- Email: [email protected]
CAIPIRINHA-Dixon-TWIST (CDT)-volume-interpolated breath-hold examination (VIBE) for dynamic liver imaging: Comparison of gadoterate meglumine, gadobutrol and gadoxetic acid
European Journal of Radiology 83 (2014) 2007–2012
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
CAIPIRINHA-Dixon-TWIST (CDT)-volume-interpolated breath-hold examination (VIBE) for dynamic liver imaging: Comparison of gadoterate meglumine, gadobutrol and gadoxetic acid Johannes Budjan a,∗ , Melissa Ong a , Philipp Riffel a , John N. Morelli b , Henrik J. Michaely a , Stefan O. Schoenberg a , Stefan Haneder a a Institute of Clinical Radiology and Nuclear Medicine, University Medical Center Mannheim, Medical Faculty Mannheim—Heidelberg University, Mannheim, Germany b Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, Baltimore, MD, USA
a r t i c l e
i n f o
Article history: Received 1 April 2014 Received in revised form 8 July 2014 Accepted 7 August 2014 Keywords: Magnetic resonance imaging Dynamic imaging Time-resolved imaging Gadolinium-based contrast agents Liver imaging
a b s t r a c t Purpose: CAIPIRINHA-Dixon-TWIST (CDT)-VIBE is a robust method for abdominal magnetic resonance imaging providing both high spatial and high temporal resolution. The purpose of this study was to examine the influence of different gadolinium based contrast agents (GBCA) on image quality (IQ) with CDT-VIBE. Materials and methods: In this IRB-approved, retrospective, inter-individual comparison study, 86 patients scanned at 3T were included. Within 28 s, 14 high-resolution 3D datasets were acquired using CDT-VIBE. 37 patients received 0.1 mmol/kg gadoterate meglumine, 28 patients 0.1 mmol/kg gadobutrol, and 19 patients 0.1 mL/kg gadoxetic acid. Two blinded, board-certified radiologists assessed the image quality on a 5 point scale, as well as the number of hepatic arterial dominant (HAD) phases. Results: Regardless of the GBCA utilized, CDT-VIBE resulted in good IQ in terms of best IQ achieved among all 14 datasets (gadobutrol 4.3, gadoterate meglumine 3.9, gadoxetic acid 3.7). With respect to worst IQ, the three groups showed statistically significant differences with gadobutrol receiving the highest rating (3.6) and gadoxetic acid the lowest (2.4) (gadoterate meglumine 3.0; 0.0014 < p < 0.0485). No statistically significant differences were found in the mean number of acquired HAD phases (gadobutrol 3.4, gadoterate meglumine 3.9, gadoxetic acid 3.1; 0.18 < p < 0.57). Conclusion: Different gadolinium-based contrast agents can be utilized for dynamic liver imaging with CDT-VIBE resulting in good image quality. © 2014 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Magnetic resonance imaging (MRI) has become the modality of choice for assessment of various upper abdominal pathologies, in particular those involving the liver. Standard MRI protocols consist of dynamic, multiphase, axial T1-weighted contrast enhanced imaging after the injection of a gadolinium-based contrast agent (GBCA). This increases sensitivity and specificity [1–3], most notably in the detection and differentiation of focal liver lesions. High quality arterial or hepatic arterial dominant phases (HAD) improve assessment and diagnosis of many liver lesions such as hepatic cell carcinoma (HCC). For dynamic imaging,
∗ Corresponding author. Tel.: +49 621 383 2067; fax: +49 621 383 3817. E-mail address: [email protected] (J. Budjan). http://dx.doi.org/10.1016/j.ejrad.2014.08.003 0720-048X/© 2014 Elsevier Ireland Ltd. All rights reserved.
gradient-recalled-echo (GRE) sequences using spectral fat saturation and an acceleration technique (SENSE or GRAPPA) play a substantial role in the clinical routine. These are known as THRIVE (T1W High Resolution Isotropic Volume Examination, Philipps), LAVA (Liver Acquisition with Volume Acquisition, GE), or VIBE (Volumetric Interpolated Breath Hold Examination, Siemens) depending on the vendor. An alternative, novel sequence called CAIPIRINHA-Dixon-TWIST (CDT) VIBE has been recently described for contrast-enhanced abdominal imaging [4]. This sequence combines a new parallel imaging reconstruction technique called CAIPIRINHA (Controlled aliasing in parallel imaging results in higher acceleration) [5,6], Dixon fat suppression, and view-sharing techniques (TWIST, time-resolved imaging with interleaved stochastic trajectories). CDT-VIBE allows the acquisition of 14 complete, high resolution 3D-datasets during a single breath-hold (28 s), resulting in a
2008
J. Budjan et al. / European Journal of Radiology 83 (2014) 2007–2012
temporal resolution of one dataset every 2.1 s. Consequently CDTVIBE enables acquisition and selection of the optimal arterial or hepatic arterial dominant phase after the injection of GBCA [4]. CDT-VIBE has been shown to be reliable and robust when utilized with gadoterate meglumine as the contrast agent [4]. Currently approved GBCAs demonstrate different properties with respect to chemical structure, stability, molar concentration, and r1 relaxivity. Besides the conventional, extracellular, nonspecific GBCA, several other indications for organ specific contrast agents are available. For example, gadoxetic acid – a linear GBCA which is partially metabolized by functioning hepatocytes and excreted in the biliary system – can be administered for dedicated hepatic MRI, improving the differentiation of benign and malignant liver lesions [7]. The advent of nephrogenic systemic fibrosis (NSF) and correlation between the incidence of NSF and the use of gadolinium chelates in patients with impaired renal function has drawn scrutiny to the use of such agents [8,9]. As a result, new guidelines have been adopted in the United States and Europe, wherein macrocyclic gadolinium chelates are considered the preferred contrast agent in patients with impaired renal function or on dialysis. The purpose of this retrospective inter-individual study is to assess the recently introduced CDT-VIBE sequence for hepatic imaging in the clinical routine utilizing three different contrast agents. The assessment was performed in terms of image quality and in comparison to a fast VIBE sequence using CAIPIRINHA-Dixon technique. The contrast agents were two macrocyclic GBCAs with different r1 relaxivities and availability at different concentrations, specifically gadoterate meglumine and gadobutrol, and one linear, hepatocyte-specific GBCA, gadoxetic acid. 2. Materials and methods
Table 1 Sequence parameters.
TR/TE/TE2 (ms) Flip angle Spatial resolution (mm3 ) Acceleration factor Field of view (mm × mm) Matrix Number of frames Number of slices TWIST size of central k-space region TWIST sampling density k-space periphery
CDT-VIBE
CD-VIBE
4.1/1.33/2.56 9◦ 1.2 × 1.2 × 3 CAIPIRINHA 4 340 × 255 288 × 162 14 72 20% 20%
3.95/1.27/2.5 9◦ 1.2 × 1.2 × 3 CAIPIRINHA 4 390 × 270 320 × 167 1 72 – –
Note—TR: repetition time; TE: echo time; CDT-VIBE: dynamic CAIPIRINHA-DIXONTWIST-VIBE; CD-VIBE: standard portal venous T1 vibe.
