Comparison of test-injection method and fixed-time method for depiction of hepatocellular carcinoma using dynamic steady-state free precession magnetic resonance imaging

Objective: The purpose of this study was to clarify the usefulness of the test-injection method as compared with the fixed-time method in dynamic magnetic resonance (MR) imaging of hepatocellular carcinoma (HCC). Methods: Ninety-seven patients with a total of 118 hepatocellular carcinomas underwent 3-dimensional fast imaging with steady-state free precession (3D-FISP) for dynamic study of the liver as well as catheter-assisted computed tomography hepatic angiography (CTHA) for preoperative evaluation. In 42 cases, the fixed-time method (30-second scan time delay in the hepatic arterial phase [HAP]) was performed (group 1), and in 55 cases, the test-injection method was performed (group 2). The following parameters were evaluated: 1) the adequacy of the HAP, 2) tumor vascularity using CTHA findings as a gold standard, and 3) the contrast-to-noise ratio (CNR) of the HCC during the HAP of dynamic MR imaging. Results: In group 1, 79% (33 of 42) of the cases were obtained at the optimal HAP; the percentage in group 2 was 98% (54 of 55) of the cases. This difference was statistically significant (P < 0.05). The vascularity of 82% of the tumors in group 1 and 89% of those in group 2 was diagnosed correctly. Regarding hypervascular tumors, correct evaluation of tumor vascularity was made in 87% of group 1 cases and 95% of group 2 cases. No significant difference was present between the 2 groups (total: P = 0.43, hypervascular HCC: P = 0.29). 3) The CNR calculated for all HCCs in group 2 (mean ± SD: 8.66 ± 11.0) was significantly higher than that for HCCs in group 1 (4.29 ± 9.44; P < 0.05). As for the hypervascular tumors, the CNR calculated for group 2 (mean ± SD: 9.89 ± 10.6) was also significantly higher than that for group 1 (5.52 ± 9.81; P < 0.05). Conclusion: The 3D-FISP dynamic MR imaging using the test-injection method resulted in better demonstration of HCC than the 3D-FISP using the fixed-time method.