Evaluation of combined Gd-EOB-DTPA and gadobutrol magnetic resonance imaging for the prediction of hepatocellular carcinoma grading

Abstract

Background

Tumor biopsy is not essential for the diagnosis of hepatocellular carcinoma (HCC); however, grading remains important for the prognosis.

Purpose

To investigate whether combined Gd-EOB-DTPA and gadobutrol liver magnetic resonance imaging (MRI) can predict HCC grading.

Material and Methods

Thirty patients (66.6 ± 7.3 years) with histologically confirmed HCC (grade 1, n = 5; grade 1–2, n = 6; grade 2, n = 13; grade 2–3, n = 2; grade 3, n = 4) underwent two liver MRIs, one with gadobutrol and one with Gd-EOB-DTPA, on consecutive days. Blinded to grading, two radiologists reviewed the gadobutrol and Gd-EOB-DTPA images in consensus with respect to: (i) HCC hyper-/iso-/hypointensity in the arterial, portal-venous/delayed, and Gd-EOB-DTPA hepatocellular phase; and (ii) morphologic tumor features (encapsulated growth, vessel invasion, heterogeneity, liver capsule infiltration, satellite metastases).

Results

A significant correlation with grading was not found for either the combined dynamic information of all gadobutrol phases (r = −0.187, P = 0.331) or all the Gd-EOB-DTPA phases (r = 0.052, P = 0.802). No correlation with grading was found for a combination of arterial and hepatocellular phase in Gd-EOB-DTPA MRI (r = 0.209, P = 0.305), a combination of both arterial phases (gadobutrol and Gd-EOB-DTPA) with the Gd-EOB-DTPA hepatocellular phase (r = 0.240, P = 0.248), or a combination of all available gadobutrol and Gd-EOB-DTPA phases (r = 0.086, P = 0.691). For all gadobutrol information (dynamic phases and morphology; r = 0.049, P = 0.801) and for all Gd-EOB-DTPA information (r = 0.040, P = 0.845), no correlation with grading was found. Hepatocellular Gd-EOB-DTPA phase iso-/hyperintensity never occurred in grade 3 HCCs.

Conclusion

Histological HCC grading cannot be predicted by combined Gd-EOB-DTPA/gadobutrol MRI. However, Gd-EOB-DTPA hepatocellular phase iso-/hyperintensity was never detected in grade 3 HCCs.

Keywords

Hepatocellular carcinoma grading magnetic resonance imaging (MRI)Gd-EOB-DTPA gadoxetate gadobutrol

Introduction

Hepatocellular carcinoma (HCC) is the sixth most common cancer entity worldwide and the third leading cause of cancer related death (1). The current European guideline allows the diagnosis of HCC in cirrhotic patients without a tumor biopsy if the tumor is larger than 1 cm and presents with a typical arterial enhancement and portal-venous or delayed phase wash-out (“radiological hallmarks”) in at least one to two different cross-sectional imaging modalities (2). With respect to the stage of tumor disease, liver function, and treatment options, the histological grading of the tumor still remains a significant factor for the prediction of survival in HCC patients (3). Given the risk of complications in biopsies, it would be desirable to predict the HCC grading by non-invasive cross-sectional imaging.

Gd-EOB-DTPA (gadolinium-ethoxybenzyl-diethylenetriamine pentaacetic acid) is a liver-specific intracellular magnetic resonance imaging (MRI) contrast agent that initially distributes into vascular and extracellular spaces, but subsequently accumulates in the hepatocytes and bile ducts (with up to 50% hepatobiliary excretion). Accordingly, Gd-EOB-DTPA presents a special hepatocellular parenchymal contrast in late-phase images and was found to improve sensitivity and specificity in the detection and classification of liver lesions (4 –9). Recently, few studies have dealt with the question whether imaging features, especially enhancement patterns, in Gd-EOB-DTPA MRI are linked to HCC grading. In these studies inconsistent results have been reported (10 –12). But it should be taken into consideration that detection of arterial HCC blushing in Gd-EOB-DTPA MRI might be impaired by more frequent breathing artifacts, blurring the arterial phase images, and a lower signal intensity increase in normal liver tissue (13,14). Additionally, HCC washout evaluation in Gd-EOB-DTPA MRI might be affected by an early onset of liver-specific enhancement of the surrounding liver tissue or liver-specific enhancement of the HCC itself. Therefore, the combination with a non-liver-specific extracellular contrast agent like gadobutrol appears promising, particularly because gadobutrol has been reported to provide a higher arterial enhancement in normal liver tissue than Gd-EOB-DTPA (15). We aimed to assess whether the combined MRI enhancement patterns of the contrast agents gadobutrol and Gd-EOB-DTPA correlate with HCC grading.

