The relationship between effective renal plasma flow (ERPF) and cardiac output was examined in 46 patients (22 with congestive heart failure and 24 following cardiac surgical procedures) by simultaneously measuring the global ERPF by the single-injection method and cardiac output by the thermodilution method. Of the patients in the heart-failure group, 21 also had pulmonary artery end diastolic pressure (PAEDP) recorded at the same time. ERPF and cardiac output were found to be related by the regression equation: cardiac output=2.08+0.0065 ERPF (r, 0.80), with a SE of estimate of 0.81 l/min. ERPF and PAEDP were related by the regression equation: PAEDP= 45.02-0.0675 ERPF (r, 0.86), with a SE of estimate of 5.5 mm Hg. ERPF may be a useful noninvasive method of estimating cardiac output if it is known that no intrinsic kidney disease is present, and if the error of 0.81 l/min (1 SE of estimate) is within the range of clinical usefulness. The error is principally attributable to the determination of cardiac output by the thermodilution method.