Objective To investigate the clinical value of gadolinium ethoxybenzyl diethy-lanetriaminepentaacetic acid (Gd-EOB-DTPA) enhanced magnetic resonance imaging (MRI) in the preoperative diagnosis of macrotrabecular-massive hepatocellular carcinoma (MTM-HCC).

Methods The diagnostic test was conducted. The clinicopathological data of 150 HCC patients who were admitted to the First Affiliated Hospital of Army Medical University from January 2019 to December 2020 were collected. There were 116 males and 34 females, aged (53±10)years. There were 38 MTM-HCC patients and 112 non-macrotrabecular-massive hepatocellular carcinoma (nMTM-HCC) patients. All patients received Gd-EOB-DTPA enhanced MRI examination. Observation indicators: (1) clinicopathological features of MTM-HCC and nMTM-HCC; (2) imaging features of MTM-HCC and nMTM-HCC; (3) imaging features for diagnosis of MTM-HCC. The normality test of continuous data was analyzed by the Kolmogorov-Smirnov test. Measurement data with normal distribution and homoscedasticity were represented as Mean±SD, and comparison between groups was analyzed using the independent sample t test. Measurement data with skewed distribution were represented as M(P25,P75), and comparison between groups was analyzed using Mann-Whitney U test. Count data were described as absolute numbers, and comparison between groups was analyzed using the chi-square test. Multivariate analysis was conducted using the Logistic regression model. The receiver operating characteristic (ROC) curve was used to evaluate diagnostic value of indicators, and Delong test was used for comparison.

Results (1) Clinicopathological features of MTM-HCC and nMTM-HCC: the alpha-fetoprotein and cases with microvascular invasion were 329 μg/L(20 μg/L,1 034 μg/L) and 24 for MTM-HCC patients, versus 25 μg/L(8 μg/L,200 μg/L) and 31 for nMTM-HCC patients, showing significant differences between the two groups (Z=-3.306, χ²=15.380, P<0.05). (2) Imaging features of MTM-HCC and nMTM-HCC: cases with regular morphology of tumor, intra-tumoral fat, arterial phase peritumoral enhancement, complete capsule, intratumoral necrosis or ischemia, peritumoral hypointense at hepatobiliary phase (HBP) were 6, 4, 20, 5, 28, 17 for MTM-HCC patients, versus 44, 40, 21, 43, 26, 11 for nMTM-HCC patients, showing significant differences between the two groups (χ²=7.049, 8.684, 16.399, 8.303, 31.368, 22.783, P<0.05). (3) Imaging features for diagnosis of MTM-HCC. ① Results of multivariate analysis showed that intratumoral fat, intratumoral necrosis or ischemia were independent predictors for MTM-HCC (hazard ratio=4.033,0.215, 95% confidence interval as 1.196-13.603, 0.079-0.588, P<0.05). ② Diagnostic efficacy: the arear under ROC curve, sensitivity and specificity of intratumoral fat combined with intratumoral necrosis or ischemia for diagnosis of MTM-HCC were 0.799(95% confidence interval as 0.718-0.880, P<0.05), 73.7%, 76.8%. The above indicators of intratumoral fat for diagnosis of MTM-HCC were 0.626(95% confidence interval as 0.530-0.721, P<0.05), 89.5%, 35.7%. The above indicators of intratumoral necrosis or ischemia for diagnosis of MTM-HCC were 0.752(95% confidence interval as 0.659-0.845, P<0.05), 73.7%, 76.8%. There were significant differences in the diagnostic efficacy between the intratumoral fat combined with intratumoral necrosis or ischemia and single intratumoral fat, between the intratumoral fat combined with intratumoral necrosis or ischemia and single intratumoral necrosis or ischemia, respectively (P<0.05).

Conclusions Intratumoral fat, intratumoral necrosis or ischemia on Gd-EOB-DTPA MRI are independent predictors for MTM-HCC. The two combined features has higher diagnostic efficacy. Gd-EOB-DTPA MRI can be used for pre-operative diagnosis of MTM-HCC.