Kupffer-phase Sonazoid ultrasound LI-RADS for liver cancer: diagnostic performance and algorithm modification

PURPOSE: This study aimed to enhance the noninvasive identification of hepatocellular carcinoma (HCC) by modifying the Sonazoid-based contrast-enhanced ultrasound (CEUS) Liver Imaging Reporting and Data System (LI-RADS) through integration of Kupffer-phase (KP) imaging and optimization of washout timing criteria. METHODS: This retrospective two-center study enrolled 558 patients with solitary liver nodules who underwent Sonazoid-based CEUS between August 2022 and September 2024. CEUS features were assessed according to LI-RADS v2017 using predefined washout time windows (2, 5, and 10 minutes). A modified LI-RADS algorithm incorporating KP hypoenhancement was developed. Interobserver agreement was evaluated using the Cohen kappa. Diagnostic performance metrics (area under the curve [AUC], sensitivity, specificity, and positive predictive value) were compared between the modified CEUS LI-RADS algorithms and the Japan Society of Hepatology (JSH) and Korean Liver Cancer Association (KLCA) guidelines using the McNemar and DeLong tests. RESULTS: For key CEUS features, interobserver agreement was moderate to almost perfect (κ=0.580-0.815). The modified CEUS LI-RADS-10min algorithm demonstrated superior performance over the other algorithms. By reclassifying LR-M nodules with early washout and mild KP hypoenhancement as LR-5, the modified CEUS LI-RADS-10min algorithm achieved the highest diagnostic performance for HCC (AUC, 0.782), improving sensitivity while maintaining specificity (73.7%). Its performance was comparable to that of the KLCA guideline. The JSH guidelines showed the highest sensitivity (91.7%) but the lowest specificity (53.9%). CONCLUSION: Integrating KP hypoenhancement with a 10-minute washout window improves the sensitivity of Sonazoid-based CEUS LI-RADS for diagnosing HCC while preserving specificity, providing feasible refinement aligned with established guidelines and promoting standardization.