Ferumoxides-enhanced MR in the Detection of Hepatocellular Carcinoma: Comparison with Combined CT During Arterial Portography and CT Hepatic Arteriography.
10.3348/jkrs.2001.44.2.177
- Author:
Yoong Ki JEONG
1
;
Seung Hoon KIM
;
Jong Hwa LEE
;
Jae Cheol HWANG
;
Soo Youn HAM
;
Neung Hwa PARK
;
Chang Woo NAM
;
Jae Hee SEO
;
Seoung Oh YANG
Author Information
1. Department of Radiology, Ulsan University Hospital, University of Ulsan College of Medicine, Korea.
- Publication Type:Original Article
- Keywords:
Iron;
Liver, MR;
Liver, CT;
Liver neoplasms;
Magnetic resonance (MR), contrast enhancement
- MeSH:
Angiography*;
Carcinoma, Hepatocellular*;
Follow-Up Studies;
Humans;
Iron;
Laparotomy;
Liver;
Liver Neoplasms;
Magnetic Resonance Imaging;
Portography*;
Protons;
ROC Curve;
Sensitivity and Specificity;
Ultrasonography
- From:Journal of the Korean Radiological Society
2001;44(2):177-186
- CountryRepublic of Korea
- Language:Korean
-
Abstract:
PURPOSE: Purpose: To compare the diagnostic accuracy of ferumoxides-enhanced MR with that of combined CT during arterial portography (CTAP) and CT hepatic arteriography (CTHA) in the preoperative detection of hepatocellular carcinoma (HCC). MATERIALS AND METHODS: For preoperative evaluation, 20 patients with HCC underwent ferumoxides-enhanced MR and combined CTAP and CTHA. The MR protocol included fat-suppressed respiratory-triggered fast spin echo, T2*-weighted fast multiplanar gradient-recalled acquisition in the steady state, proton density-weighted fast multiplanar spoiled gradient-recalled echo, and breath-hold in-phase T1-weighted fast multiplanar spoiled gradient-recalled echo. In all patients, laparotomy was performed. The presence or absence of HCC was confirmed by pathologic examination in the resected liver and by intraoperative ultrasonography of remaining liver, or by follow up. Images were reviewed by three radiologists working independently; regarding the presence or absence of HCC in each segment, each observer assigned one of five confidence levels. A receiver operating characteristic (ROC) curve was fitted to these confidence ratings, and the diagnostic accuracy of each modality was evaluated by calculating the Az value (area under the ROC curve) and compared with that of other modalities. The sensitivity and specificity of each modality in the detection of HCC were also calculated and compared, and using a κstatistic, inter-observer agreement for each modality was assessed. RESULTS: In 28 of 160 liver segments, 30 HCCs were present. For ferumoxide-enhanced MR the mean Az value was 0.958, and for combined CTAP and CTHA this value was 0.948. The difference was not statistically significant. The mean sensitivities of ferumoxide-enhanced MR and combined CTAP and CTHA were 92.9% and 90.9%, respectively, the difference being statistically insignificant. The mean specificities of these modalities were, respectively, 98.9% and 93.6%. The difference was statistically significant. For both ferumoxide-enhanced MR and combined CTAP and CTHA, interobserver agreement was excellent. CONCLUSION: In the preoperative detection of HCC, ferumoxide-enhanced MR imaging of the liver showed a diagnostic accuracy similar to that of combined CTAP and CTHA. Its specificity, however, was higher.
