ratio of systolic to diastolic flow duration (r = -0.556). The ratio of systolic to diastolic peak velocity and the time to maximal flow velocity were identified as strong independent predictors of mean left atrial pressure. Left atrial compliance was not found to be an independent predictor of mean left atrial pressure. CONCLUSIONS Flow velocity in the left upper pulmonary vein can be reliably recorded by transesophageal pulsed wave Doppler echocardiography. Our data reveal further evidence that mean left atrial pressure can be estimated by the pattern of pulmonary venous flow velocity.