To elucidate how individual determinants might lower the rate of glomerular ultrafiltration (GFR) in some patients following cardiac surgery, we performed hemodynamic measurements and clearance of inulin (as a measure of GFR), PAH (as a measure of effective renal plasma flow [ERPF]), and dextran-40. Two groups of 17 patients each were distinguished by the presence or absence of prerenal azotemia. Glomerular hypofiltration (GFR = 21 +/- 2 vs. 76 +/- 6 ml/min/1.73 m2, P less than 0.001) in the former was accompanied by depressed left ventricular function, arterial pressure, and ERPF (152 +/- 26 vs. 317 +/- 32 ml/min/1.73 m2, P less than 0.001). To determine if factors beside ERPF play a role in lowering GFR, we calculated the efferent oncotic pressure (pie). Failure of GFR to change over a 24-hour period despite increases in ERPF suggested that both patient groups were at filtration pressure disequilibrium (FPD). This condition permits calculation of a unique glomerular ultrafiltration coefficient (Kf). Over a range of pressures for transcapillary hydraulic pressure (deltaP), such that 3 less than or equal to (deltaP - pie) less than or equal to 10 mm Hg (to simulate FPD), Kf was less than 0.08 ml . sec-1 . mm Hg-1 . 1.73 m-2 in azotemic, but exceeded this value in nonazotemic patients. Although a selective reduction of Kf is predicted to lower the fractional clearance of dextrans, these were significant elevated in azotemic relative to nonazotemic patients (molecular radii 30 - 40 A). A theoretical analysis of the latter data suggests that over the foregoing range of FPD, a 15 to 30% decline in deltaP combined with a 30 to 0% reduction in Kf from values in nonazotemic patients best explains the experimental findings in azotemic patients.