OBJECTIVE: The purpose of this study was to investigate the relative enhancement characteristics of the hepatic parenchyma and portal vein during gadoxetate disodium-enhanced abdominal MRI and to assess whether contrast between the portal vein and the hepatic parenchyma can be improved with higher doses of gadoxetate disodium. MATERIALS AND METHODS: A total of 102 patients (61 women, 41 men) underwent gadoxetate disodium-enhanced abdominal MRI. They received a weight-independent dose of 10 mL of gadoxetate disodium, corresponding to a dose spectrum of 0.02-0.06 mmol/kg body weight. The patients were assigned to one of three dose groups: recommended dose (0.02-0.03 mmol/kg), intermediate dose (0.03-0.045 mmol/kg), or high dose (0.045-0.06 mmol/kg). The signal-to-noise ratios for the portal vein, liver, and the portal vein-to-liver contrast-to-noise ratio were calculated for three consecutive arterial phases, one portal venous phase, and four delayed imaging phases. RESULTS: The delayed phase images of the liver showed statistically significant dose dependency and greater enhancement in the intermediate- and high-dose groups (p < 0.01). Analogously, the portal vein also exhibited greater enhancement in the two higher-dose groups, but the difference was not statistically significant (p > 0.05). Regarding portal vein-to-liver contrast, all three groups had a dose-independent fast parallel increase from baseline toward maximum contrast followed by a steady decline in contrast with no statistically significant differences between dose groups (p > 0.05). CONCLUSION: Portal vein-to-liver contrast during gadoxetate disodium-enhanced hepatic MRI cannot be improved within a dose spectrum of 0.025-0.06 mmol/kg body weight.