Research on hepatocellular carcinoma of CT perfusion and reproducibility

Objective To prospectively assess computed tomographic(CT) perfusion for evaluation of tumor vascularity of advanced hepatocellular carcinoma(HCC) and to correlate CT perfusion parameters with tumor grade and serum markers. Methods Total of 30 patients with unresectable HCC and hepatic metastatic(25 cases of HCC, 5 cases of hepatic metastases) were accept dynamic by CT perfusion scan for the fi rst time after intravenous contrast agent. CT perfusion parameters(tumor tissue, hepatic tissue blood fl ow, blood volume, mean transit time, permeability surface area product) were collected. Repeat examination was performed in four patients within 30 hours to test reproducibility of CT perfusion. CT perfusion parameters were compared among tumors of different grades, with presence or absence of portal vein invasion, with presence or absence of cirrhosis, and of various extrahepatic metastases. Parameters were correlated with HCC serum markers. One-way analysis of variance was used to calculate variations in CT perfusion parameters. Results Good correlation(r = 0.9, P 0.01) was observed between repeat examination results and fi rst CT examination results. There was a significant difference(P 0.05) in CT perfusion parameters between primary HCC and background liver parenchyma. Well-differentiated HCC showed significantly higher perfusion values(P 0.05) than other grades. There was no signifi cant difference in tumor perfusion between presence or absence of portal vein invasion or cirrhosis. CT perfusion value of lymph node metastasis demonstrated was lower than that of other extrahepatic metastasis. There was no signifi cant correlation between CT perfusion parameters and serum markers(P 0.05). Conclusions CT perfusion is a feasible and, from the limited data, reproducible technique for quantifying tumor vascularity and angiogenesis in advanced HCC.