BACKGROUND: This study investigated the usefulness of postvascular images of contrast-enhanced ultrasonography (CE-US) in the gross classification of hepatocellular carcinoma (HCC) in comparison with contrast-enhanced CT (CE-CT) findings. METHODS: This is a prospective study with consecutive HCC patients who had both CE-US and CE-CT prior to surgical resection. Fifty-one patients (32 men, 19 women; mean age, 68.9 years) with 61 HCCs were enrolled. The maximal diameters of all tumors ranged from 1.0 to 5.0 cm (mean ± SD, 2.5 cm ± 1.1). Weighted kappa statistics were used to assess the agreement of the sonographic or CT findings versus the results of macroscopic configurations. RESULTS: Thirty-nine tumors were macroscopically diagnosed as simple nodule type; 19 tumors were macroscopically diagnosed as simple nodular type with extranodular growth, and 3 were macroscopically diagnosed as confluent multinodular type from the resected specimen. The diagnostic accuracy was 86.9% (53/61) for CE-US and 65.6% (40/61) for CE-CT. The differences in accuracy between CE-US and CE-CT were statistically significant (McNemar; p = 0.007). Agreement analysis between gross classification using CE-US and final macroscopic results gave a kappa value of 0.74 (95% CI: 0.65–0.82), which was considered a good agreement. On the other hand, kappa coefficient value was 0.38 (95% CI: 0.28–0.48) between gross classification using CE-CT and final macroscopic results. CONCLUSIONS: CE-US is a more reliable tool than CE-CT to evaluate the gross type of HCC than CE-CT. Accurate gross classification using imaging is considered to be essential for the determination of the correct treatment strategy and the estimates of the patients' prognosis.
BACKGROUND: This study investigated the usefulness of postvascular images of contrast-enhanced ultrasonography (CE-US) in the gross classification of hepatocellular carcinoma (HCC) in comparison with contrast-enhanced CT (CE-CT) findings. METHODS: This is a prospective study with consecutive HCC patients who had both CE-US and CE-CT prior to surgical resection. Fifty-one patients (32 men, 19 women; mean age, 68.9 years) with 61 HCCs were enrolled. The maximal diameters of all tumors ranged from 1.0 to 5.0 cm (mean ± SD, 2.5 cm ± 1.1). Weighted kappa statistics were used to assess the agreement of the sonographic or CT findings versus the results of macroscopic configurations. RESULTS: Thirty-nine tumors were macroscopically diagnosed as simple nodule type; 19 tumors were macroscopically diagnosed as simple nodular type with extranodular growth, and 3 were macroscopically diagnosed as confluent multinodular type from the resected specimen. The diagnostic accuracy was 86.9% (53/61) for CE-US and 65.6% (40/61) for CE-CT. The differences in accuracy between CE-US and CE-CT were statistically significant (McNemar; p = 0.007). Agreement analysis between gross classification using CE-US and final macroscopic results gave a kappa value of 0.74 (95% CI: 0.65–0.82), which was considered a good agreement. On the other hand, kappa coefficient value was 0.38 (95% CI: 0.28–0.48) between gross classification using CE-CT and final macroscopic results. CONCLUSIONS: CE-US is a more reliable tool than CE-CT to evaluate the gross type of HCC than CE-CT. Accurate gross classification using imaging is considered to be essential for the determination of the correct treatment strategy and the estimates of the patients' prognosis.