Supplementary Material for: How to Best Detect Portal Vein Tumor Thrombosis in Patients with Hepatocellular Carcinoma Meeting the Milan Criteria: Gadoxetic Acid-Enhanced MRI versus Contrast-Enhanced CT

Introduction: The detection of portal vein tumor thrombosis (PVTT) in patients with hepatocellular carcinoma (HCC) meeting the Milan criteria is of critical importance as PVTT is known to be a contraindication to transplantation and an indicator of a dismal prognosis. Objective: To determine which modality may best detect PVTT, we compared the diagnostic performance of gadoxetic acid-enhanced MRI (GA-MRI) and contrast-enhanced CT (CECT) in HCC patients meeting the Milan criteria. Methods: We retrospectively enrolled 310 patients with HCCs meeting the Milan criteria who underwent both GA-MRI and CECT between June 2007 and May 2017. Among them, 44 patients were demonstrated to have PVTT while 266 patients had no PVTT. Two radiologists then assessed GA-MRI and CECT images for the presence of PVTT on a 5-point scale as well as vessel expansion, continuity with tumors, and enhancement on both modalities, as well as T2 hyperintensity and diffusion restriction on GA-MRI. The McNemar test was used to compare sensitivity and specificity of GA-MRI and CECT for the detection of PVTT, and Fisher’s exact test was used to compare their imaging features. Results: GA-MRI showed higher sensitivity values than CECT in detecting PVTT (reviewer 1, 93.2% [41/44] vs. 77.3% [34/44]; reviewer 2, 88.6% [39/44] vs. 70.5% [31/44]) (both p = 0.039). Specificity of GA-MRI and CECT demonstrated no difference (reviewer 1, 95.5% [254/266] vs. 95.1% [253/266]; reviewer 2, 97.0% [258/266] vs. 97.4% [259/266]) (both p > 0.999). Continuity with tumors and enhancement were more frequently observed on GA-MRI than on CECT (p < 0.050, both reviewers). Conclusion: GA-MRI detected PVTT more often than CECT in HCC patients meeting the Milan criteria and better depicted PVTT in continuity with tumors and those showing enhancement than CECT.