CT liver perfusion (CTLP) has been well validated for hepatocellular carcinoma (HCC) detection, characterization, and treatment response evaluation. However, its role in HCC management algorithms remains unclear. This study aims to assess the diagnostic performance of CTLP alone or as an adjunct to MRI in patients considered for- or undergoing locoregional treatment for HCC. Thirty-nine patients under HCC surveillance (36 male, 31 cirrhotic, 16 pretreatment, 19 post-transarterial chemoembolization, 2 post-ablation) underwent MRI and CTLP in a single center within a 45-day interval. Two readers identified and characterized all observations on MRI using Liver Imaging Reporting and Data System (LI-RADS) v2018 criteria. CTLP assessment was based on Mean Slope of Increase (MSI), Time To Peak (TTP), Hepatic arterial Blood Flow (HaBF) and Hepatic Arterial Fraction (HAF) maps and established cut-offs. Diagnostic performance of MRI, CTLP, and their combination was evaluated for treated and untreated lesions using imaging or pathology as reference standard. Of the total 33 treated and 61 untreated lesions, 13 and 41 were considered viable HCCs. CTLP demonstrated 75.9 % sensitivity and 95 % specificity compared to 72.2 % and 100 % for MRI (p > 0.05). Combining both modalities increased sensitivity to 85.2 % (p < 0.05) and maintained specificity at 97.5 % (p > 0.05). The combined approach led to an LR category change in 5 treated and 19 untreated lesions and affected management in 5 cases. CTLP and MRI have comparable diagnostic performance for HCC. A combined approach improves sensitivity, without sacrificing specificity. This approach might enable more efficient patient selection for early and individualized loco-regional treatment.
Read full abstract