The plasma fractional disappearance rate of indocyanine green (kICG, min-1) and the absolute hepatic ICG removal rate (VICG, mg per kg per min) were determined in 26 patients with Gilbert's syndrome (GS) and 19 normal control volunteers following intravenous administration of doses of 0.5, 2.0, 3.5, and 5.0 mg per kg of dye. The diagnosis of GS was based on studies of radiobilirubin kinetics in all cases and liver biopsy in 22 cases. The patients were further classified into 3 subgroups, based on patterns of sulfobromophthalein (BSP) kinetics, as follows: GS I (15 patients) had normal BSP metabolism; GS II (5 patients) had a defect in BSP metabolism beyond the stage of initial hepatic uptake; and GS III (6 patients) had a defect in the initial hepatic uptake of BSP (Gastroenterology 63:472-481, 1972). Both kICG and VICG were significantly reduced, compared to normal controls, in the GS III group with defective BSP uptake, but did not differ significantly from normal in the GS I and GS II groups. Michaelis-Menten analysis of the data indicated that VMAX for ICG uptake in the GS III group (1.2 +/- 0.6 mg per min per kg) was significantly reduced compared to the previously established normal value of 3.6 +/- 0.6 mg per min per kg; (P less than 0.01). For the total population of 26 patients with Gilbets' syndrome, there was a highly significant correlation (r= 0.77, P less than 0.01) between kICG and lambda21BSP, the fractional hepatic uptake rate for BSP. These studies confirm previous work indicating that patients with Gilbert's syndrome constitute a heterogeneous population with regard to defects in hepatic organic anion transport, some of the defects not being attributable to glucuronyl transferese deficiency. Future studies of Gilbert's syndrome must take into account the existence of these subgroups, since they may have different underlying pathogenetic mechanisms.