Publication: Differentiation of hepatocellular carcinoma from non-hepatocellular malignant tumours of liver by chemical-shift mri at 3 t
No Thumbnail Available
Date
2019-10-01
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
W B Saunders Co Ltd
Abstract
AIM: To evaluate the diagnostic performance of chemical shift magnetic resonance imaging (MRI) in distinguishing hepatocellular carcinomas (HCCs) from non-hepatocellular malignant tumours (non-HCCs) of the liver.MATERIALS AND METHODS: Patients with a diagnosis of malignant liver tumours examined at 3 T MRI were included in this retrospective study. Forty-seven HCCs and 75 non-HCCs that were studied with chemical-shift MRI between January 2012 and October 2016 were retrieved from the radiology database. Two blinded observers measured the signal intensities of the tumours, adjacent normal-looking liver parenchyma, and spleen on chemical-shift MRI. The fat quantification for HCCs, non-HCCs, and adjacent normal-looking liver parenchyma were calculated by using the spleen as a reference standard. The subtraction scores were calculated by subtracting fat percentages in liver parenchyma from those in tumours. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of the fat percentage subtraction scores in distinguishing HCCs from non-HCCs were calculated.RESULTS: According to the optimal cut-off value acquired from both readers, a subtraction score >-0.26 was considered to be a HCC. Fat signal percentage subtraction scores were >=-0.26 in 45 of 47 HCCs and were <-0.26 in 69 of 75 non-HCCs. The sensitivity, specificity, PPV, and NPV of fat signal percentage subtraction score to differentiate HCCs from non-HCCs were found to be 95.7%, 89.3%, 84.9%, and 97.1%, respectively.CONCLUSION: Intracytoplasmic lipid in HCCs demonstrated by quantitative chemical-shift MRI may be a potentially powerful imaging biomarker to distinguish HCCs from the other malignant liver tumours.
Description
Keywords
Fat quantification, Cirrhotic liver, Gradient-echo, Diagnosis, Phase, Activation, Mechanism, Accuracy, Nodules, Cancer, Science & technology, Life sciences & biomedicine, Radiology, nuclear medicine & medical imaging
Citation
Collections
Metrikler