Identification of Imaging Predictors Discriminating Different Primary Liver Tumours in Patients with Chronic Liver Disease on Gadoxetic Acid-enhanced MRI: a Classification Tree Analysis
- 579 Downloads
To identify predictors for the discrimination of intrahepatic cholangiocarcinoma (IMCC) and combined hepatocellular-cholangiocarcinoma (CHC) from hepatocellular carcinoma (HCC) for primary liver cancers on gadoxetic acid-enhanced MRI among high-risk chronic liver disease (CLD) patients using classification tree analysis (CTA).
A total of 152 patients with histopathologically proven IMCC (n = 40), CHC (n = 24) and HCC (n = 91) were enrolled. Tumour marker and MRI variables including morphologic features, signal intensity, and enhancement pattern were used to identify tumours suspicious for IMCC and CHC using CTA.
On CTA, arterial rim enhancement (ARE) was the initial splitting predictor for assessing the probability of tumours being IMCC or CHC. Of 43 tumours that were classified in a subgroup on CTA based on the presence of ARE, non-intralesional fat, and non-globular shape, 41 (95.3 %) were IMCCs (n = 29) or CHCs (n = 12). All 24 tumours showing fat on MRI were HCCs. The CTA model demonstrated sensitivity of 84.4 %, specificity of 97.8 %, and accuracy of 92.3 % for discriminating IMCCs and CHCs from HCCs.
We established a simple CTA model for classifying a high-risk group of CLD patients with IMCC and CHC. This model may be useful for guiding diagnosis for primary liver cancers in patients with CLD.
• Arterial rim enhancement was the initial splitting predictor on CTA.
• CTA model achieved high sensitivity, specificity, and accuracy for discrimination of tumours.
• This model may be useful for guiding diagnosis of primary liver cancers.
KeywordsHepatocellular carcinoma Cholangiocarcinoma Combined hepatocellular-cholangiocarcinoma Gadoxetic acid Classification tree analysis
The scientific guarantor of this publication is J.M. Jang. The authors of this manuscript declare no relationships with any companies whose products or services may be related to the subject matter of the article. The authors state that this work has not received any funding. One of the authors, J. Kim, has significant statistical expertise.. Institutional Review Board approval was obtained. Written informed consent was waived by the Institutional Review Board. Methodology: retrospective, diagnostic or prognostic study, performed at one institution.
- 25.Leo Breiman , Jerome Friedman, Charles J. Stone and R.A. Olshen (1984) Classification and Regression Trees. Chapman and Hall/CRCGoogle Scholar
- 29.American College of Radiology (2013) Liver Imaging Reporting and Data System version 2013.1. http://www.acr.org/Quality-Safety/Resources/LIRADS/. Accessed Jan 2013
- 36.An C, Kim DW, Park YN, Chung YE, Rhee H, Kim MJ (2015) Single Hepatocellular Carcinoma: Preoperative MR Imaging to Predict Early Recurrence after Curative Resection. Radiology 142394Google Scholar