Material and Methods

Study design

The local ethics committee approved prospective data collection and analysis. All patients provided written informed consent for the study participation and MRI. Study inclusion criteria were known HCC, which was confirmed and graded by core needle biopsy, sufficient renal function (serum creatinine <2 mg/dl, creatinine clearance >30 mL/min), and patient age ≥18 years. Exclusion criteria were contraindications for MRI and/or gadobutrol/Gd-EOB-DTPA administration. HCCs were histologically classified according to the WHO criteria (grade 1, well differentiated; grade 2, moderately differentiated; grade 3, poorly differentiated) (16). All patients underwent two contrast-enhanced liver MRIs on consecutive days, one with gadobutrol and one with Gd-EOB-DTPA.

Patients

Between July 2010 and June 2013, 30 patients (3 women, 27 men) with a mean age of 66.6 ± 7.3 years (age range, 47–80 years) were included. Twenty-three (76.7%) patients suffered from liver cirrhosis; 19 patients were classified as Child-Pugh A and four patients as Child-Pugh B (with a maximum score of 8 Child-Pugh points). Causes of liver cirrhosis were chronic viral hepatitis (n = 7), alcohol abuse (n = 6), non-alcoholic steatohepatitis (n = 6), hemochromatosis (n = 1), and cardiogenic (n = 1). In two patients, etiology of liver cirrhosis remained cryptogenic.

MRI with Gd-EOB-DTPA and gadobutrol

All MR scans were performed on a 1.5-Tesla system (Magnetom Avanto, Siemens Healthcare, Erlangen, Germany). The scanning protocol consisted of axial T1-weighted (T1W) dual gradient echo in- and opposed-phase and axial T2-weighted (T2W) turbo spin echo (TSE) sequences (Table 1). Subsequently, an axial T1W 3D volume interpolated breath-hold examination (VIBE) (Table 1) was acquired before and after gadolinium-based contrast medium injection. A CARE Bolus technique (Siemens Healthcare) was used for the exact timing of the arterial phase. Portal-venous and delayed images were acquired 30 and 120 s after the start of the arterial phase acquisition. In Gd-EOB-DTPA MRI (Primovist® 0.25 mmol/mL injection solution, Bayer HealthCare, Leverkusen, Germany) additional axial and coronal T1W fast low angle shot (FLASH) 2D images were acquired 20 min after Gd-EOB-DTPA administration (Table 1). In all patients, a manufactured syringe with 10 mL contrast agent was used for Gd-EOB-DTPA imaging. For gadobutrol imaging (Gadovist® 1.0 mmol/mL injection solution, Bayer HealthCare) 0.1 mL contrast agent per kg body weight was administered. Both contrast media were injected at a rate of 2.0 mL/s, followed by 20 mL saline at the same rate.

Table 1.

MRI sequence protocol of the gadobutrol and Gd-EOB-DTPA liver MRI (T1W FLASH 2D after contrast agent administration was acquired only in Gd-EOB-DTPA MRI).

	Unenhanced			Post contrast
	T1-in-phase	T1-opposed-phase	T2 TSE	T1 3D dynamic	T1 2D
Sequence type	FLASH 2D	FLASH 2D	T2 TSE fs	3D VIBE	FLASH 2D
TR [ms]	117	117	3520	3.96	120/79
TE [ms]	4.92	2.22	82	1.5	2.47/2.39
Flip angle [°]	70	70	150	12	70
FOV [mm]	400	400	400	400	400
Slice thickness [mm]	7	7	7	3.5	7
Orientation	axial	axial	axial	axial	axial/coronal

FLASH, fast low angle shot; FOV, field of view fs, fat saturation; T1W, T1-weighted; T2W, T2-weighted; TE, echo time; TR, repetition time; TSE, turbo spin echo; VIBE, volume interpolated breath hold examination (acquired before and after gadolinium based contrast medium injection).

