Diagnostic Performance of Gadoxetic Acid–enhanced Liver MR Imaging versus Multidetector CT in the Detection of Dysplastic Nodules and Early Hepatocellular Carcinoma

Abstract

Purpose To compare the diagnostic performance of gadoxetic acid-enhanced liver magnetic resonance (MR) imaging with that of contrast material-enhanced multidetector computed tomography (CT) in the detection of borderline hepatocellular nodules in patients with liver cirrhosis and to determine the Liver Imaging Reporting and Data System (LI-RADS) categories of these detected nodules. Materials and Methods The institutional review board approved this retrospective study and waived the informed consent requirement. Sixty-eight patients with pathologically proven dysplastic nodules (DNs) (low-grade DNs, n = 20; high-grade DNs, n = 17), early hepatocellular carcinomas (HCCs) (n = 42), or progressed HCCs (n = 33) underwent gadoxetic acid-enhanced MR imaging and multidetector CT. An additional 57 patients without any DNs or HCCs in the explanted livers were included as control subjects. Three radiologists independently graded the presence of liver nodules on a five-point confidence scale and assigned LI-RADS categories by using imaging findings. Jackknife alternative free-response receiver operating characteristics (JAFROC) software was used to compare the diagnostic accuracy of each modality in lesion detection. Results Reader-averaged figures of merit estimated with JAFROC software to detect hepatocellular nodules were 0.774 for multidetector CT and 0.842 for MR imaging (P = .002). Readers had significantly higher detection sensitivity for early HCCs with MR imaging than with multidetector CT (78.6% vs 52.4% [P = .001], 71.4% vs 50.0% [P = .011], and 73.8% vs 50.0% [P = .001], respectively). A high proportion of overall detected early HCCs at multidetector CT (59.4%) and MR imaging (72.3%) were categorized as LI-RADS category 4. Most early HCCs (76.2%) and high-grade DNs (82.4%) demonstrated hypointensity on hepatobiliary phase images. In total, 30 more LI-RADS category 4 early HCCs were identified with MR imaging than with multidetector CT across all readers. Conclusion Gadoxetic acid-enhanced MR imaging performed significantly better in the detection of high-risk borderline nodules, especially early HCCs, than did multidetector CT. RSNA, 2017 Online supplemental material is available for this article.

Prospective evaluation of gadoxetic acid magnetic resonance for the diagnosis of hepatocellular carcinoma in newly detected nodules ≤2 cm in cirrhosis

“…Up to now, the reported sensitivity of MR with gadoxetic acid‐enhanced MR (EOB‐MR) for the diagnosis HCC ranges between 70% and 100% . However, the retrospective design of the studies may overestimate the sensitivity and impedes the evaluation of specificity.…”

“…Up to now, the reported sensitivity of MR with gadoxetic acid‐enhanced MR (EOB‐MR) for the diagnosis HCC ranges between 70% and 100% . However, the retrospective design of the studies may overestimate the sensitivity and impedes the evaluation of specificity. In addition, some authors have considered the finding of hypo‐signal intensity (SI) in the HBP as equivalent to portal washout or even diagnostic of HCC regardless of the presence of arterial contrast enhancement in patients with cirrhosis .…”

Use of gadoxetate disodium in patients with chronic liver disease and its implications for liver imaging reporting and data system (LI‐RADS)

“…Therefore, the TP and HBP may detect early HCCs, not visible on any other sequence, and up to 38% of early HCCs may be seen only on the HBP . Eighty‐two percent of high‐grade dysplastic nodules and 76% of early HCCs demonstrate HBP hypointensity . Furthermore, in a multivariate analysis HBP hypointensity was an independent predictor of early HCC, adjusting for arterial phase hyperenhancement (APHE), restricted diffusion, and lesion size …”

“…26 Eightytwo percent of high-grade dysplastic nodules and 76% of early HCCs demonstrate HBP hypointensity. 27 Furthermore, in a multivariate analysis HBP hypointensity was an independent predictor of early HCC, adjusting for arterial phase hyperenhancement (APHE), restricted diffusion, and lesion size. 28,29 HBP hypointense nodules without APHE constitute an entity that is unique to HBA MRI.…”

Introduction to the Liver Imaging Reporting and Data System for Hepatocellular Carcinoma

“…Although recent cross-sectional studies using histology sampling as the reference method have indicated that LI-RADS CT/MRI and CEUS categories stratify the likelihood of malignancy accurately, 31,33,[41][42][43][44][45][46][47][48][49][50][51] it is not yet understood how LI-RADS diagnostic performance is affected by patient (demographics, anthropometrics, etiology and severity of cirrhosis), technical (imaging modality, field strength, type of contrast media), and radiologist (subspecialization, experience, practice setting) factors. In addition, because most LR-3 and LR-4 observations are followed, the natural history of such observations needs to be understood.…”