PA) was used to administer gadoterate meglumine and gadobutrol agent through an antecubital venous catheter. Contrast media was injected at a rate of 1.5 mL/s, followed by a 30 mL saline chaser at the same injection rate. As gadoxetic acid is the less frequently administered GBCA at our institution, we utilize prefilled syringes that are injected manually at a rate of approximately 1.5 mL/s. All images were acquired per our institution’s standard liver MR protocol. Among other morphological and functional sequences, this includes a non-enhanced T1-weighted VIBE sequence with Dixon fat separation and CAIPRINHA 4 fold acceleration (CD-VIBE) before contrast administration. After a delay of 5 s post injection, the acquisition of the CDT-VIBE sequence is started. Additional portal venous and more delayed CD-VIBE sequences are acquired at 50 s and 100 s following the contrast media injection, respectively. In patients receiving gadoxetic acid, additional CD-VIBE sequences are acquired 10 and 20 min after contrast media injection. All sequences are acquired in the axial plane. An overview of the sequence parameters is shown in Table 1.
2.1. Patients 2.3. Image analysis The institutional review board (IRB) waived the requirement of informed patient consent in this retrospective study. From September 2012 until July 2013, all patients referred for hepatic MR were included, in which the CDT-VIBE sequence was part of the standard clinical MR protocol (this was only available on one MR scanner). While gadoterate meglumine is the standard contrast media utilized at our institution, patients with suspected HCC or previously uncharacterized liver lesions receive gadoxetic acid. Patients with impaired renal function or a history of adverse reaction after gadoterate meglumine injection receive gadobutrol. Thus, depending on the contrast agent utilized for a given patient’s examination, patients were assigned to the gadoterate meglumine, gadobutrol or gadoxetic acid-group. Indications for the MRI examinations were the characterization or follow-up of liver lesions as well as first-time oncologic staging or follow up examinations for different malignancies. 2.2. Magnetic resonance imaging All examinations were acquired on a 48-channel 3T MR system (MAGNETOM Skyra; Siemens Healthcare Sector, Erlangen, Germany) with an 18-element body matrix coil and the inbuilt 32-element spine matrix coil. The patients were positioned headfirst and supine. The patients received 0.1 mmol/kg body weight of gadoterate meglumine (Dotarem® , Guerbet), 0.1 mmol/kg body weight gadobutrol (Gadovist® [EU] or Gadavist® [US], Bayer HeathCare), or 0.025 mmol/kg body weight gadoxetic acid (Primovist® [EU] or Eovist® [US], Bayer HealthCare). Gadoterate meglumine and gadobutrol were diluted 1:1 with a 0.9% NaCl solution. An automated injector pump (Spectris Solaris EP; Medrad, Indianola,
Two radiologists (12 and 6 years of experience with body magnetic resonance imaging) assessed the CDT-VIBE and portal-venous CD-VIBE images in a consensus reading. The radiologists were unaware of which GBCA was utilized, the final diagnosis, and the clinical history. For this assessment, the fat-separated water-only series alone was evaluated. This series contained all 14 3D-datasets. It was loaded into the 4D viewer of OsiriX 4.0 (OsiriX Foundation, Geneva, Switzerland). The radiologists assessed the best and worst image quality (IQ) present within the 14 data sets obtained for each patient on a 5-point Likert-scale (5, excellent; 4, good image quality; 3, moderate image quality with slightly detrimental artifacts; 2, poor image quality with artifacts yet still diagnostic; and 1, nondiagnostic image quality). The IQ of the CD-VIBE sequence was rated accordingly. Additionally, the number of hepatic arterial dominant phases (HAD), as defined by Semelka et al. [10], was assessed. The presence of ascites was rated on a 3-point scale (0, no ascites; 1, moderate ascites; 2, severe ascites). 2.4. Statistical analysis The statistical analysis was performed using JMP 10.0 (SAS Institute, Cary, NC). Descriptive statistics are given as means and standard deviations if not indicated otherwise. The Shapiro–Wilk test was applied to identify normally distributed data. Data on patient characteristics was compared using univariate analysis of variance (ANOVA) for continuous variables and Chi-square test for categorical variables. Data on IQ, number of HAD phases, and the presence of ascites was examined with the Kruskal–Wallis test with the post hoc Mann–Whitney U test. IQ of CDT-VIBE and CD-VIBE
J. Budjan et al. / European Journal of Radiology 83 (2014) 2007–2012
2009
Table 2 Patient characteristics.
Numbers Age [years] mean range Gender [male/female]
Gadoterate meglumine
Gadobutrol
Gadoxetic acid
Total
37 64 24–92 22m/15f
28 63 27–87 22m/7f
19 62 32–79 11m/8f
84 B3 24–92 54m/30f
p-Value 0.86* 0.34**
Note: * ANOVA. ** Chi-square test.
was compared using the Wilcoxon signed rank test. Two-tailed p-values of <0.05 were considered statistically significant. 3. Results The patient characteristics are shown in Table 2. The groups did not demonstrate statistically significant differences with respect to gender or age. All MRI exams were acquired successfully without any complications (i.e. adverse or significant adverse events). 3.1. Image quality An overview of the image quality (IQ) ratings for the individual contrast media agents is shown in Fig. 1. Overall, the CDT-VIBE sequence provided good image quality regardless of the contrast media agent used (see Fig. 2). Among all examinations, the best image quality achieved throughout the 14 CDT-VIBE datasets received a mean rating of 4.0 ± 0.8, while the worst received a mean rating of 3.1 ± 1.1. In the individual groups, the best IQ achieved in
the datasets received a mean rating of 3.9 ± 0.6 for gadoterate meglumine, 4.3 ± 0.6 for gadobutrol, and 3.7 ± 1.2 for gadoxetic acid. The worst IQ achieved in the datasets received a mean rating of 3.0 ± 0.8 for gadoterate meglumine, 3.6 ± 1.0 for gadobutrol and 2.4 ± 1.2 for gadoxetic acid. As summarized in Table 3, analyses found statistically significant differences between the groups. With respect to mean best IQ, gadobutrol received a higher rating than gadoterate meglumine (p = 0.007). No statistically significant differences regarding best IQ were found between the other groups. With respect to worst IQ, there were statistically significant differences between all three groups, with gadobutrol receiving the highest and gadoxetic acid receiving the lowest mean rating. For the CD-VIBE-Sequence, the groups did not show statistically significant differences (gadoterate meglumine 4.3 ± 0.8; gadobutrol 4.4 ± 0.7; gadoxetic acid 3.9 ± 1.1). Comparing the best IQ of the CDT-VIBE with IQ of the standard portal venous CD-VIBE, the CD-VIBE received a statistically significant higher rating in the gadoterate meglumine group (CD-VIBE 4.3,
Fig. 1. Bars in waterfall plots present the best (left) and worst (right) image quality ratings among the 14 CDT-VIBE 3D datasets for all individual examinations. Image quality is rated on a 5-point Likert-scale, the rating of 3 (moderate image quality with slightly detrimental artifacts) is shown as baseline.