Data evaluation

Two radiologists (11 and 6 years of experience in abdominal imaging) evaluated all MR images in consensus while blinded to HCC grading. In patients with more than one HCC lesion, identification of the sampled lesion was performed using biopsy reports and data correlation. Arterial images were evaluated for the presence and severity of breathing artifacts (mild, moderate, severe) and the degree of visual arterial HCC enhancement (unsatisfactory, moderate, excellent). In cases without arterial HCC enhancement, the degree of visual arterial enhancement of the surrounding liver tissue, especially of existing regenerative nodules, was assessed. HCC signal intensities were assessed in comparison to the surrounding tissue in arterial and portal-venous/delayed phase images for both contrast media and in the hepatocellular Gd-EOB-DTPA phase. Accordingly, HCCs were rated as hypo-, iso-, or hyperintense. Unenhanced and arterial phase images were compared to assess the presence of arterial HCC enhancement. Arterial and portal-venous/delayed phase images were compared to assess the presence of HCC washout. In visually uncertain cases HCC signal intensities were measured and compared in the different phases. Additionally, HCCs were evaluated for the presence of the morphologic features encapsulated growth, vessel invasion, lesion heterogeneity, liver capsule infiltration, and satellite metastases.

According to Choi et al., the dynamic HCC enhancement patterns in gadobutrol and Gd-EOB-DTPA MRI were rated as being typical (arterial enhancing nodule with portal-venous/delayed phase washout, (Figs. 1 and 2) or atypical (11). Additionally, a scoring system of MRI findings was established, which was intended to reflect the HCC grade and which included the contrast patterns as well as the additional morphologic HCC features. In this score, dynamic and morphologic HCC imaging features, which are more likely found in poorly differentiated HCCs, were rated with +1 or +0.5. HCC imaging features, which are more likely seen in well differentiated HCCs, were rated with −1 (e.g. iso- or hyperintensity in the hepatocellular Gd-EOB-DTPA phase, which is known to be associated with lower histologic grade, was rated with −1 point) (Table 2) (2,11,17,18). Using these scoring points, totals for different combinations of the dynamic gadobutrol and Gd-EOB-DTPA phases and the additional morphologic HCC features were calculated by summing up the scoring points.

Fig. 1.

A 63-year-old female patient suffering from multifocal HCC. The biopsied HCC nodule in segment seven (grade 1, arrow) showed a typical dynamic enhancement pattern with arterial enhancement and portal-venous washout: (a) unenhanced T1W in-phase, (b) arterial phase gadobutrol, (c) portal-venous phase gadobutrol, (d) arterial phase Gd-EOB, (e) portal-venous phase Gd-EOB – but presented hyperintense in the hepatocellular Gd-EOB phase (f).

Fig. 2.

A 72-year-old male patient with multifocal HCC (grade 3, biopsy was taken from the large tumor nodule central in the right liver lobe, arrow). This HCC presented with a typical enhancement pattern (arterial enhancement, portal-venous washout, hypointensity in hepatocellular Gd-EOB phase) and a pseudocapsule in both gadobutrol ((a) unenhanced T1W in-phase, (b) arterial phase, (c) portal-venous phase) and Gd-EOB-DTPA MRI ((d) arterial phase, (e) portal-venous phase, (f) hepatocellular phase).

Table 2.

Point allocation of the HCC score, which aimed to reflect the HCC grading.

HCC nodule features	Points
Arterial hyperintensity (gadobutrol and Gd-EOB-DTPA)	+1
Portal-venous/delayed hypointensity (after arterial iso- or hyperintensity, gadobutrol and Gd-EOB-DTPA)	+1
Portal-venous/delayed isointensity (after arterial hyperintensity, gadobutrol and Gd-EOB-DTPA)	+0.5
Iso- or hyperintensity in hepatocellular Gd-EOB-DTPA phase	−1
Hypointensity in hepatocellular Gd-EOB-DTPA phase	+1
Encapsulated growth	+0.5
Vessel invasion	+1
Signal heterogeneity of the HCC	+1
Liver capsule infiltration	+1
Satellite metastases	+1

