In order to assess the role of the pulmonary venous flow (PVF) velocity pattern in the evaluation of patients with congestive heart failure (CHF), we studied 41 CHF patients by means of transthoracic echocardiography (TTE) and multiplane transesophageal echocardiography (TEE). The etiology of CHF was idiopathic or ischemic dilated cardiomyopathy in 19 patients and hypertensive heart disease in 22. Sixteen subjects without cardiovascular disease were selected as normal controls. PVF peak systolic and peak early diastolic (D) velocities were recorded by TEE and TTE and the systolic fraction (SF) was measured (i.e., the systolic velocity-time integral-VTI-expressed as a fraction of the sum of systolic and early diastolic (VTI). TEE tracings were obtained in all patients and had more laminar-appearing spectral signals, thus were used for analysis. By TEE the mitral flow velocity patterns were also evaluated: peak early diastolic velocity (E), peak velocity at atrial contraction, E velocity normalized for VTI (E/VTI), deceleration time (DT), and left ventricular isovolumic relaxation time (LVIRT). The left ventricular ejection fraction (LVEF) was calculated by two-dimensional echocardiographic images using the modified Simpson method. The SF was lower in CHF patients as compared with normal controls (p < 0.0001). The E/VTI ratio was higher, and DT and LVIRT were shorter (p < 0.0001) in CHF patients. A significant correlation was observed between SF and LVEF in CHF patients (r = 0.76, p < 0.001). Two different PVF velocity patterns (type A:SF << 50%, D > 50 cm/s; type B:SF approximately 50%, D > 50 cm/s) were recognized in patients with a low LVEF (type A) and a nearly normal or normal LVEF (type B). Patients with LVEF < 40% showed mean SF values significantly lower than patients with LVEF > 40% (33.26 +/- 10.84 vs. 51.00 +/- 4.00%, p < 0.0001). Mean DT and LVIRT values were not significantly different in patients with LVEF < 40% and > 40%. Thus in CHF patients TEE PVF velocity patterns help in distinguishing patients with systolic dysfunction (low LVEF and SF) from patients with predominant diastolic impairment (normal or nearly normal LVEF, high D velocities).