PURPOSE: To evaluate whether hepatic magnetic resonance angiography (MRA) performed at 3 T and acquired with a parallel imaging acceleration factor of 2 provides satisfactory image quality and adequate contrast-to-noise, compared with an acquisition without parallel imaging. MATERIALS AND METHODS: Ten volunteers underwent both a standard (A) and an experimental (B) MRA protocol. Both protocols used a gadoteridol dose of 20 mL injected at 2 mL/s in a 3-T magnetic resonance (MR) system. Protocol B was identical to protocol A, except protocol B was performed with a parallel imaging acceleration factor of 2. Contrast-to-noise ratios (CNRs) were calculated in the suprarenal abdominal aorta, the common hepatic artery, and the right and left hepatic arteries. The same 4 vessels were rated for quality of arterial visualization using a 5-point scale (1 = poor to 5 = excellent). The paired t test and Wilcoxon rank sum test were used for statistical analysis. RESULTS: The CNRs and qualitative scores were higher in all 4 vessels using protocol B. Mean CNRs in the suprarenal abdominal aorta, common hepatic artery, and the right and left hepatic arteries were 31, 33, 28, and 22 for protocol A and 36, 40, 36, and 25 for protocol B, respectively (P < 0.05 except the left hepatic arteries [P = 0.35]). Mean qualitative scores of the same vessels were 4.3, 3.7, 3.1, and 2.9 using protocol A and 4.4, 3.8, 3.5, and 3.2 for protocol B, respectively (P > 0.44). CONCLUSIONS: Parallel imaging performed in a 3-T MR system improves hepatic MRA both quantitatively and qualitatively.