[Diagnostic Value of Liver Imaging Reporting and Data System MRI on Primary Hepatocellular Carcinoma].

Wei Zhao,Wenzheng Li,Xiaoping Yi,Yigang Pei,Hui Liu,Liqing Zhang,Huaping Liu

DOI: https://doi.org/10.11817/j.issn.1672-7347.2016.04.007

2016-01-01

Abstract:OBJECTIVE To evaluate the diagnostic value of liver imaging reporting and classification criteria of data system (LI-RADS) MRI on primary hepatic carcinoma (HCC).
 METHODS The MRI images of 100 patients (72 patients with HCC, 4 with regenerative nodules, 13 with hemangioma, 4 with liver cyst, 3 with Liver abscess, 2 with inflammatory pseudotumor, 1 with hepatic arteriovenous fistula, 1 with Limitations of steatosis) with liver disease were collected in Xiangya Hospital from January, 2014 to January, 2015. All patients were at the risk of HCC. The images were reviewed according to the LI-RADS categorization by American College of Radiology (ACR) under a condition of blindness to the pathological data. The sensitivity, specificity and the area under the ROC curve in the diagnosis of HCC with the LI-RADS categorization were calculated by three associate professors from the Department of Radiology, Xiangya Hospital.
 RESULTS The area under the ROC curve was 0.925 in the 100 patients with HCC. Of the 100 pathologically confirmed patients, 20 cases were LR1 and LR2. However, the post-operation pathological examination showed benign tumor (negative predictive value was 100%). There were 25 HCC in 31 cases of LR3, 27 HCC in 29 cases of LR4 (positive predictive value was 93.1%). All of 20 cases of LR5 were HCC (positive predictive value was 100%). If LR1 and LR2 were considered negative, and LR3, 4, 5 were positive, then the sensitivity, the specificity and the area under the ROC curve was 100%, 71.4% and 0.925, respectively. If LR1, 2 and 3 were considered negative and LR4 and LR5 were positive, the sensitivity, the specificity and the area under the ROC curve was 100%, 90.9% and 0.974, respectively.
 CONCLUSION The MRI LI-RADS categorization is highly sensitive and specific in the diagnosis of HCC, which can improve the diagnostic accuracy.