2010
J. Budjan et al. / European Journal of Radiology 83 (2014) 2007–2012
Fig. 2. Sample images for all three contrast media agents showing the first HAD phase (left column) and last set of the CDT-VIBE dataset (center). The standard CD-VIBE portal venous sequence is shown in the right column. The presented images are maximum intensity projections over 3 images. Table 3 IQ and HAD phases per group. Gadoterate meglumine
Best IQ Worst IQ CD-VIBE IQ HAD phases
3.9 3 4.3 3.9
Gadobutrol
4.3 3.6 4.4 3.4
Gadoxetic acid
3.7 2.4 3,9 3.1
p-Values
Gadoterate m. vs. gadobutrol
Gadoterate m. vs. gadoxetic a.
Gadobutrol vs. gadoxetic a.
0.0065 0.0073 0.53 0.57
0.79 0.0485 0.19 0.32
0.11 0.0014 0.08 0.18
Note: p-Values were calculated using Mann–Whitney U test, statistically significant values are bold; IQ: image quality; HAD: hepatic arterial dominant; CD-VIBE: standard portal venous T1 vibe; m.: meglumine; a.: acid.
best IQ CDT-VIBE 3.9, p = 0.006). For gadobutrol and gadoxetic acid, CD-VIBE and CDT-VIBE ratings were not statistically significantly different (p = 0.37 and p = 0.38, respectively).
3.2. Number of hepatic arterial dominant phases The average number of HAD phases in the individual 14 datasets obtained was not statistically significant different regardless of the contrast media utilized. For gadoterate meglumine-enhanced examinations, an average of 3.9 out of the 14 datasets showed a HAD phase vs. 3.4 with gadobutrol and 3.1 with gadoxetic acid.
3.3. Presence of ascites Statistically significant differences were found between the three groups regarding the presence and extent of ascites. The gadoxetic acid-group patients had more severe ascites than those of either the gadobutrol (p = 0.04) or the gadoterate meglumine groups (p = 0.009). Patients receiving gadobutrol- and gadoterate meglumine exhibited a similar prevalence and severity of ascites.
4. Discussion A recently introduced time-resolved approach for contrastenhanced, dynamic liver MR imaging, referred to as CAIPIRINHADixon-TWIST (CDT) –VIBE, combines both high spatial as well as high temporal resolution imaging [4–6]. CDT-VIBE allows the acquisition of 14 complete datasets within 28 s and has demonstrated high quality, reliable imaging in an initial feasibility study using gadoterate meglumine as the contrast agent [4]. The current study shows that this technique provides good image quality with three different gadolinium-based contrast agents (GBCA) for routine clinical imaging. Of both macrocyclic GBCAs evaluated, gadobutrol led to a higher image quality compared to gadoterate meglumine. The statistical superiority of gadobutrol in terms of image quality compared to gadoterate meglumine might be attributed to the higher r1 relaxivity and its availability at a higher concentration of 0.5 mol/L vs. 0.25 mol/L for gadoterate meglumine and 0.25 mol/L for gadoxetic acid. Recent publications have shown that gadobutrol may provide higher image quality in different imaging settings, e.g. in renal MR-angiography [11,12] or brain imaging [13,14]. In other imaging settings like peripheral MRA however, no definite benefit of
J. Budjan et al. / European Journal of Radiology 83 (2014) 2007–2012
2011
Fig. 3. Example images using all three contrast media agents to illustrate the differences in image quality found among the 14 datasets of an individual CDT-VIBE sequence. Best (left column) and worst (center) rated CDT-VIBE datasets as well as the standard CD-VIBE portal venous sequence (right) are depicted. The presented images are maximum intensity projections over 3 images. Image quality of the datasets was rated as 4–2-4 for gadoterate meglumine, 4-2-4 for gadobutrol, and 4-2-5 for gadoxetic acid.
gadobutrol in comparison to gadoterate meglumine has been found [15,16]. Compared to the two other compounds, gadoxetic acid provided the poorest mean image quality. To some extent, this might be attributable to the higher prevalence of ascites in the gadoxetic acid group. While new techniques like fully dynamic transmission with two independent radio frequency transmitters, as is available in the utilized second-generation scanner, help to reduce the imaging artifacts in patients with ascites, it still affects image quality to a certain extent. However, our data shows that even despite this, the CDT-VIBE sequence in the gadoxetic acid group, which featured highest presence and extent of ascites, resulted in good IQ. Not only might ascites itself lead to lower image quality due to artifacts, especially at higher field strengths [17], but also patients with ascites may have been more critically ill than the other patients within the study. Thus, their ability to cooperate with MRI examinations may have been impaired. Due to the resulting increased intra-abdominal pressure with ascites, patients with ascites also likely experience more difficulties completing the requisite breathholds for CDT-VIBE. This is reflected in the datasets with the worst IQ, which often represented datasets at the end of a breath hold (see Fig. 3). When comparing the standard CD-VIBE sequence, which requires a shorter breath hold, IQ ratings between the groups were not statistically significantly different. With arterial-phase imaging of the liver, there is a need for precise timing in order to acquire proper arterial or hepatic arterial dominant phase images. CDT-VIBE is useful in this regard, as multiple images can be acquired even within the first 10–15 s of breath hold. Thus, even though the CDT-VIBE IQ is acceptable but overall slightly lower than CD-VIBE, the probability of acquiring the requisite arterial or HAD phase images is increased with the former. That is, timing is not as critical for the CDT-VIBE as it is for the CD-VIBE. In this particular study, the breath hold for the CD-VIBE is substantially shorter than that of the CDT-VIBE (10 vs. 28 s). In critically ill
patients who have problems complying with the required breath hold, this makes CD-VIBE more practical for imaging in the portal venous phase, where accurate timing is less imperative. Using the CDT-VIBE for dynamic, multi-phase imaging of the upper abdomen, an average of 3–4 high-resolution HAD phase datasets per examination could be acquired regardless of the contrast media utilized. In the detection of small hepatocellular carcinomas (HCC), the use of the hepatocyte-specific contrast agent gadoxetic acid can increase sensitivity and specificity by combining multi-phase dynamic imaging with hepatobiliary phase imaging [18–20]. A precise timing of the acquisition of the HAD phases is important for the detection of small lesions [21]. Depending on factors such as the cardiovascular status of the patient, this can be difficult to achieve using conventional fixed-delay 3-phase approaches [22,23]. In our study, 3 HAD phases per examination were acquired using gadoxetic acid. By increasing the chance of acquiring proper HAD phases with gadoxetic acid, examinations performed with this agent may become more robust. 4.1. Study limitations Some study limitations warrant further discussion. First, due to the clinical setting, only inter-individual and not intra-individual comparisons could be performed. While an intra-individual comparison would be the optimal approach, restrictions in the use of contrast agents in healthy volunteers or in patients make it impractical and potentially unethical given concerns of NSF—especially when comparing more than two contrast agents. In this fundamental question (intra- vs. inter-individual) we chose – in our opinion – the second best way to get a clinical driven statement about the influence of contrast agents on the image quality of a new imaging sequence. Second, the clinical setting allowed the acquisition of a portal-venous CD-VIBE sequence only after the CDT-VIBE sequence rather than a direct comparison