Statistical analysis

Statistical analysis was performed using SPSS Statistics 19.0 (IBM Corporation, Armonk, NY, USA). A normal distribution fit for the data was tested using the Kolmogorov-Smirnov test. Mean values and standard deviations were calculated for a normal distribution. Fisher’s exact test was used to compare the frequency of breathing artifacts and the degree of arterial HCC enhancement between contrast agents. Testing for frequency differences of arterial enhancement, portal-venous/delayed phase washout, and hepatocellular Gd-EOB-DTPA hypo- or iso-/hyperintensity with respect to grading was performed using the Chi-square test. For the Chi-square test, cases with an intermediate histological grading, such as grade 1–2, were adjusted to the higher grade, e.g. 2 in this example. Correlation between grading and the different score totals (representing different combinations of gadobutrol and Gd-EOB-DTPA enhancement findings and morphologic HCC features) was evaluated using Spearman’s correlation test. A P value of less than 0.05 was considered significant.

Results

Five HCCs were pathologically rated as grade 1, six as grade 1–2, 13 as grade 2, two as grade 2–3, and four as grade 3. The mean HCC size was 5.7 ± 3.3 cm (range, 1.9–17.0 cm).

Comparing the image quality of the arterial phase datasets, a non-significant tendency towards more frequent moderate to severe breathing artifacts was seen in Gd-EOB-DTPA MRI (moderate, n = 4; severe, n = 2) compared to gadobutrol MRI (moderate, n = 1; severe, n = 0; P = 0.103). No significant difference between the visual degree of arterial HCC enhancement was detected comparing Gd-EOB-DTPA MRI (moderate to excellent enhancement in 24 cases) and gadobutrol MRI (moderate to excellent enhancement in 25 cases, P = 1.00).

In gadobutrol MRI 21 HCCs (70%) and in Gd-EOB-DTPA MRI 24 HCCs (80%) presented with a typical dynamic enhancement pattern. Both typical and atypical enhancement patterns were detected in all HCC grades, irrespective of the applied contrast media. No correlation between the presence of a typical dynamic enhancement pattern and grading was found (gadobutrol: P = 0.349, Gd-EOB-DTPA: P = 0.645). Contingency table evaluation for HCC enhancement features in gadobutrol/Gd-EOB-DTPA MRI and grading (Tables 3 and 4) showed no significant distribution difference. Considering only the arterial phase, 27 HCCs enhanced arterially in both contrast agents. One poorly differentiated HCC did not enhance arterially with any contrast media. In two HCCs arterial enhancement was detected exclusively by gadobutrol or Gd-EOB-DTPA MRI (in one moderately differentiated HCC exclusively by gadobutrol MRI, and in one well differentiated HCC exclusively by Gd-EOB-DTPA MRI), but the arterial HCC enhancement was not detected by the arterial phase of the other MRI, which was blurred in both cases due to breathing artifacts. Considering only the portal-venous/delayed phase, washout was detected in 22 HCCs by gadobutrol MRI (73.3%) and in 25 HCCs by Gd-EOB-DTPA MRI (83.8%). In four cases (grade 1 to 3) no washout was seen in both contrast agents. In four cases washout was only detected by Gd-EOB-DTPA MRI but not by gadobutrol MRI. Grading of these four HCCs was in the range of grades 1–3 and all HCCs were hypointense in the liver-specific Gd-EOB-DTPA phase. Additionally, one HCC (grade 1), which was isointense in the liver-specific Gd-EOB-DTPA phase, showed washout exclusively by gadobutrol MRI but not by Gd-EOB-DTPA MRI. Regarding the hepatocellular Gd-EOB-DTPA phase, one out of five grade 1 HCCs appeared hyperintense (Fig. 1) and three HCCs (one HCC grade 1 and two HCCs grade 2) appeared isointense. Grade 3 HCCs never appeared iso- or hyperintense on hepatocellular Gd-EOB-DTPA images.

Table 3.

Contingency table of HCC enhancement features in Gd-EOB-DTPA MRI and HCC grading.

