Characterization of 1- to 2-cm Liver Nodules Detected on HCC Surveillance Ultrasound According to the Criteria of the American Association for the Study of Liver Disease: Is Quadriphasic CT Necessary?

OBJECTIVE. The purpose of this study was to identify the essential number of phases from multiphasic CT for 1- to 2-cm hepatocellular carcinoma (HCC) on surveillance ultrasound and to compare the results with the American Association for the Study of Liver Disease (AASLD) standard (arterial phase hypervascularity and portal venous phase [PVP] or delayed phase hypovascularity). MATERIALS AND METHODS. The study included 110 newly detected nodules (1-2 cm; 36 HCC, 74 benign) in 96 patients detected in an HCC surveillance program. Three radiologists prospectively evaluated the attenuation of each nodule relative to the liver on each phase of quadriphasic CT. Univariate and multivariate logistic regression analyses were used to identify parameters associated with HCC. Multiple combinations of phases were compared with the AASLD standard. RESULTS. Only arterial phase hypervascularity and delayed phase hypovascularity were significantly associated with HCC both on univariate (odds ratio, arterial phase 7.51 [95% CI, 2.79-20.20]; delayed phase, 2.80 [1.14-6.90]) and multivariate analyses (arterial phase, 11.30 [4.30-29.68]; delayed phase, 4.39 [1.20-16.13]). The combination of arterial phase and delayed phase yielded the highest specificity (99%) and sensitivity (57%). There was no significant difference between AASLD standard (sensitivity, 57%; specificity, 98%) versus biphasic (arterial phase hypervascularity and delayed phase hypovascularity: sensitivity, 57%; p = 1 and specificity, 99%; p = 0.32), triphasic (arterial phase hypervascularity and unenhanced or PVP hypovascularity: sensitivity, 53%; p = 0.325 and specificity, 97%; p = 0.32), or quadriphasic combination (arterial phase hypervascularity and unenhanced, PVP or delayed phase hypovascularity: sensitivity, 57%; specificity, 97%), whereas the sensitivity of biphasic arterial phase and PVP was significantly lower (39% vs 57%, p = 0.022). CONCLUSION. For diagnosing 1- to 2-cm HCC detected on surveillance ultrasound, arterial phase and delayed phase are two essential phases, providing higher sensitivity than the combination of arterial phase and PVP, and equal performance with triphasic and quadriphasic combinations. The biphasic combination of arterial phase and delayed phase may replace quadriphasic CT recommended by AASLD.