Prediction of HCC microvascular invasion with gadobenate-enhanced MRI: correlation with pathology.

ABSTRACT

PURPOSE: To assess the accuracy of gadobenate-enhanced MRI for predicting microvascular invasion (MVI) in patients operated for hepatocellular carcinoma (HCC). METHODS: The 164 patients who met the inclusion criteria were assigned to one of two groups the MVI-positive group and the MVI-negative group. Imaging results were compared between the two groups using the Kruskal test, chi-square test, independent sample t test, and logistic regression analysis. RESULTS: Differences in the capsule (p = 0.037) and margin (p = 0.004) of the tumor, rim enhancement (p = 0.002), peritumoral enhancement in the arterial phase (p < 0.001), and peritumoral hypointensity in the hepatobiliary phase (HBP) (p < 0.001) were statistically significant. The results of multivariate analysis identified rim enhancement in the arterial phase (odds ratio (OR) = 2.115; 95% confidence interval (CI), 1.002-4.464; p = 0.049) and peritumoral hypointensity in the HBP (OR = 5.836; 95% CI, 2.442-13.948; p < 0.001) as independent risk factors for MVI. Use of the two predictors in combination identified 32.79% (20/61) of HCCs with MVI with a specificity of 95.15% (98/103). CONCLUSIONS: Rim enhancement in the arterial phase and peritumoral hypointensity in the HBP were identified as independent risk factors for MVI in patients with HCC. KEY POINTS • Rim enhancement in the arterial phase and peritumoral hypointensity in the hepatobiliary phase were independent risk factors for microvascular invasion in patients with HCC. • Use of the two predictors in combination had a sensitivity of 32.79% and a specificity of 95.15% for predicting microvascular invasion.