Assessment of cancer cell differentiation in small hepatocellular carcinoma by computed tomography and magnetic resonance imaging

Aim: To study whether cancer cell differentiation in small hepatocellular carcinoma (HCC) can be assessed by computed tomography (CT) and magnetic resonance (MR) imaging. Methods: We retrospectively evaluated the relationship between cancer cell differentiation in 127 HCC 3 cm or less in diameter (113 patients) and CT and MR images. Images were reviewed in a consensus conference by three authors (SA, TY, and ME). Histopathological diagnosis of HCC was made from liver specimens obtained by sonographically guided biopsy. Results: The degree of histological differentiation of cancer cells was significantly different between HCC that were isodense with liver parenchyma in both artery-dominant and equilibrium phases in contrast-enhanced CT and tumors that were hyperdense in the artery-dominant phase and iso- or hypodense in the equilibrium phase (P = 0.0054), as well as tumors that were iso- or hypodense artery-dominant and hypodense equilibrium (P = 0.0002). Histological differentiation of lesions that were hyperintense in T1-weighted images and hypointense in T2-weighted images differed significantly from those with the opposite MR characteristics (P = 0.0122). In T1-weighted fat-suppression images and T2-weighted images, respectively, the degree of histological differentiation was significant between the hypointense/hyperintense and the hyperintense/hypointense patterns (P < 0.0001), as well as the hyperintense/isointense (P = 0.0296), the hyperintense/hyperintense (P = 0.0434), and the isointense/hyperintense (P = 0.0171). Using these differences an equation was developed that could determine with 76% accuracy whether the tumors were well or less-well differentiated. Conclusion: CT and MR imaging patterns were useful in predicting the degree of histological differentiation of cancer cells in HCC. (C) 2003 Blackwell Publishing Asia Pty Ltd.