	Grade 1	Grade 2	Grade 3	P value
Arterial enhancement
Yes	5 (100%)	18 (95%)	5 (83%)	0.501
No	0 (0%)	1 (5%)	1 (17%)	0.501
Portal-venous/delayed wash-out
Yes	4 (80%)	17 (89%)	4 (67%)	0.416
No	1 (20%)	2 (11%)	2 (33%)	0.416
Hypointensity in hepatocellular phase
Yes	3 (60%)	17 (89%)	6 (100%)	0.127
No	2 (40%)	2 (11%)	0 (0%)	0.127

Absolute numbers of cases (according percentage values) are displayed.

Table 4.

Contingency table of HCC enhancement features in gadobutrol MRI and HCC grading.

	Grade 1	Grade 2	Grade 3	P value
Arterial enhancement
Yes	4 (80%)	19 (100%)	5 (83%)	0.153
No	1 (20%)	0 (0%)	1 (17%)	0.153
Portal-venous/delayed washout
Yes	4 (80%)	15 (79%)	3 (50%)	0.352
No	1 (20%)	4 (21%)	3 (50%)	0.352

Absolute numbers of cases (according percentage values) are displayed.

An encapsulated HCC growth was detected in seven HCCs (23.3%) by gadobutrol MRI and only in six HCCs (20%) by Gd-EOB-DTPA MRI. HCC vessel invasion was not found by gadobutrol MRI nor Gd-EOB-DTPA MRI. HCC heterogeneity was detected by both contrast agents in 13 cases (43.3%) and infiltration of liver capsule was suspected in 10 HCCs (33.3%) by both contrast media. Satellite metastases were found in 10 HCCs (33.3%) by gadobutrol MRI and in nine HCCs (30%) by Gd-EOB-DTPA MRI. Interestingly, in the HCC in which the satellite metastases were missed by Gd-EOB-DTPA MRI, the arterial Gd-EOB-DTPA phase was severely affected by breathing artifacts and the satellite metastases could not be detected in the portal-venous/delayed or liver-specific phase because the HCC and the satellite metastases did not washout in the portal-venous/delayed phase and the satellite metastases appeared isointense in the liver-specific Gd-EOB-DTPA phase.

None of the different score sums (representing different combinations of enhancement information of both contrast media and the morphologic HCC features) correlated significantly with the HCC grading (Table 5).

Table 5.

Results of the correlation analysis between the different HCC score sums (representing different combinations of enhancement information of both contrast media and morphologic HCC features) and the HCC grading.

Different HCC score formations versus HCC grading	r	P value
Arterial + portal-venous/delayed phase by gadobutrol	−0.187	0.331
Arterial + hepatocellular phases by Gd-EOB	0.209	0.305
Arterial + portal-venous/delayed + hepatocellular phase by Gd-EOB	0.052	0.802
Arterial + hepatocellular phase by Gd-EOB + arterial gadobutrol phase	0.240	0.248
All available dynamic phases by gadobutrol and Gd-EOB	0.086	0.691
All dynamic phase + morphologic feature information by gadobutrol	0.049	0.801
All dynamic phase + morphologic feature information by GD-EOB	0.040	0.845

Gd-EOB, Gd-EOB-DTPA.

Discussion

Thirty patients with histologically confirmed and graded HCC were examined by gadobutrol and Gd-EOB-DTPA MRI to investigate whether the combination of the enhancement patterns of both contrast agents and HCC grading are correlated. This study did not find any correlation between the enhancement patterns of both contrast media and tumor grading, nor was a correlation seen between a score, based on the contrast patterns in both contrast agents and morphologic HCC features, and grading. The single potentially helpful finding with respect to grading was that iso- or hyperintensity in the hepatocellular Gd-EOB-DTPA phase never occurred in undifferentiated HCCs, but only in HCCs grades 1 and 2.