2012
J. Budjan et al. / European Journal of Radiology 83 (2014) 2007–2012
of both techniques. The aforementioned limitations should be further investigated in a separate comparison study focusing on this technical point. Third, even though the patients of all three groups did not differ in regards of age or gender, there were very likely differences in illness severity among groups. As indicated in the preceding discussion, more critically ill patients were likely included in the gadoxetic acid group, thus likely resulting in a diminished IQ in this group due to inability to comply fully with the MR examination. Fourth, to account for the different gadolinium concentrations, previous studies comparing different GBCA altered the injection rates to achieve equivalent gadolinium delivery rates regardless of the different molar concentrations [15]. As the examinations in the current study were performed as part of the clinical routine, we used a fixed injection rate for gadoterate meglumine and gadobutrol, while gadoxetic acid was injected manually at approximately the same rate. 5. Conclusion In general, CDT-VIBE is a robust sequence for routine abdominal imaging in various settings. Both macrocyclic GBCAs examined in this study – the newly FDA-approved macrocyclic gadoterate meglumine and the high-molar gadobutrol – as well as the hepatocyte-specific GBCA gadoxetic acid result in good IQ. Limitations in IQ using CDT-VIBE seem to be patient-specific (i.e. related to incomplete breath hold, ascites) rather than related to differences in specific GBCAs. References [1] Mori K, Yoshioka H, Takahashi N, et al. Triple arterial phase dynamic MRI with sensitivity encoding for hypervascular hepatocellular carcinoma: comparison of the diagnostic accuracy among the early, middle, late, and whole triple arterial phase imaging. AJR Am J Roentgenol 2005;184(1):63–9. [2] Brodsky EK, Bultman EM, Johnson KM, et al. High-spatial and hightemporal resolution dynamic contrast-enhanced perfusion imaging of the liver with time-resolved three-dimensional radial MRI. Magn Reson Med 2013;71(3):934–41 (official journal of the Society of Magnetic Resonance in Medicine/Society of Magnetic Resonance in Medicine). [3] Yu JS, Kim KW, Kim EK, Lee JT, Yoo HS. Contrast enhancement of small hepatocellular carcinoma: usefulness of three successive early image acquisitions during multiphase dynamic MR imaging. AJR Am J Roentgenol 1999;173(3):597–604. [4] Michaely HJ, Morelli JN, Budjan J, et al. CAIPIRINHA-Dixon-TWIST (CDT)volume-interpolated breath-hold examination (VIBE): a new technique for fast time-resolved dynamic 3-dimensional imaging of the abdomen with high spatial resolution. Invest Radiol 2013;48(8):590–7. [5] Breuer FA, Blaimer M, Heidemann RM, Mueller MF, Griswold MA, Jakob PM. Controlled aliasing in parallel imaging results in higher acceleration (CAIPIRINHA) for multi-slice imaging. Magn Reson Med 2005;53(3):684–91 (official journal of the Society of Magnetic Resonance in Medicine/Society of Magnetic Resonance in Medicine).
[6] Breuer FA, Blaimer M, Mueller MF, et al. Controlled aliasing in volumetric parallel imaging (2D CAIPIRINHA). Magn Reson Med 2006;55(3):549–56 (official journal of the Society of Magnetic Resonance in Medicine/Society of Magnetic Resonance in Medicine). [7] Zech CJ, Herrmann KA, Reiser MF, Schoenberg SO. MR imaging in patients with suspected liver metastases: value of liver-specific contrast agent Gd-EOBDTPA. Magn Reson Med Sci 2007;6(1):43–52 (MRMS: an official journal of Japan Society of Magnetic Resonance in Medicine). [8] Grobner T. Gadolinium—a specific trigger for the development of nephrogenic fibrosing dermopathy and nephrogenic systemic fibrosis? Nephrol Dial Transplantation 2006;21(4):1104–8 (official publication of the European Dialysis and Transplant Association—European Renal Association). [9] Kuo PH, Kanal E, Abu-Alfa AK, Cowper SE, Gadolinium-based MR. contrast agents and nephrogenic systemic fibrosis. Radiology 2007;242(3):647–9. [10] Semelka RC, Martin DR, Balci NC. Magnetic resonance imaging of the liver: how I do it. J Gastroenterol Hepatol 2006;21(4):632–7. [11] Morelli JN, Runge VM, Ai F, et al. Magnetic resonance evaluation of renal artery stenosis in a swine model: performance of low-dose gadobutrol versus gadoterate meglumine in comparison with digital subtraction intra-arterial catheter angiography. Invest Radiol 2012;47(6):376–82. [12] Hadizadeh DR, Von Falkenhausen M, Kukuk GM, et al. Contrast material for abdominal dynamic contrast-enhanced 3D MR angiography with parallel imaging: intraindividual equimolar comparison of a macrocyclic 1.0 M gadolinium chelate and a linear ionic 0.5 M gadolinium chelate. AJR Am J Roentgenol 2010;194(3):821–9. [13] Kramer H, Runge VM, Naul LG, Loynachan AT, Reiser MF, Wintersperger BJ. Brain MRI with single-dose (0.1 mmol/kg) gadobutrol at 1.5 T and 3 T: comparison with 0.15 mmol/kg gadoterate meglumine. AJR Am J Roentgenol 2010;194(5):1337–42. [14] Kim ES, Chang JH, Choi HS, Kim J, Lee SK. Diagnostic yield of double-dose gadobutrol in the detection of brain metastasis: intraindividual comparison with double-dose gadopentetate dimeglumine. AJNR Am J Neuroradiol 2010;31(6):1055–8. [15] Haneder S, Attenberger UI, Schoenberg SO, Loewe C, Arnaiz J, Michaely HJ. Comparison of 0.5 M gadoterate and 1.0 M gadobutrol in peripheral MRA: a prospective, single-center, randomized, crossover, double-blind study. J Magn Reson Imaging: JMRI 2012;36(5):1213–21. [16] Szucs-Farkas Z, Froehlich JM, Ulrich M, et al. 1.0-M gadobutrol versus 0.5-M gadoterate for peripheral magnetic resonance angiography: a prospective randomized controlled clinical trial. J Magn Reson Imaging: JMRI 2008;27(6):1399–405. [17] Merkle EM, Dale BM. Abdominal MRI at 3.0 T: the basics revisited. AJR Am J Roentgenol 2006;186(6):1524–32. [18] Zech CJ, Bartolozzi C, Bioulac-Sage P, et al. Consensus report of the Fifth International Forum for Liver MRI. AJR Am J Roentgenol 2013;201(1):97–107. [19] Sun HY, Lee JM, Shin CI, et al. Gadoxetic acid-enhanced magnetic resonance imaging for differentiating small hepatocellular carcinomas (< or =2 cm in diameter) from arterial enhancing pseudolesions: special emphasis on hepatobiliary phase imaging. Invest Radiol 2010;45(2):96–103. [20] Purysko AS, Remer EM, Veniero JC. Focal liver lesion detection and characterization with GD-EOB-DTPA. Clin Radiol 2011;66(7):673–84. [21] Kanematsu M, Semelka RC, Matsuo M, et al. Gadolinium-enhanced MR imaging of the liver: optimizing imaging delay for hepatic arterial and portal venous phases—a prospective randomized study in patients with chronic liver damage. Radiology 2002;225(2):407–15. [22] Materne R, Horsmans Y, Jamart J, Smith AM, Gigot JF, Van Beers BE. Gadolinium-enhanced arterial-phase MR imaging of hypervascular liver tumors: comparison between tailored and fixed scanning delays in the same patients. J Magn Reson Imaging: JMRI 2000;11(3):244–9. [23] Campos JT, Sirlin CB, Choi JY. Focal hepatic lesions in Gd-EOB-DTPA enhanced MRI: the atlas. Insights Imaging 2012;3(5):451–74.
