We evaluated the effects of positive end-expiratory pressure (PEEP) on left ventricular function in 15 patients with acute respiratory insufficiency secondary to pulmonary edema with invasive (pressure; flow) measurements and radionuclide angiography (RA). Using RNA allowed a definition of the left ventricular ejection fraction (LVEF), and then calculation of the left ventricular end-diastolic volume (LVEDV), both before and after PEEP. With a mean PEEP of 14.2 +/- 1.8 cm H2O (mean +/- SD) (range, 10 to 15), a fall in the cardiac index (4.34 +/- 1.5 to 3.84 +/- 1.4 L/min/M2; p less than 0.001) was accompanied by a significant decrease in the stroke volume index (42 +/- 13 to 39 +/- 12 ml/beat M2; p less than 0.01) and pulse rate (103.4 +/- 14.3 to 98 +/- 13.5 beats/min; p less than 0.01). The decrease in the stroke volume index was primarily due to a significant decrease in left ventricular preload (LVEDV) from 85.9 +/- 19 to 71.4 +/- 21.4 ml/m2 (p less than 0.01). Simultaneously, the mean LVEF increased from 0.47 +/- 0.10 to 0.53 +/- 0.08 (p less than 0.05), despite a significant increase in the systemic vascular resistance (1,619 +/- 575 to 1,864 +/- 617 dynes . s. cm-5/M2; p less than 0.01). We concluded that the use of PEEP in patients with acute pulmonary edema, to the degree used in this study, may depress cardiac output by simply decreasing left ventricular preload. We were unable to produce any evidence that would support a change in the contractile state of the left ventricle as a cause of depressed forward flow with the use of PEEP.