To the best of our knowledge, this is the first study analyzing a combination of the enhancement patterns of both a non-liver-specific extracellular and a liver-specific intracellular contrast agent for its correlation with HCC grading. Choi et al. reported 304 histologically proven HCCs investigated by Gd-EOB-DTPA MRI and classified them depending on their dynamic enhancement pattern in two major groups of typically and atypically enhancing lesions (11). A lesion was defined as typically enhancing if there was an arterial enhancement and a washout in the portal-venous/delayed phase (72%). Otherwise a lesion was classified as atypical. In our study, a typical dynamic HCC enhancement pattern was found in 70% of HCCs by gadobutrol and in 80% by Gd-EOB-DTPA. This discrepancy between both contrast agents was caused by different ratings of portal-venous/delayed phase washout and not by different ratings of arterial HCC enhancement. Considering the arterial phase, no difference in the visual degree of arterial HCC enhancement was detected between gadobutrol and Gd-EOB-DTPA. Although a higher arterial enhancement of normal liver tissue was reported repeatedly for gadobutrol, no study exists comparing the arterial enhancement of HCCs between both contrast agents (14,15). In our cohort a tendency towards more frequent breathing artifacts in Gd-EOB-DTPA MRI was seen, as described by Davenport et al. (13), but this did not result in differences in the current arterial enhancement pattern reading. The difference in the classification into typically or atypically enhancing HCCs was caused by five differently rated HCC washouts by gadobutrol and Gd-EOB-DTPA MRI. It remains unclear whether this was caused by a potential masking of delayed washout by the liver-specific Gd-EOB-DTPA enhancement in one case or by a potential overestimation of washout caused by an early onset of liver-specific Gd-EOB-DTPA enhancement of the surrounding liver tissue in the other cases. In terms of additional morphologic HCC features, in one HCC the satellite metastases were missed by a single Gd-EOB-DTPA MRI, which was severely affected by breathing artifacts. Taking all these aspects into account, in single cases the combined gadobutrol and Gd-EOB-DTPA MRI can compensate for more frequent breathing artifacts occurring in Gd-EOB-DTPA MRI, but otherwise the combination of both contrast agents is not superior to the single use of Gd-EOB-DTPA.

Choi et al. reported that HCCs with an atypical dynamic Gd-EOB-DTPA enhancement pattern had a lower histologic grade compared to HCCs with a typical enhancement pattern and that HCCs appearing iso- or hyperintense on hepatocellular phase images were more differentiated compared to HCCs appearing hypointense on late Gd-EOB-DTPA images (11). Additionally, they found that patients with iso- or hyperintense HCCs in the Gd-EOB-DTPA hepatocellular phase had a significantly longer time span until tumor recurrence after surgery, and concluded that the HCC signal intensity on late Gd-EOB-DTPA images might be a useful prognostic imaging marker. An et al. found a tendency towards higher HCC grades in tumors presenting with an arterial enhancement in Gd-EOB-DTPA subtraction imaging in 175 patients with 201 subsequently surgically resected HCCs, but did not describe a correlation between hypointensity in the hepatocellular Gd-EOB-DTPA phase and grading (12). However, Kim et al., who investigated the enhancement patterns in 286 surgically confirmed HCCs by means of signal- and contrast-to-noise ratios, reported that these were independent from tumor grading (10).

Choi et al. reported that well differentiated HCCs tended to have an increased contrast enhancement in the hepatocellular Gd-EOB-DTPA phase, but that there was no significant correlation between the contrast enhancement ratio of the HCC and the grading (19). Huppertz et al. found a hepatocyte-selective uptake only in well differentiated HCCs (in two out of four) but not in moderately or poorly differentiated HCCs, and Lee et al. reported high signal intensity on hepatobiliary Gd-EOB-DTPA images in four well differentiated HCCs (17,18).

The present study is not without limitations. First, the number of included patients was small. However, recruiting patients who recently received a tumor biopsy to consent to participate in an MRI study being performed on 2 consecutive days involving the administration of two different contrast agents is challenging. Second, moderately differentiated HCCs were overrepresented in our cohort compared to the general distribution. Finally, since our patients were undergoing non-surgical treatments like systemic therapy or selective internal radiation therapy, we only had biopsy samples and no surgical specimens to compare with the MRI. And HCC grading is especially challenging for the pathologist in biopsy samples.

In conclusion, no correlation between combined gadobutrol and Gd-EOB-DTPA liver MRI features and HCC grading was seen. However, a lower histologic HCC grade was associated with iso- or hyperintensity on late Gd-EOB-DTPA images.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by Bayer Healthcare, Leverkusen, Germany.

References

Altekruse

McGlynn

Reichman

. Hepatocellular carcinoma incidence, mortality, and survival trends in the United States from 1975 to 2005. J Clin Oncol 2009; 27: 1485–1491.

