Diagnosis of hepatocellular carcinoma – usefulness of magnetic resonance T2-weighted images, diffusion weighted images, and T1 post-contrast injection sequences

hyper-enhancement followed by wash-out in the portal or late venous phases (the so called vascular diagnosis criterion for HCC). However, there are many studies showing that small HCC lesions (with a diameter of less than 2 cm) present in less than 50% of the cases the classical model of contrast-enhancement, presenting either hyper-enhancement during the arterial phase not accompanied by wash-out in portal/late phase, either without arterial enhancement at all, being isoor even hypo-vascular to the hepatic parenchyma during the arterial phase (Bolondi et al 2005; Hammerstingl et al 2008; Huppertz et al 2004). From a histological point of view, early HCC represents a distinct subtype of HCC, in general with a smaller size, with few internally neo-formation arteries, aspect which explains its hypo-vascular nature (Krinsky et al 2001). Considering these, the strict application of the vascular criterion will lead to whether the increase of the biopsies number in cases of suspected HCC nodules, or to the sub-diagnosis/ delayed diagnosis of HCC nodules (Lorenz et al 2000). If we would try to improve the sensitivity of the technique by using hyper-enhancement during the arterial phase as the only Abstract. Objectives: To comparatively assess the sensitivity of different magnetic resonance imaging sequences for the diagnosis of the hepatocellular carcinoma. Material and method: A total number of 60 hepatic nodules in 49 patients, with a final diagnosis of hepatocellular carcinoma have been included in the present study. All included lesions had a final histological diagnosis of hepatocellular carcinoma or they were considered as being hepatocellular carcinomas after a consensual reading made by two radiologists with experience in abdominal imaging . All nodules have been characterized through the following properties: the signal on T2-weighted images, the signal on diffusion weighted sequence, and the apparent diffusion coefficient obtained from the diffusion map. The vascular behavior in the arterial, portal and late interstitial phase has been also analyzed. Results: Of all assessed lesions, 75% were hyperintense on T2-weighted sequences and 68.3% were hypointense on the apparent diffusion coefficient map. The vascular criterion in favor of hepatocellular carcinoma (arterial hyper-vascularization accompanied by washing-out in the portal or the late phase) was met in 77.6% of the lesions, with arterial hyper-vascularization seen in 86.7% of the cases. Most of the arterial hyper-vascularized lesions with no washing-out, presented hyper-signal on the T2-weighted sequences or hypo-signal on the apparent diffusion coefficient map. Conclusions: The percentage of hepatocellular carcinoma nodules that follow the vascular criterion is under 80%. The hyper-signal on the T2 sequence and the hypo-signal on the apparent diffusion coefficient map are criteria with a reasonably good sensitivity for the diagnostic of HCC, and can be used as auxiliary diagnosis criteria.