Evaluating the Negative Predictive Value of ACR LI-RADS 5 on Triphasic MRI for Hepatocellular Carcinoma in a Histopathology-Proven Cohort

Abstract

Hepatocellular carcinoma (HCC) is a leading cause of cancer-related mortality worldwide. The American College of Radiology Liver Imaging Reporting and Data System (ACR LI-RADS) provides standardized criteria for HCC diagnosis on imaging, but the negative predictive value (NPV) of LI-RADS 5 categorization requires validation against histopathology. The objective of the study was to evaluate the negative predictive value of ACR LI-RADS 5 category on triphasic MRI for diagnosing hepatocellular carcinoma using histopathology as the reference standard. This cross-sectional analytic study was conducted from April 2024 to September 2025 in the Department of Radiology & Imaging, BMU. Thirty patients with suspected HCC underwent triphasic MRI and were classified per LI-RADS v2018. Ultrasound-guided biopsies targeted the highest-category observations (LR-4/LR-5). Diagnostic accuracy parameters (sensitivity, specificity, PPV, NPV, likelihood ratios) were calculated using histopathology as the gold standard. Among 30 observations, 16 (53.3%) were categorized as LI-RADS 5 and 14 (46.7%) as LI-RADS 4. Histopathology confirmed HCC in 18 cases (60.0%). The NPV of LI-RADS 5 was 78.6% (11/14), with a sensitivity of 83.3%, specificity of 91.7%, and PPV of 93.8%. The positive and negative likelihood ratios were 10.0 and 0.18, respectively, with an overall accuracy of 86.7% (p<0.001). ACR LI-RADS 5 on triphasic MRI shows high specificity (91.7%) and PPV (93.8%) for HCC. The 78.6% NPV suggests LR-5 reliably confirms HCC, though non-LR-5 lesions may still be malignant, necessitating histopathological correlation in equivocal cases.  

CBMJ 2026 July: Vol. 15 No. 02 P:213-220