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
CAIPIRINHA-Dixon-TWIST (CDT)-volume-interpolated breath-hold examination (VIBE) for dynamic liver imaging: Comparison of gadoterate meglumine, gadobutrol and gadoxetic acid Johannes Budjan a,∗ , Melissa Ong a , Philipp Riffel a , John N. Morelli b , Henrik J. Michaely a , Stefan O. Schoenberg a , Stefan Haneder a a Institute of Clinical Radiology and Nuclear Medicine, University Medical Center Mannheim, Medical Faculty Mannheim—Heidelberg University, Mannheim, Germany b Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, Baltimore, MD, USA
a r t i c l e
i n f o
Article history: Received 1 April 2014 Received in revised form 8 July 2014 Accepted 7 August 2014 Keywords: Magnetic resonance imaging Dynamic imaging Time-resolved imaging Gadolinium-based contrast agents Liver imaging
a b s t r a c t Purpose: CAIPIRINHA-Dixon-TWIST (CDT)-VIBE is a robust method for abdominal magnetic resonance imaging providing both high spatial and high temporal resolution. The purpose of this study was to examine the influence of different gadolinium based contrast agents (GBCA) on image quality (IQ) with CDT-VIBE. Materials and methods: In this IRB-approved, retrospective, inter-individual comparison study, 86 patients scanned at 3T were included. Within 28 s, 14 high-resolution 3D datasets were acquired using CDT-VIBE. 37 patients received 0.1 mmol/kg gadoterate meglumine, 28 patients 0.1 mmol/kg gadobutrol, and 19 patients 0.1 mL/kg gadoxetic acid. Two blinded, board-certified radiologists assessed the image quality on a 5 point scale, as well as the number of hepatic arterial dominant (HAD) phases. Results: Regardless of the GBCA utilized, CDT-VIBE resulted in good IQ in terms of best IQ achieved among all 14 datasets (gadobutrol 4.3, gadoterate meglumine 3.9, gadoxetic acid 3.7). With respect to worst IQ, the three groups showed statistically significant differences with gadobutrol receiving the highest rating (3.6) and gadoxetic acid the lowest (2.4) (gadoterate meglumine 3.0; 0.0014 < p < 0.0485). No statistically significant differences were found in the mean number of acquired HAD phases (gadobutrol 3.4, gadoterate meglumine 3.9, gadoxetic acid 3.1; 0.18 < p < 0.57). Conclusion: Different gadolinium-based contrast agents can be utilized for dynamic liver imaging with CDT-VIBE resulting in good image quality. © 2014 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Magnetic resonance imaging (MRI) has become the modality of choice for assessment of various upper abdominal pathologies, in particular those involving the liver. Standard MRI protocols consist of dynamic, multiphase, axial T1-weighted contrast enhanced imaging after the injection of a gadolinium-based contrast agent (GBCA). This increases sensitivity and specificity [1–3], most notably in the detection and differentiation of focal liver lesions. High quality arterial or hepatic arterial dominant phases (HAD) improve assessment and diagnosis of many liver lesions such as hepatic cell carcinoma (HCC). For dynamic imaging,
∗ Corresponding author. Tel.: +49 621 383 2067; fax: +49 621 383 3817. E-mail address: [email protected] (J. Budjan). http://dx.doi.org/10.1016/j.ejrad.2014.08.003 0720-048X/© 2014 Elsevier Ireland Ltd. All rights reserved.
gradient-recalled-echo (GRE) sequences using spectral fat saturation and an acceleration technique (SENSE or GRAPPA) play a substantial role in the clinical routine. These are known as THRIVE (T1W High Resolution Isotropic Volume Examination, Philipps), LAVA (Liver Acquisition with Volume Acquisition, GE), or VIBE (Volumetric Interpolated Breath Hold Examination, Siemens) depending on the vendor. An alternative, novel sequence called CAIPIRINHA-Dixon-TWIST (CDT) VIBE has been recently described for contrast-enhanced abdominal imaging [4]. This sequence combines a new parallel imaging reconstruction technique called CAIPIRINHA (Controlled aliasing in parallel imaging results in higher acceleration) [5,6], Dixon fat suppression, and view-sharing techniques (TWIST, time-resolved imaging with interleaved stochastic trajectories). CDT-VIBE allows the acquisition of 14 complete, high resolution 3D-datasets during a single breath-hold (28 s), resulting in a
2008
J. Budjan et al. / European Journal of Radiology 83 (2014) 2007–2012
temporal resolution of one dataset every 2.1 s. Consequently CDTVIBE enables acquisition and selection of the optimal arterial or hepatic arterial dominant phase after the injection of GBCA [4]. CDT-VIBE has been shown to be reliable and robust when utilized with gadoterate meglumine as the contrast agent [4]. Currently approved GBCAs demonstrate different properties with respect to chemical structure, stability, molar concentration, and r1 relaxivity. Besides the conventional, extracellular, nonspecific GBCA, several other indications for organ specific contrast agents are available. For example, gadoxetic acid – a linear GBCA which is partially metabolized by functioning hepatocytes and excreted in the biliary system – can be administered for dedicated hepatic MRI, improving the differentiation of benign and malignant liver lesions [7]. The advent of nephrogenic systemic fibrosis (NSF) and correlation between the incidence of NSF and the use of gadolinium chelates in patients with impaired renal function has drawn scrutiny to the use of such agents [8,9]. As a result, new guidelines have been adopted in the United States and Europe, wherein macrocyclic gadolinium chelates are considered the preferred contrast agent in patients with impaired renal function or on dialysis. The purpose of this retrospective inter-individual study is to assess the recently introduced CDT-VIBE sequence for hepatic imaging in the clinical routine utilizing three different contrast agents. The assessment was performed in terms of image quality and in comparison to a fast VIBE sequence using CAIPIRINHA-Dixon technique. The contrast agents were two macrocyclic GBCAs with different r1 relaxivities and availability at different concentrations, specifically gadoterate meglumine and gadobutrol, and one linear, hepatocyte-specific GBCA, gadoxetic acid. 2. Materials and methods
Table 1 Sequence parameters.
TR/TE/TE2 (ms) Flip angle Spatial resolution (mm3 ) Acceleration factor Field of view (mm × mm) Matrix Number of frames Number of slices TWIST size of central k-space region TWIST sampling density k-space periphery
CDT-VIBE
CD-VIBE
4.1/1.33/2.56 9◦ 1.2 × 1.2 × 3 CAIPIRINHA 4 340 × 255 288 × 162 14 72 20% 20%
3.95/1.27/2.5 9◦ 1.2 × 1.2 × 3 CAIPIRINHA 4 390 × 270 320 × 167 1 72 – –
Note—TR: repetition time; TE: echo time; CDT-VIBE: dynamic CAIPIRINHA-DIXONTWIST-VIBE; CD-VIBE: standard portal venous T1 vibe.
