Aims: to evaluate the accuracy of LR-5 category from the latest Contrast-Enhanced Ultrasound algorithm (ACR CEUS LI-RADSv 2017) for the noninvasive diagnosis of hepatocellular carcinoma (HCC), in a real-life cohort of high-risk patients.
Material and methods: We retrospectively re-analysed the CEUS studies of 464 focal liver lesions (FLL) in 382 patients at high-risk for HCC (liver cirrhosis of any aetiology, chronic B or C hepatitis with severe fibrosis) using the ACR CEUS LI-RADSv 2017 algorithm. CEUS LI-RADS categories used for the diagnosis of HCC were: CEUS LR-5 (definitely HCC) and CEUS LR-TIV (HCC with macrovascular invasion). Contrast-enhanced CT, contrast-enhanced MRI, or histology were used as diagnostic reference methods to evaluate the CEUS LI-RADS classification of the 464 lesions.
Results: According to the reference method, the 464 lesions were classified as follows: 359 HCCs, 68 non-HCC-non-malignant lesions and 37 non-HCC malignant lesions. The diagnostic accuracy of LR-5 category for the diagnosis of hepatocellular carcinoma was 76.9%. Sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) were 71.9%, 94.3 %, 97.7% and 49.5%, respectively.
Conclusions: LR-5 category from ACR CEUS LI-RADSv 2017 algorithm, has good sensitivity, excellent specificity, and PPV for the diagnosis of HCC. The HCC rate increases from LR-3 to LR-5.