This study showed that the presence of a halo and an intratumoral echo pattern on ultrasonography were useful predictors for local recurrence after percutaneous ethanol injection therapy for small hepatocellular carcinoma, and that a high serum alpha-fetoprotein level was associated with a higher frequency of new nodular recurrences.

Thrombocytopenia in liver diseases is considered to be due to splenic platelet pooling and accelerated destruction. Since thrombopoietin (TPO), a regulator of thrombopoiesis, is produced mainly in the liver, decreased production of TPO may account for thrombocytopenia in liver diseases. To address this issue, we measured serum TPO, using a sensitive sandwich ELISA, in 108 patients with chronic viral hepatitis, which included chronic hepatitis (CH) and liver cirrhosis (LC), and hepatocellular carcinoma (HCC), and in 29 normal controls. TPO mRNA in 78 liver samples was examined by reverse transcription (RT)-PCR. Platelet counts in CH, LC, HCC and controls were 176 +/- 15 x 10(9)/l, 81 +/- 8 x 10(9)/l, 99 +/- 7 x 10(9)/l and 234 +/- 9 x 10(9)/l respectively. Serum TPO levels in CH, LC and HCC were 2.79 +/- 0.4 fmol/ml, 1.49 +/- 0.2 fmol/ml and 1.97 +/- 0.2 fmol/ml, and were higher than those of controls. Serum TPO levels were positively correlated with prothrombin time and serum albumin (P < 0.05, in each case), and negatively correlated with Indocyanine Green test and Pugh score (P < 0.01 and P < 0.05 respectively). However, RT-PCR and immunohistochemistry showed that expression of TPO mRNA and protein were similar in the different liver diseases, suggesting that serum TPO is a reflection of the total mass of functional liver. Platelet counts were negatively correlated with spleen index, but not with serum TPO. These results suggest that thrombocytopenia in liver disease is not directly associated with serum TPO but is associated with hypersplenism.

Carcinogenesis is a multistep process. Most hepatocellular carcinoma (HCC) is preceded by liver cirrhosis, but the genetic changes involved in cirrhosis are not known well. The present study was conducted to evaluate aberration of the retinoblastoma (RB) gene in HCC and adjacent non-tumorous liver using 22 patients with chronic liver damage accompanying HCC. The specimens obtained by microdissection from paraffin-embedded tissues were analyzed using an assay based on the polymerase chain reaction for highly polymorphic nucleotide sequences of microsatellites in the RB gene. Out of 22 cases, 15 showed constitutional heterozygosity for the microsatellite markers. In 11 (73.3%) of these 15 informative cases, the primary HCC foci showed loss of heterozygosity (LOH). In 8 of these 11 doubly informative (informative and LOH-positive in primary HCC) cases, LOH was found in 20 (64.5%) of 31 microdissected non-tumorous foci. All of the non-tumorous foci showing RB loss were cirrhotic lesions but there were no foci of chronic hepatitis. The remaining 4 cases without LOH in HCC foci showed no LOH in non-tumorous lesions. In our study, LOH of the RB gene was frequently observed in liver cirrhosis surrounding tumor.

No abstract

No abstract