Prospective, open-label, multi-centre study evaluating the accuracy of unenhanced versus SonoVue® enhanced ultrasonography (US) in the characterisation of focal liver lesions

Abstract

Objective: The aim of this study was to compare the accuracy of unenhanced versus SonoVueTM® enhanced ultrasonography in the characterisation of focal liver lesions.Methods: 127 patients with focal liver lesions were studied using top-of-range equipments with non-linear imaging facilities. Standardised baseline scans and SonoVue® enhanced US (dose: 2 × 2.4 ml and 1 × 4.8 ml) with optimised non-linear imaging modes were performed. On the basis of the tumour vascularity and pattern of enhancement in the vascular phase and degree of contrast uptake within the lesion relative to that of normal liver in the late phase, diagnosis was made as to whether the lesion was benign or malignant. SonoVue® enhanced diagnosis was compared with that of baseline using combined enhanced helical CT, MR, and/or histology as the reference standard.Results: Of the total of 134 lesions studied, 82 were malignant and 52 were benign on the basis of the reference standard. The sensitivity for correctly classifying malignant lesions for unenhanced US was 28% compared with 90% for SonoVue® enhanced US; the difference was highly significant (McNemar’s test: p<0.0001). The specificity for correctly classifying benign lesions for unenhanced US was 35% versus 81% for enhanced US; the difference was highly significant (McNemar’s test: p<0.0001). Of the 25 haemangiomas, diagnosis was correct for unenhanced US in 36% versus 84% for enhanced US with the characteristic display of peripheral nodular centripetal enhancement. Of the 13 FNHs, unenhanced US correctly diagnosed 31% whilst enhanced US was accurate in 77%.Conclusions: These results suggest that SonoVue® enhanced US with non-linear imaging modes is superior to unenhanced US in the characterization of focal liver tumours with significant reduction in the number of indeterminate diagnoses. Objective: The aim of this study was to compare the accuracy of unenhanced versus SonoVueTM® enhanced ultrasonography in the characterisation of focal liver lesions. Methods: 127 patients with focal liver lesions were studied using top-of-range equipments with non-linear imaging facilities. Standardised baseline scans and SonoVue® enhanced US (dose: 2 × 2.4 ml and 1 × 4.8 ml) with optimised non-linear imaging modes were performed. On the basis of the tumour vascularity and pattern of enhancement in the vascular phase and degree of contrast uptake within the lesion relative to that of normal liver in the late phase, diagnosis was made as to whether the lesion was benign or malignant. SonoVue® enhanced diagnosis was compared with that of baseline using combined enhanced helical CT, MR, and/or histology as the reference standard. Results: Of the total of 134 lesions studied, 82 were malignant and 52 were benign on the basis of the reference standard. The sensitivity for correctly classifying malignant lesions for unenhanced US was 28% compared with 90% for SonoVue® enhanced US; the difference was highly significant (McNemar’s test: p<0.0001). The specificity for correctly classifying benign lesions for unenhanced US was 35% versus 81% for enhanced US; the difference was highly significant (McNemar’s test: p<0.0001). Of the 25 haemangiomas, diagnosis was correct for unenhanced US in 36% versus 84% for enhanced US with the characteristic display of peripheral nodular centripetal enhancement. Of the 13 FNHs, unenhanced US correctly diagnosed 31% whilst enhanced US was accurate in 77%. Conclusions: These results suggest that SonoVue® enhanced US with non-linear imaging modes is superior to unenhanced US in the characterization of focal liver tumours with significant reduction in the number of indeterminate diagnoses.