Transesophageal echocardiography (TEE) is used to diagnose hypovolemia despite the lack of validation studies. The objective was to determine the effects of acute graded hypovolemia on TEE and conventional hemodynamic determinants of left ventricular (LV) preload in anesthetized patients with normal and abnormal LV function. Determinants of LV preload derived from TEE and hemodynamic monitoring were measured serially in 35 anesthetized cardiac surgical patients without valvular heart disease. Patients were stratified into two groups: those with normal LV function (group 1, n = 17) and those with LV wall motion abnormalities (group 2, n = 13). Patients in groups 1 and 2 were subjected to graded hypovolemia produced by collecting 6 aliquots of blood, each equal to 2.5% of their estimated blood volume (EBV). A third group of patients (group 3, n = 5), not subjected to graded hypovolemia, were studied to test for time-dependent changes. Group 2 had a significantly greater baseline (mean +/- SD) pulmonary artery occlusion pressure (17 +/- 6 vs. 11 +/- 6 mmHg), LV end-diastolic area (23 +/- 5 vs. 18 +/- 4 cm2), LV end-diastolic wall stress (23 +/- 10 vs. 14 +/- 6 x 10(3) dyne.cm-2), and smaller fractional area change (35 +/- 13 vs. 59 +/- 7%). In groups 1 and 2, the LV end-diastolic area, pulmonary artery occlusion pressure, and LV end-diastolic wall stress decreased linearly in response to blood loss in the range of 0-15% of the EBV. No significant changes in the measured parameters occurred in group 3. A significant decrease in the central venous pressure, pulmonary artery occlusion pressure, and LV end-diastolic area was detected in response to a 2.5% EBV deficit (approximately 1.75 ml.kg-1) in groups 1 and 2. The mean change in LV end-diastolic area (0.3 cm2/1.0% EBV deficit) in response to equivalent EBV deficits was the same in groups 1 and 2. In contrast, the mean change in cardiac output and LV end-diastolic wall stress was less in group 2 despite a greater decrease in pulmonary artery occlusion pressure. Compared to group 1, a greater EBV deficit (7.5% to 12.5% vs. 2.5% to 5%) was required in group 2 to cause a significant decrease in the cardiac output, stroke volume, mixed venous oxygen saturation, and LV end-diastolic wall stress. TEE and hemodynamic determinants of LV preload detected changes in LV function caused by acute blood loss. Acute blood loss caused directional changes in LV end-diastolic area, pulmonary artery occlusion pressure, and LV end-diastolic wall stress even in patients with LV wall motion abnormalities. Changes in LV end-diastolic wall stress, derived from both TEE and hemodynamic measurements corresponded to changes in cardiac output, stroke volume, and mixed venous oxygen saturation that occurred during acute blood loss.