Characterization of liver lesions by real-time contrast-enhanced ultrasonography

2007 
Objective The aim of this study was to identify the most common patterns of various common liver lesions at real-time contrast-enhanced ultrasonography with second-generation contrast agents and their role in the differentiation of malignant from benign lesions. Patients and methods The enhancement pattern in the arterial phase and its modifications in subsequent portal and sinusoidal phases were separately evaluated in (i) 171 liver lesions detected at conventional ultrasonography in 125 noncirrhotic patients (87 metastases, six cholangiocellular carcinoma, 38 focal nodular hyperplasia, 30 hemangiomas, seven focal fatty sparing/changes, two hepatocellular adenomas and one hepatocellular carcinoma) and (ii) 75 lesions detected in 67 cirrhotic patients (66 hepatocellular carcinoma and nine dysplastic nodules). The final diagnosis was made by contrast-enhanced helical computed tomography and/or magnetic resonance imaging or by ultrasonography-guided biopsy when the diagnosis was equivocal at conventional imaging techniques (45 lesions). Results In noncirrhotic patients, the hypoechoic pattern in portal and sinusoidal phase (rapid washout) or the markedly hypoechoic or anechoic pattern in sinusoidal phase (marked late washout) showed a sensitivity, specificity and accuracy of 96.8, 100 and 98.2% for the diagnosis of malignancy. In cirrhotic patients, early arterial enhancement showed a sensitivity of 93.9% for the diagnosis of malignancy, with a specificity as low as 55.5% given the presence of arterial enhancement in 5/9 nodules resulted dysplastic at histological analysis. Conclusion Real-time contrast-enhanced ultrasonography provides sensitive and specific criteria for the differential diagnosis between benign and malignant liver lesions, and in most cases it may replace more expensive and invasive imaging techniques.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    47
    References
    71
    Citations
    NaN
    KQI
    []