Contrast-enhanced ultrasonography (CEUS)-CT/MRI fusion imaging substantially improved the visibility of small tumors that were inconspicuous on both US and US-CT/MRI fusion imaging, contributing to successful percutaneous radiofrequency ablation.

Percutaneous radiofrequency ablation (RFA) has been widely implemented as a curative treatment for liver malignancies. The overall survival post-RFA is comparable to that of surgical outcomes for small hepatocellular carcinomas (HCCs) (≤ 3 cm). US guidance for RFA is common in Asia because the real-time capability of US allows precise electrode placement. However, precisely targeting small liver malignancies with poor sonographic conspicuity in US-guided RFA is often challenging. To solve this problem, CEUS or real-time fusion imaging (FI) of US and CT/MRI have been used. Both approaches have been reported to improve the confidence for tumor localization in RFA for small HCCs with poor B-mode US visibility. However, when an index tumor is entirely invisible on B-mode US, neither method is ideal, especially for multiple-electrode RFA, due to inherent registration errors on FI or insonation-induced bubble destruction on CEUS with SonoVue. Therefore, there is a clinical need to compensate for the inherent registration errors of US-CT/MRI FI for inconspicuous tumors on both US and US-CT/MRI FI.

Recent studies reported that after adding CEUS using Sonazoid (GE Healthcare, Waukesha, WI, USA) or SonoVue to FI, 83.3% to 90% of target liver cancers that were initially difficult to visualize on FI became conspicuous, allowing them to be directly targeted for RFA. However, these studies had limitations, including a retrospective study design, a small sample size (< 30 inconspicuous tumors), or a short-term follow-up (< 1 year). Furthermore, no prospective study has explored the value of CEUS-CT/MRI FI, using both SonoVue and Sonazoid as contrast agents, for localizing index tumors and guiding RFA procedures in patients with inconspicuous tumors on B-mode US and reported its therapeutic impact on local tumor control.

Therefore, the present study investigated whether CEUS-CT/MRI FI improved the visibility of small (≤ 3 cm) primary and secondary liver tumors that were invisible or difficult to visualize on B-mode US for percutaneous RFA.

Inclusion Criteria:

• inconspicuous or invisible index tumors for ablation on US
• tumor size ≤ 3 cm
• pathologic diagnosis of primary or secondary liver malignancy or imaging-based diagnosis of HCC according to the American Association for the Study of Liver Disease guidelines or viable HCC according to the Liver Imaging Reporting and Data System treatment response algorithm
• consideration of curative-intent RFA

Exclusion Criteria:

• well-visible tumors (visibility score of 3 or 4) on the planning B-mode US
• the absence of available data from multiphase CT or MRI performed in the 3-month pre-procedure period
• poor quality registration of US-CT/MRI fusion imaging and RFA planned with palliative intent
• contraindications for the conventional RFA procedure at the authors' institution, which are uncontrollable coagulopathy (platelet count < 50,000/μL or international standard ratio ≥ 1.6), a low level of cooperation, impracticality of sedation, portal vein thrombus, the tumor abutting the portal vein, or larger bile ducts than the segmental branches