PA) was used to administer gadoterate meglumine and gadobutrol agent through an antecubital venous catheter. Contrast media was injected at a rate of 1.5 mL/s, followed by a 30 mL saline chaser at the same injection rate. As gadoxetic acid is the less frequently administered GBCA at our institution, we utilize prefilled syringes that are injected manually at a rate of approximately 1.5 mL/s. All images were acquired per our institution’s standard liver MR protocol. Among other morphological and functional sequences, this includes a non-enhanced T1-weighted VIBE sequence with Dixon fat separation and CAIPRINHA 4 fold acceleration (CD-VIBE) before contrast administration. After a delay of 5 s post injection, the acquisition of the CDT-VIBE sequence is started. Additional portal venous and more delayed CD-VIBE sequences are acquired at 50 s and 100 s following the contrast media injection, respectively. In patients receiving gadoxetic acid, additional CD-VIBE sequences are acquired 10 and 20 min after contrast media injection. All sequences are acquired in the axial plane. An overview of the sequence parameters is shown in Table 1.
2.1. Patients 2.3. Image analysis The institutional review board (IRB) waived the requirement of informed patient consent in this retrospective study. From September 2012 until July 2013, all patients referred for hepatic MR were included, in which the CDT-VIBE sequence was part of the standard clinical MR protocol (this was only available on one MR scanner). While gadoterate meglumine is the standard contrast media utilized at our institution, patients with suspected HCC or previously uncharacterized liver lesions receive gadoxetic acid. Patients with impaired renal function or a history of adverse reaction after gadoterate meglumine injection receive gadobutrol. Thus, depending on the contrast agent utilized for a given patient’s examination, patients were assigned to the gadoterate meglumine, gadobutrol or gadoxetic acid-group. Indications for the MRI examinations were the characterization or follow-up of liver lesions as well as first-time oncologic staging or follow up examinations for different malignancies. 2.2. Magnetic resonance imaging All examinations were acquired on a 48-channel 3T MR system (MAGNETOM Skyra; Siemens Healthcare Sector, Erlangen, Germany) with an 18-element body matrix coil and the inbuilt 32-element spine matrix coil. The patients were positioned headfirst and supine. The patients received 0.1 mmol/kg body weight of gadoterate meglumine (Dotarem® , Guerbet), 0.1 mmol/kg body weight gadobutrol (Gadovist® [EU] or Gadavist® [US], Bayer HeathCare), or 0.025 mmol/kg body weight gadoxetic acid (Primovist® [EU] or Eovist® [US], Bayer HealthCare). Gadoterate meglumine and gadobutrol were diluted 1:1 with a 0.9% NaCl solution. An automated injector pump (Spectris Solaris EP; Medrad, Indianola,
Two radiologists (12 and 6 years of experience with body magnetic resonance imaging) assessed the CDT-VIBE and portal-venous CD-VIBE images in a consensus reading. The radiologists were unaware of which GBCA was utilized, the final diagnosis, and the clinical history. For this assessment, the fat-separated water-only series alone was evaluated. This series contained all 14 3D-datasets. It was loaded into the 4D viewer of OsiriX 4.0 (OsiriX Foundation, Geneva, Switzerland). The radiologists assessed the best and worst image quality (IQ) present within the 14 data sets obtained for each patient on a 5-point Likert-scale (5, excellent; 4, good image quality; 3, moderate image quality with slightly detrimental artifacts; 2, poor image quality with artifacts yet still diagnostic; and 1, nondiagnostic image quality). The IQ of the CD-VIBE sequence was rated accordingly. Additionally, the number of hepatic arterial dominant phases (HAD), as defined by Semelka et al. [10], was assessed. The presence of ascites was rated on a 3-point scale (0, no ascites; 1, moderate ascites; 2, severe ascites). 2.4. Statistical analysis The statistical analysis was performed using JMP 10.0 (SAS Institute, Cary, NC). Descriptive statistics are given as means and standard deviations if not indicated otherwise. The Shapiro–Wilk test was applied to identify normally distributed data. Data on patient characteristics was compared using univariate analysis of variance (ANOVA) for continuous variables and Chi-square test for categorical variables. Data on IQ, number of HAD phases, and the presence of ascites was examined with the Kruskal–Wallis test with the post hoc Mann–Whitney U test. IQ of CDT-VIBE and CD-VIBE
J. Budjan et al. / European Journal of Radiology 83 (2014) 2007–2012
2009
Table 2 Patient characteristics.
Numbers Age [years] mean range Gender [male/female]
Gadoterate meglumine
Gadobutrol
Gadoxetic acid
Total
37 64 24–92 22m/15f
28 63 27–87 22m/7f
19 62 32–79 11m/8f
84 B3 24–92 54m/30f
p-Value 0.86* 0.34**
Note: * ANOVA. ** Chi-square test.
was compared using the Wilcoxon signed rank test. Two-tailed p-values of <0.05 were considered statistically significant. 3. Results The patient characteristics are shown in Table 2. The groups did not demonstrate statistically significant differences with respect to gender or age. All MRI exams were acquired successfully without any complications (i.e. adverse or significant adverse events). 3.1. Image quality An overview of the image quality (IQ) ratings for the individual contrast media agents is shown in Fig. 1. Overall, the CDT-VIBE sequence provided good image quality regardless of the contrast media agent used (see Fig. 2). Among all examinations, the best image quality achieved throughout the 14 CDT-VIBE datasets received a mean rating of 4.0 ± 0.8, while the worst received a mean rating of 3.1 ± 1.1. In the individual groups, the best IQ achieved in
the datasets received a mean rating of 3.9 ± 0.6 for gadoterate meglumine, 4.3 ± 0.6 for gadobutrol, and 3.7 ± 1.2 for gadoxetic acid. The worst IQ achieved in the datasets received a mean rating of 3.0 ± 0.8 for gadoterate meglumine, 3.6 ± 1.0 for gadobutrol and 2.4 ± 1.2 for gadoxetic acid. As summarized in Table 3, analyses found statistically significant differences between the groups. With respect to mean best IQ, gadobutrol received a higher rating than gadoterate meglumine (p = 0.007). No statistically significant differences regarding best IQ were found between the other groups. With respect to worst IQ, there were statistically significant differences between all three groups, with gadobutrol receiving the highest and gadoxetic acid receiving the lowest mean rating. For the CD-VIBE-Sequence, the groups did not show statistically significant differences (gadoterate meglumine 4.3 ± 0.8; gadobutrol 4.4 ± 0.7; gadoxetic acid 3.9 ± 1.1). Comparing the best IQ of the CDT-VIBE with IQ of the standard portal venous CD-VIBE, the CD-VIBE received a statistically significant higher rating in the gadoterate meglumine group (CD-VIBE 4.3,
Fig. 1. Bars in waterfall plots present the best (left) and worst (right) image quality ratings among the 14 CDT-VIBE 3D datasets for all individual examinations. Image quality is rated on a 5-point Likert-scale, the rating of 3 (moderate image quality with slightly detrimental artifacts) is shown as baseline.
