Predicting Microvascular Invasion in Hepatocellular Carcinoma Using Triphasic Computed Tomography Perfusion Indices

Abstract

Background: Hepatocellular carcinoma (HCC) is a leading global cause of cancer mortality, and microvascular invasion (MVI) is a crucial prognostic factor linked to recurrence and poor outcomes. Current imaging modalities lack the accuracy needed for reliable preoperative MVI prediction. Objective: To assess the diagnostic performance of triphasic CT-derived perfusion parameters in detecting microvascular invasion in patients with HCC. Methods: In this retrospective study, 128 patients with histopathologically confirmed HCC underwent preoperative triphasic CT scans. Patients were stratified into MVI-positive and MVI-negative groups. Nine perfusion parameters, including hepatic arterial perfusion (HAP), portal venous perfusion (PVP), total liver perfusion (TLP), ΔHF, AEF, and rHF, were extracted. Group comparisons were conducted using t-tests or Mann–Whitney U tests. ROC analysis was performed to determine diagnostic performance. Results: ΔHF, AEF, and rHF were significantly elevated in MVI-positive patients and showed the highest diagnostic value (AUCs 0.740, 0.749, and 0.758, respectively). In contrast, PVP showed limited discrimination (AUC 0.594), and HAP and HPI showed no predictive power. Tumor size, AFP level, lesion number, and pathological grade were also significantly associated with MVI status (p<0.001). Conclusion: Triphasic CT-derived perfusion parameters—particularly ΔHF, AEF, and rHF—demonstrate promising noninvasive predictive value for MVI in HCC patients. These metrics may enhance preoperative risk stratification and guide clinical decision-making.