Reliability of LI-RADS Ancillary Features in HCC Detection

LI-RADS ancillary features help radiologists categorize liver observations during imaging assessments. These features provide essential support when major criteria do not yield a clear diagnosis. However, radiologists often question how much different readers agree on these markers. Consequently, a comprehensive meta-analysis recently investigated interreader agreement for individual features. This study examined data from 47 different original research papers. It included over 11,000 patients and nearly 14,000 hepatic observations. Radiologists in India can use this data to refine their diagnostic confidence.

Consistency of LI-RADS Ancillary Features

The researchers calculated pooled kappa values for 21 unique features. Notably, hepatobiliary-phase hypointensity showed the highest agreement with a kappa of 0.84. Other features like restricted diffusion and mild-moderate T2 hyperintensity also performed well. These markers demonstrated substantial consistency between different imaging experts. Furthermore, blood products in mass and fat in mass showed moderate to high agreement. However, some ancillary markers revealed significant variability among readers. Specifically, “enhancement paralleling blood pool” showed poor agreement with a kappa of 0.26. Therefore, clinicians must exercise caution when relying on this particular marker.

Factors Influencing Diagnostic Agreement

The study identified several factors that improved interreader consistency during evaluation. For instance, a higher proportion of patients with cirrhosis significantly increased agreement levels. Additionally, the meta-analysis highlighted gaps in current medical literature. For example, three specific features lacked any reported data in the reviewed studies. Specifically, these include ultrasound visibility, size stability, and size reduction. Consequently, further research must address these less-documented features. Therefore, standardizing the interpretation of these markers remains a priority. This effort will ultimately improve diagnostic accuracy for liver cancer in India.

Frequently Asked Questions

Q1: Which LI-RADS ancillary feature is most reliable among readers?

Hepatobiliary-phase hypointensity is the most reliable feature. It showed the highest interreader agreement in the meta-analysis with a substantial kappa value of 0.84.

Q2: Is “enhancement paralleling blood pool” a consistent diagnostic marker?

No, this feature demonstrated the lowest consistency in the study. It yielded a pooled kappa of only 0.26, suggesting significant variability between different readers.

References

1. Jung YJ et al. Interreader Agreement for Individual Ancillary Features in LI-RADS CT/MRI Version 2018: A Systematic Review and Meta-Analysis. AJR Am J Roentgenol. 2026 Apr 22. doi: 10.2214/AJR.26.34591. PMID: 42018773.
2. Cerny M, et al. LI-RADS Version 2018 Ancillary Features at MRI. Radiographics. 2018;38(7):1973-2001.
3. Indian National Association for Study of the Liver (INASL) and Indian Radiological and Imaging Association (IRIA). Joint Consensus Statement for the Diagnosis of Hepatocellular Carcinoma. J Clin Exp Hepatol. 2024.