2010
J. Budjan et al. / European Journal of Radiology 83 (2014) 2007–2012
Fig. 2. Sample images for all three contrast media agents showing the first HAD phase (left column) and last set of the CDT-VIBE dataset (center). The standard CD-VIBE portal venous sequence is shown in the right column. The presented images are maximum intensity projections over 3 images. Table 3 IQ and HAD phases per group. Gadoterate meglumine
Best IQ Worst IQ CD-VIBE IQ HAD phases
3.9 3 4.3 3.9
Gadobutrol
4.3 3.6 4.4 3.4
Gadoxetic acid
3.7 2.4 3,9 3.1
p-Values
Gadoterate m. vs. gadobutrol
Gadoterate m. vs. gadoxetic a.
Gadobutrol vs. gadoxetic a.
0.0065 0.0073 0.53 0.57
0.79 0.0485 0.19 0.32
0.11 0.0014 0.08 0.18
Note: p-Values were calculated using Mann–Whitney U test, statistically significant values are bold; IQ: image quality; HAD: hepatic arterial dominant; CD-VIBE: standard portal venous T1 vibe; m.: meglumine; a.: acid.
best IQ CDT-VIBE 3.9, p = 0.006). For gadobutrol and gadoxetic acid, CD-VIBE and CDT-VIBE ratings were not statistically significantly different (p = 0.37 and p = 0.38, respectively).
3.2. Number of hepatic arterial dominant phases The average number of HAD phases in the individual 14 datasets obtained was not statistically significant different regardless of the contrast media utilized. For gadoterate meglumine-enhanced examinations, an average of 3.9 out of the 14 datasets showed a HAD phase vs. 3.4 with gadobutrol and 3.1 with gadoxetic acid.
3.3. Presence of ascites Statistically significant differences were found between the three groups regarding the presence and extent of ascites. The gadoxetic acid-group patients had more severe ascites than those of either the gadobutrol (p = 0.04) or the gadoterate meglumine groups (p = 0.009). Patients receiving gadobutrol- and gadoterate meglumine exhibited a similar prevalence and severity of ascites.
4. Discussion A recently introduced time-resolved approach for contrastenhanced, dynamic liver MR imaging, referred to as CAIPIRINHADixon-TWIST (CDT) –VIBE, combines both high spatial as well as high temporal resolution imaging [4–6]. CDT-VIBE allows the acquisition of 14 complete datasets within 28 s and has demonstrated high quality, reliable imaging in an initial feasibility study using gadoterate meglumine as the contrast agent [4]. The current study shows that this technique provides good image quality with three different gadolinium-based contrast agents (GBCA) for routine clinical imaging. Of both macrocyclic GBCAs evaluated, gadobutrol led to a higher image quality compared to gadoterate meglumine. The statistical superiority of gadobutrol in terms of image quality compared to gadoterate meglumine might be attributed to the higher r1 relaxivity and its availability at a higher concentration of 0.5 mol/L vs. 0.25 mol/L for gadoterate meglumine and 0.25 mol/L for gadoxetic acid. Recent publications have shown that gadobutrol may provide higher image quality in different imaging settings, e.g. in renal MR-angiography [11,12] or brain imaging [13,14]. In other imaging settings like peripheral MRA however, no definite benefit of
J. Budjan et al. / European Journal of Radiology 83 (2014) 2007–2012
2011
Fig. 3. Example images using all three contrast media agents to illustrate the differences in image quality found among the 14 datasets of an individual CDT-VIBE sequence. Best (left column) and worst (center) rated CDT-VIBE datasets as well as the standard CD-VIBE portal venous sequence (right) are depicted. The presented images are maximum intensity projections over 3 images. Image quality of the datasets was rated as 4–2-4 for gadoterate meglumine, 4-2-4 for gadobutrol, and 4-2-5 for gadoxetic acid.
gadobutrol in comparison to gadoterate meglumine has been found [15,16]. Compared to the two other compounds, gadoxetic acid provided the poorest mean image quality. To some extent, this might be attributable to the higher prevalence of ascites in the gadoxetic acid group. While new techniques like fully dynamic transmission with two independent radio frequency transmitters, as is available in the utilized second-generation scanner, help to reduce the imaging artifacts in patients with ascites, it still affects image quality to a certain extent. However, our data shows that even despite this, the CDT-VIBE sequence in the gadoxetic acid group, which featured highest presence and extent of ascites, resulted in good IQ. Not only might ascites itself lead to lower image quality due to artifacts, especially at higher field strengths [17], but also patients with ascites may have been more critically ill than the other patients within the study. Thus, their ability to cooperate with MRI examinations may have been impaired. Due to the resulting increased intra-abdominal pressure with ascites, patients with ascites also likely experience more difficulties completing the requisite breathholds for CDT-VIBE. This is reflected in the datasets with the worst IQ, which often represented datasets at the end of a breath hold (see Fig. 3). When comparing the standard CD-VIBE sequence, which requires a shorter breath hold, IQ ratings between the groups were not statistically significantly different. With arterial-phase imaging of the liver, there is a need for precise timing in order to acquire proper arterial or hepatic arterial dominant phase images. CDT-VIBE is useful in this regard, as multiple images can be acquired even within the first 10–15 s of breath hold. Thus, even though the CDT-VIBE IQ is acceptable but overall slightly lower than CD-VIBE, the probability of acquiring the requisite arterial or HAD phase images is increased with the former. That is, timing is not as critical for the CDT-VIBE as it is for the CD-VIBE. In this particular study, the breath hold for the CD-VIBE is substantially shorter than that of the CDT-VIBE (10 vs. 28 s). In critically ill
patients who have problems complying with the required breath hold, this makes CD-VIBE more practical for imaging in the portal venous phase, where accurate timing is less imperative. Using the CDT-VIBE for dynamic, multi-phase imaging of the upper abdomen, an average of 3–4 high-resolution HAD phase datasets per examination could be acquired regardless of the contrast media utilized. In the detection of small hepatocellular carcinomas (HCC), the use of the hepatocyte-specific contrast agent gadoxetic acid can increase sensitivity and specificity by combining multi-phase dynamic imaging with hepatobiliary phase imaging [18–20]. A precise timing of the acquisition of the HAD phases is important for the detection of small lesions [21]. Depending on factors such as the cardiovascular status of the patient, this can be difficult to achieve using conventional fixed-delay 3-phase approaches [22,23]. In our study, 3 HAD phases per examination were acquired using gadoxetic acid. By increasing the chance of acquiring proper HAD phases with gadoxetic acid, examinations performed with this agent may become more robust. 4.1. Study limitations Some study limitations warrant further discussion. First, due to the clinical setting, only inter-individual and not intra-individual comparisons could be performed. While an intra-individual comparison would be the optimal approach, restrictions in the use of contrast agents in healthy volunteers or in patients make it impractical and potentially unethical given concerns of NSF—especially when comparing more than two contrast agents. In this fundamental question (intra- vs. inter-individual) we chose – in our opinion – the second best way to get a clinical driven statement about the influence of contrast agents on the image quality of a new imaging sequence. Second, the clinical setting allowed the acquisition of a portal-venous CD-VIBE sequence only after the CDT-VIBE sequence rather than a direct comparison
