Portal and hepatic indicator dilution curves (IDC) were obtained after injection of a mixture of 51Cr-labeled red blood cells ([51Cr]RBC) and 125I-albumin microaggregates (125I-AMA) into the cranial mesenteric artery in dogs. The extraction (E) of 125I-AMA from portal blood was measured during one passage through the hepatic reticuloendothelial system. Using [51Cr]RBC as a vascular reference substance, E-125I-AMA was calculated by comparing simultaneous [51Cr]RBC and 125I-AMA portal and hepatic IDC, and was expressed as percentage of 125I-AMA flowing through the portal vein. In 44 experiments (15 dogs), the colloid was almost completely extracted (E-125I-AMA = 92.3 +/- 1.0% (mean +/- SE)). This approach was applied in 15 patients with severe portal hypertension undergoing combined umbilicoportal, hepatic vein, and superior mesenteric artery catheterization. Eleven patients had alcoholic cirrhosis (AC) and 4 patients had idiopathic noncirrhotic portal hypertension (IPH). Using [51Cr]RBC-IDC, the portal fraction of hepatic blood flow varied between 34.1 and 100% (mean 62.6%) in AC patients and between 56.5 and 91.2% (mean 74.2%) in IPH patients. E-125I-AMA varied from 5.2 to 100% (mean 45.1%) in AC patients, although normal values were obtained in IPH patients (mean 93.2%). In all patients the extraction of Indocyanine green (E-ICG) was calculated using a continuous infusion for the estimation of hepatic blood flow. E-ICG was decreased in AC patients (mean 22.1%), although normal values were obtained in IPH patients (mean 49.5%). A highly significant correlation was found between E-125I-AMA and E-IGC (r = 0.977, P less than 0.001). Also, a significant correlation was found in all patients between E-125I-AMA and the relative clearance of ICG (r = 0.906, P less than 0.001). The correlations between the extraction or clearance of substances removed by two different cell population suggest that their decreases are mainly due to changes in liver microcirculation. In cirrhotics, the decreased E-125I-AMA can be related to part of portal blood bypassing Kupffer cells (intrahepatic portohepatic shunts) and/or to sinusoidal changes responsible for ineffective phagocytosis. Thus, E-125I-AMA can be used as an estimation of the functional portal blood supply to the liver in cirrhotics. Using portal and hepatic IDC after injection of [51Cr]RBC and 125I-AMA into the superior mesenteric artery, the portal fraction of hepatic blood flow and the functional portal blood supply can be estimated simultaneously in patients with portal hypertension before portacaval shunts.