Background: Outcome of patients with hepatocellular carcinoma is influenced by their histological grade. Invasive biopsy of the lesions is the gold standard in this regard.
Aims: We therefore analysed the diagnostic accuracy of contrast enhanced ultrasound for non-invasive grading of hypervascular hepatocellular carcinoma in liver cirrhosis.
Methods: According to the tumour perfusion kinetics on contrast enhanced ultrasound two grading groups were prospectively defined: well-differentiated hepatocellular carcinoma (US-G1) and higher grade hepatocellular carcinoma (US-G2/G3). Immediately after contrast enhanced ultrasound-grading, biopsies of hepatocellular carcinoma-lesions (n=95, 1.2-12.5 cm) were obtained and analysed for tumour grading (G). Descriptive statistics, sensitivity, specificity positive and negative predictive values, diagnostic likelihood ratios and interoperator reproducibility were calculated (κ).
Results: Histologically 77 (81.1%) patients had G2-G3 and 18 (18.9%) had G1 tumours. Higher grade hepatocellular carcinoma showed more often a washout in the portal or late phase (p<0.0001). The sensitivity, specificity, positive predictive values and negative predictive values of contrast enhanced ultrasound for grading of hepatocellular carcinoma for all patients were 94% (CI: 72-99%), 95% (CI: 88-99%), 81% and 99% and for patients with tumours<5 cm 100%(95% CI: 79-100), 96% (95% CI: 80-99), 92% and 100%. Positive and negative diagnostic likelihood ratios' were 18 and 26 and 0.06 and 0, respectively. κ=0.941 (p<0.001).
Conclusions: Contrast enhanced ultrasound has a high diagnostic value and reproducibility for non-invasive grading of hypervascular hepatocellular carcinoma >1cm in patients with liver cirrhosis.
Copyright © 2011 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved.