Assessment of liver hemodynamics can be obtained by analysis of first pass flow studies through the liver and spleen using 99mTc compounds which are not actually trapped by these organs. This study examines new and existing methods for determining the relevant contribution made by the hepatic artery and portal vein to total liver blood flow, from these first pass studies. Eighty-two studies were performed in 56 patients with both normal and abnormal liver function. Using region of interest analysis, time-activity curves were obtained for the lungs, liver, spleen, and left kidney. These curves were analyzed by four different methods. Two of these methods are based upon measurement of the slopes of the uptake and washout curves from the liver and spleen and the other two methods employ deconvolution analysis to permit area measurement under the deconvolved curves as an indicator of blood flow. All four methods showed a small intraobserver variation after reanalysis. In 11 patients who underwent repeat studies, the correlation between the deconvolution based methods (r = 0.79-0.89) was significantly better than that for the slope based methods (r = 0.55-0.58). The deconvolution based methods provided the most significant separation between normals and patients with various liver disorders and would appear to be the most suitable techniques for monitoring the effects of various drugs and surgical procedures on the relative arterial/portal contribution to hepatic blood flow.