Gadoxetate disodium-enhanced hepatobiliary phase MRI of hepatocellular carcinoma: correlation with histological characteristics.

OBJECTIVE: The purpose of this study was to assess whether gadoxetate disodium-enhanced hepatobiliary phase MRI could predict the histologic factors of hepatocellular carcinomas (HCCs). MATERIALS AND METHODS: Fifty-three HCCs histopathologically proved by surgery in 51 patients were evaluated retrospectively. All patients underwent gadoxetate disodium-enhanced MRI before surgical resection. The differences in contrast enhancement ratio of the lesions and differences in contrast-to-noise ratio (CNR) among the histologic grades of HCC were compared by using the Kruskal-Wallis test. The Spearman method was used to determine the correlations among contrast enhancement ratio, CNR, cell density ratio, and positivity for anti-hepatocyte antibody, keratin 7, and keratin 19. RESULTS: Of 53 HCCs, 50 showed low signal intensity on hepatobiliary phase images, whereas three HCCs were hyperintense on hepatobiliary phase images compared with surrounding hepatic parenchyma. Although well-differentiated HCCs tended to show higher contrast enhancement, there was no statistical significance between contrast enhancement ratio of the tumors and histologic grade (p = 0.414). No significant difference was observed between CNR and histologic grade (p = 0.965). The contrast enhancement ratios of the tumors were significantly lower in the keratin 19-positive group than in the keratin 19-negative group (p = 0.015). There was no significant correlation among contrast enhancement ratio, anti-hepatocyte antibody positivity, cell density ratio, and keratin 7 positivity (p > 0.05). CONCLUSION: The contrast enhancement ratio and CNR of HCCs were not correlated with histologic grades. The contrast enhancement ratio was significantly lower in keratin 19-positive HCCs.ope