Diagnostic efficacy of contrast-enhanced ultrasound versus MRI Liver Imaging Reporting and Data System (LI-RADS) for categorising hepatic observations in patients at risk of hepatocellular carcinoma

Abstract

To investigate the diagnostic efficacy of the contrast-enhanced ultrasound Liver Imaging Reporting and Data System (CEUS LI-RADS) for categorising hepatic observations in patients at risk of hepatocellular carcinoma (HCC) compared with magnetic resonance imaging (MRI) LI-RADS. CEUS and MRI data were analysed retrospectively according to the LI-RADS scheme. Follow-up results and pathological findings served as the reference standard. Receiver operating characteristic (ROC) curve analysis was used to reveal the area under the curve (AUC). The sensitivity, specificity, accuracy, and positive (PPV) and negative predictive values (NPV) of LR-5 for determining HCC were calculated. The intra-observer agreement of CEUS LI-RADS was also evaluated. Eighty-four patients with 86 liver observations were enrolled. Forty-two observations were classified as LR-5 by CEUS and MRI, respectively. Based on the reference standard, 53 nodules were HCC. The AUCs were 0.876 for CEUS and 0.873 for MRI, without a significant difference (Z=0.050, p=0.960). The sensitivity, specificity, PPV, NPV, and accuracy of LR-5 was 75.47%, 93.94%, 95.24%, 70.45%, 82.56% with CEUS and 73.58%, 90.9%, 92.86%, 68.18%, 80.23% with MRI, respectively. There was a significant difference in specificity between CEUS and MRI (p=0.006). There was almost perfect agreement for arterial phase hyperenhancement (k=0.870), substantial agreement for washout (k=0.765) and CEUS LI-RADS category (k=0.787). The CEUS LI-RADS scheme is an effective diagnostic tool for HCC with substantial intra-observer reliability. The diagnostic performance of CEUS LI-RADS for determining HCC was comparable to MRI LI-RADS, and the specificity of CEUS LR-5 was significantly higher.