Compare the diagnostic performance of CEUS and MR/CT LIRADS categories for differentiating HCC and other non-HCC malignancies:a systematic review and meta-analysis

Lingling Li, Yixin Hu, Jianhua Zhou2020 · DOI

Introduction

Hepatocellular Carcinoma (HCC) represents a unique challenge and opportunity in oncology. Unlike many other cancers, HCC can often be diagnosed definitively through non-invasive imaging alone, bypassing the need for biopsy in high-risk patients. However, global clinical practice remains divided on the optimal imaging modality.

While Contrast-Enhanced Ultrasound (CEUS) is a cornerstone of diagnostic guidelines in many international regions, American guidelines have historically favored Computed Tomography (CT) and Magnetic Resonance Imaging (MRI). The primary hesitation regarding CEUS stems from the potential for non-HCC malignancies, such as intrahepatic cholangiocarcinoma, to mimic the imaging characteristics of HCC. This systematic review and meta-analysis protocol outlines a rigorous approach to evaluating the diagnostic accuracy of CEUS compared to CT and MRI using the standardized Liver Imaging Reporting and Data System (LI-RADS).

Methodology and Study Design

The proposed study is a systematic review and meta-analysis designed to synthesize evidence from primary diagnostic studies published since 2014. By focusing on the LI-RADS framework, the researchers aim to minimize the "noise" typically introduced by varying diagnostic criteria across different institutions and geographical regions.

Search and Selection Strategy

The authors will conduct a comprehensive search across major medical databases, including MEDLINE, Embase, CENTRAL, and Scopus. The inclusion criteria are strictly defined:

  • Population: Adults (over 18 years) at high risk for HCC.
  • Interventions: Diagnostic imaging using CEUS, CT, or MRI categorized according to LI-RADS.
  • Reference Standard: Confirmation via pathology or a reliable clinical reference standard.

Statistical Analysis

To provide a robust comparison of diagnostic performance, the study will employ a bivariate model. This approach is specifically chosen to account for the inherent trade-off between sensitivity (the ability to correctly identify HCC) and specificity (the ability to correctly rule out non-HCC malignancies). Furthermore, the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) tool will be used to evaluate the risk of bias and the applicability of the included studies, ensuring the final synthesis is based on high-quality evidence.

Expected Outcomes and Clinical Impact

As this document serves as a protocol, the quantitative meta-analysis results are forthcoming. The final report will provide several key metrics essential for clinical decision-making:

  • Sensitivity and Specificity: Comparative accuracy of each modality in differentiating HCC from other malignancies.
  • Summary Receiver Operating Characteristic (SROC) Curves: Visual representations of the overall diagnostic performance across different thresholds.
  • Likelihood Ratios (PLR/NLR): Practical tools for clinicians to determine how a specific imaging result shifts the post-test probability of disease.

Discussion and Limitations

The researchers anticipate certain challenges, primarily related to study heterogeneity. Variations in imaging hardware (e.g., 1.5T vs. 3T MRI) and the specific contrast agents used (e.g., SonoVue vs. Definity in CEUS) may influence results. While the current protocol does not specify subgroup analyses, the findings will be critical in identifying whether one modality offers a significant advantage in reducing misdiagnosis.

Ultimately, this meta-analysis aims to provide the high-level evidence required to reconcile differences between international clinical guidelines. By clarifying the diagnostic performance of CEUS relative to the "gold standards" of CT and MRI, this work will help standardize global clinical pathways for liver cancer screening and diagnosis.

Summary updated

Abstract

Compare the diagnostic performance of CEUS and MR/CT LIRADS categories for differentiating
BibTeX
@misc{li2020compare,
  author = {Lingling Li and Yixin Hu and Jianhua Zhou},
  title = {{Compare the diagnostic performance of CEUS and MR/CT LIRADS categories for differentiating HCC and other non-HCC malignancies:a systematic review and meta-analysis}},
  year = {2020},
  doi = {10.37766/inplasy2020.6.0056},
}
Figure