The hepatocellular entry of salicylamide sulfate conjugate, which binds to both red blood cells and albumin, was examined with the multiple-indicator dilution technique in the perfused rat liver, with medium containing both 20 red cells and 1 albumin (set A), red cells only (set B), albumin only (set C) and neither red cells nor albumin (set D). [14C]Salicylamide sulfate, 51Cr-labeled red cells (a vascular reference), 125I-labeled albumin, [3H]sucrose or [58Co] ethylenediaminetetraacetic acid (EDTA) (high and low molecular weight interstitial references, respectively) and 3H2O or D2O (a cellular reference) were injected as a bolus into the portal vein. Among all sets of outflow data, the earliest immediate vascular recoveries for the [14C] salicylamide sulfate were lower than those for the vascular reference. For sets A and B, the upslopes precessed those for the 58Co-EDTA curve, then crossed over the 58Co-EDTA curves, with lower magnitude peaks occurring at the same time as those for labeled albumin, whereas for sets C and D, in which red cells were absent, the upslopes of [14C]salicylamide sulfate lagged behind those for labeled albumin and [3H]sucrose, reaching lower magnitude peaks coincidental in time with those for labeled sucrose. The precession of the [14C]salicylamide sulfate curve over 58Co-EDTA or [3H]sucrose in the presence (sets A and B) but not in the absence (sets C and D) of red cells and the absence of precession with albumin alone (set C) suggest that a red cell effect on the upslope is more evident than that for albumin. For all experiments, the downslopes of the sulfate curve crossed over those of the labeled red cells, albumin, 58Co-EDTA and sucrose curves and then the water curve at around the peak. The downslopes of the [14C]salicylamide sulfate were similar to those for labeled water, suggesting rapid cellular influx and efflux of salicylamide sulfate. Quantitative evaluation with a barrier-limited space-variable transit-time model for rapidly equilibrating red cell and albumin binding accounted for the upslope effects on [14C]salicylamide sulfate behavior and demonstrated its relatively high liver cell permeability. Values for the unbound permeability surface area product (0.029 to 0.036 ml sec−1·gm−1) were not different, regardless of the presence or absence of red cells and albumin, and slightly exceeded values for hepatic blood flow (0.0186 ± 0.0016 ml sec−1·gm−1). Comparable influx (0.083 to 0.14 sec−1) and efflux (0.10 to 0.18 sec−1) coefficients were obtained, suggesting rapid cellular entry and efflux of salicylamide sulfate. The ratio of the derived rate constants, which corresponds to the concentration ratio across the membrane, was 0.40 to 0.55. The findings contrast sharply with those for acetaminophen sulfate, which enters liver cells poorly. (Hepatology 1994; 19:229–244).