European Association For The Study Of The Liver, European Organisation For Research And Treatment Of Cancer. EASL-EORTC clinical practice guidelines: Management of hepatocellular carcinoma. J Hepatol 2012; 56: 908–943.

Wayne

Lauwers

Ikai

. Preoperative predictors of survival after resection of small hepatocellular carcinomas. Ann Surg 2002; 235: 722,30:730–731.

Van Beers

Pastor

Hussain

. Primovist, eovist: What to expect? J Hepatol 2012; 57: 421–429.

Hammerstingl

Huppertz

Breuer

. Diagnostic efficacy of gadoxetic acid (primovist)-enhanced MRI and spiral CT for a therapeutic strategy: Comparison with intraoperative and histopathologic findings in focal liver lesions. Eur Radiol 2008; 18: 457–467.

Huppertz

Balzer

Blakeborough

. Improved detection of focal liver lesions at MR imaging: Multicenter comparison of gadoxetic acid-enhanced MR images with intraoperative findings. Radiology 2004; 230: 266–275.

Ichikawa

Saito

Yoshioka

. Detection and characterization of focal liver lesions: A Japanese phase III, multicenter comparison between gadoxetic acid disodium-enhanced magnetic resonance imaging and contrast-enhanced computed tomography predominantly in patients with hepatocellular carcinoma and chronic liver disease. Invest Radiol 2010; 45: 133–141.

Bottcher

Hansch

Pfeil

. Detection and classification of different liver lesions: Comparison of gd-EOB-DTPA-enhanced MRI versus multiphasic spiral CT in a clinical single centre investigation. Eur J Radiol 2013; 82: 1860–1869.

Reimer

Schneider

Schima

. Hepatobiliary contrast agents for contrast-enhanced MRI of the liver: Properties, clinical development and applications. Eur Radiol 2004; 14: 559–578.

10.

Kim

Choi

Kim

. Changes in the signal- and contrast-to-noise ratios of hepatocellular carcinomas on gadoxetic acid-enhanced dynamic MR imaging. Eur J Radiol 2013; 82: 62–68.

11.

Choi

Lee

Kim

. Hepatocellular carcinoma: Imaging patterns on gadoxetic acid-enhanced MR images and their value as an imaging biomarker. Radiology 2013; 267: 776–786.

12.

Park

Jeon

. Prediction of the histopathological grade of hepatocellular carcinoma using qualitative diffusion-weighted, dynamic, and hepatobiliary phase MRI. Eur Radiol 2012; 22: 1701–1708.

13.

Davenport

Viglianti

Al-Hawary

. Comparison of acute transient dyspnea after intravenous administration of gadoxetate disodium and gadobenate dimeglumine: Effect on arterial phase image quality. Radiology 2013; 266: 452–461.

14.

Zizka

Klzo

Ferda

. Dynamic and delayed contrast enhancement in upper abdominal MRI studies: Comparison of gadoxetic acid and gadobutrol. Eur J Radiol 2007; 62: 186–191.

15.

Kuhn

Hegenscheid

Siegmund

. Normal dynamic MRI enhancement patterns of the upper abdominal organs: Gadoxetic acid compared with gadobutrol. Am J Roentgenol 2009; 193: 1318–1323.

16.

Theise

Hepatocellular carcinoma. In: Bosman

Carneiro

Hruban

(eds). WHO Classification of Tumours of the Digestive System, 4th ed. Lyon: International Agency for Research on Cancer (IARC), 2010, pp. 205–216.

17.

Huppertz

Haraida

Kraus

. Enhancement of focal liver lesions at gadoxetic acid-enhanced MR imaging: Correlation with histopathologic findings and spiral CT–initial observations. Radiology 2005; 234: 468–478.

18.

Lee

Jung

. Paradoxical high signal intensity of hepatocellular carcinoma in the hepatobiliary phase of gd-EOB-DTPA enhanced MRI: Initial experience. Magn Reson Imaging 2011; 29: 83–90.

19.

Choi

Kim

Park

. Gadoxetate disodium-enhanced hepatobiliary phase MRI of hepatocellular carcinoma: Correlation with histological characteristics. Am J Roentgenol 2011; 197: 399–405.