We have already reported that the ratio of portal venous flow 30 min after oral intake of glucose 75 g to that before intake (PVFR30), measured using pulsed-Doppler ultrasonography (US), correlated significantly with other indicators of liver function and that it could be used to estimate hepatic function before surgery, including liver resection. In this study, to assess the disadvantages of pulsed-Doppler ultrasonography, PVFR30 was measured using two-dimensional (2D) phase-shift (PS) magnetic resonance imaging (MRI). PVFR30 was measured in 17 patients and 7 volunteers: 13 with liver cirrhosis (LC) and 11 without LC (non-LC). Portal venous flow could be measured in all patients without any disturbance of intestinal gas or patient fat, or the high degree of technical skill that Doppler US requires. PVFR30 was significantly lower in the LC group than in the non-LC group. In addition, it correlated significantly with other indicators of liver function, including the indocyanine green clearance test, prothrombin time, hepaplastin test, and cholinesterase activity. These results suggest that PVFR30 measured by 2D PS MRI can be used to estimate liver function, and that this MRI method can be performed more easily than pulsed-Doppler US.