Discrimination of Hepatocellular Carcinoma from Intrahepatic Cholangiocarcinoma Using Extracellular Volume Fraction (ECV) Threshold on Contrast-Enhanced CT

Objective: This study aimed to establish an extracellular volume fraction (ECV) threshold derived from triple-phase contrast-enhanced computed tomography (CT) to noninvasively differentiate hepatocellular carcinoma (HCC) from intrahepatic cholangiocarcinoma (ICC), thereby offering a clinically applicable diagnostic tool. Methods: In this retrospective analysis, 102 patients with pathologically confirmed HCC (n=68) or ICC (n=34) who underwent surgical resection were enrolled. All participants had non-contrast and dynamic contrast-enhanced CT scans. ECV was calculated from the mean CT attenuation values of the tumor and abdominal aorta across the three largest tumor sections. Group comparisons were performed using Welch’s t-test, and receiver operating characteristic （ROC） analysis was used to identify the optimal ECV threshold. Clinical-pathological and imaging features were compared with χ² test or Fisher’s exact tests, and multivariate logistic regression was employed to identify independent discriminators. Results: The mean ECV was significantly higher in ICC patients (39.7%±7.3%) compared to HCC patients (26.9%±7.0%) (P<0.0001). An ECV threshold of 34.3% yielded a sensitivity of 79.4% and specificity of 83.3% (AUC=0.895) for distinguishing HCC from ICC. Multivariate analysis identified ECV (OR: 0.688; 95% CI: 0.484–0.978; P<0.05) and washout pattern (OR: 0.009; 95% CI: 0.000–0.232; P<0.05）as independent predictors. Conclusion: An ECV value of 34.3% serves as a simple, noninvasive biomarker for discriminating between HCC and ICC. The combination of ECV and washout pattern significantly enhances diagnostic accuracy, highlighting the potential of ECV as a valuable imaging-derived parameter in hepatic tumor characterization.