2012
J. Budjan et al. / European Journal of Radiology 83 (2014) 2007–2012
of both techniques. The aforementioned limitations should be further investigated in a separate comparison study focusing on this technical point. Third, even though the patients of all three groups did not differ in regards of age or gender, there were very likely differences in illness severity among groups. As indicated in the preceding discussion, more critically ill patients were likely included in the gadoxetic acid group, thus likely resulting in a diminished IQ in this group due to inability to comply fully with the MR examination. Fourth, to account for the different gadolinium concentrations, previous studies comparing different GBCA altered the injection rates to achieve equivalent gadolinium delivery rates regardless of the different molar concentrations [15]. As the examinations in the current study were performed as part of the clinical routine, we used a fixed injection rate for gadoterate meglumine and gadobutrol, while gadoxetic acid was injected manually at approximately the same rate. 5. Conclusion In general, CDT-VIBE is a robust sequence for routine abdominal imaging in various settings. Both macrocyclic GBCAs examined in this study – the newly FDA-approved macrocyclic gadoterate meglumine and the high-molar gadobutrol – as well as the hepatocyte-specific GBCA gadoxetic acid result in good IQ. Limitations in IQ using CDT-VIBE seem to be patient-specific (i.e. related to incomplete breath hold, ascites) rather than related to differences in specific GBCAs. References [1] Mori K, Yoshioka H, Takahashi N, et al. Triple arterial phase dynamic MRI with sensitivity encoding for hypervascular hepatocellular carcinoma: comparison of the diagnostic accuracy among the early, middle, late, and whole triple arterial phase imaging. AJR Am J Roentgenol 2005;184(1):63–9. [2] Brodsky EK, Bultman EM, Johnson KM, et al. High-spatial and hightemporal resolution dynamic contrast-enhanced perfusion imaging of the liver with time-resolved three-dimensional radial MRI. Magn Reson Med 2013;71(3):934–41 (official journal of the Society of Magnetic Resonance in Medicine/Society of Magnetic Resonance in Medicine). [3] Yu JS, Kim KW, Kim EK, Lee JT, Yoo HS. Contrast enhancement of small hepatocellular carcinoma: usefulness of three successive early image acquisitions during multiphase dynamic MR imaging. AJR Am J Roentgenol 1999;173(3):597–604. [4] Michaely HJ, Morelli JN, Budjan J, et al. CAIPIRINHA-Dixon-TWIST (CDT)volume-interpolated breath-hold examination (VIBE): a new technique for fast time-resolved dynamic 3-dimensional imaging of the abdomen with high spatial resolution. Invest Radiol 2013;48(8):590–7. [5] Breuer FA, Blaimer M, Heidemann RM, Mueller MF, Griswold MA, Jakob PM. Controlled aliasing in parallel imaging results in higher acceleration (CAIPIRINHA) for multi-slice imaging. Magn Reson Med 2005;53(3):684–91 (official journal of the Society of Magnetic Resonance in Medicine/Society of Magnetic Resonance in Medicine).
[6] Breuer FA, Blaimer M, Mueller MF, et al. Controlled aliasing in volumetric parallel imaging (2D CAIPIRINHA). Magn Reson Med 2006;55(3):549–56 (official journal of the Society of Magnetic Resonance in Medicine/Society of Magnetic Resonance in Medicine). [7] Zech CJ, Herrmann KA, Reiser MF, Schoenberg SO. MR imaging in patients with suspected liver metastases: value of liver-specific contrast agent Gd-EOBDTPA. Magn Reson Med Sci 2007;6(1):43–52 (MRMS: an official journal of Japan Society of Magnetic Resonance in Medicine). [8] Grobner T. Gadolinium—a specific trigger for the development of nephrogenic fibrosing dermopathy and nephrogenic systemic fibrosis? Nephrol Dial Transplantation 2006;21(4):1104–8 (official publication of the European Dialysis and Transplant Association—European Renal Association). [9] Kuo PH, Kanal E, Abu-Alfa AK, Cowper SE, Gadolinium-based MR. contrast agents and nephrogenic systemic fibrosis. Radiology 2007;242(3):647–9. [10] Semelka RC, Martin DR, Balci NC. Magnetic resonance imaging of the liver: how I do it. J Gastroenterol Hepatol 2006;21(4):632–7. [11] Morelli JN, Runge VM, Ai F, et al. Magnetic resonance evaluation of renal artery stenosis in a swine model: performance of low-dose gadobutrol versus gadoterate meglumine in comparison with digital subtraction intra-arterial catheter angiography. Invest Radiol 2012;47(6):376–82. [12] Hadizadeh DR, Von Falkenhausen M, Kukuk GM, et al. Contrast material for abdominal dynamic contrast-enhanced 3D MR angiography with parallel imaging: intraindividual equimolar comparison of a macrocyclic 1.0 M gadolinium chelate and a linear ionic 0.5 M gadolinium chelate. AJR Am J Roentgenol 2010;194(3):821–9. [13] Kramer H, Runge VM, Naul LG, Loynachan AT, Reiser MF, Wintersperger BJ. Brain MRI with single-dose (0.1 mmol/kg) gadobutrol at 1.5 T and 3 T: comparison with 0.15 mmol/kg gadoterate meglumine. AJR Am J Roentgenol 2010;194(5):1337–42. [14] Kim ES, Chang JH, Choi HS, Kim J, Lee SK. Diagnostic yield of double-dose gadobutrol in the detection of brain metastasis: intraindividual comparison with double-dose gadopentetate dimeglumine. AJNR Am J Neuroradiol 2010;31(6):1055–8. [15] Haneder S, Attenberger UI, Schoenberg SO, Loewe C, Arnaiz J, Michaely HJ. Comparison of 0.5 M gadoterate and 1.0 M gadobutrol in peripheral MRA: a prospective, single-center, randomized, crossover, double-blind study. J Magn Reson Imaging: JMRI 2012;36(5):1213–21. [16] Szucs-Farkas Z, Froehlich JM, Ulrich M, et al. 1.0-M gadobutrol versus 0.5-M gadoterate for peripheral magnetic resonance angiography: a prospective randomized controlled clinical trial. J Magn Reson Imaging: JMRI 2008;27(6):1399–405. [17] Merkle EM, Dale BM. Abdominal MRI at 3.0 T: the basics revisited. AJR Am J Roentgenol 2006;186(6):1524–32. [18] Zech CJ, Bartolozzi C, Bioulac-Sage P, et al. Consensus report of the Fifth International Forum for Liver MRI. AJR Am J Roentgenol 2013;201(1):97–107. [19] Sun HY, Lee JM, Shin CI, et al. Gadoxetic acid-enhanced magnetic resonance imaging for differentiating small hepatocellular carcinomas (< or =2 cm in diameter) from arterial enhancing pseudolesions: special emphasis on hepatobiliary phase imaging. Invest Radiol 2010;45(2):96–103. [20] Purysko AS, Remer EM, Veniero JC. Focal liver lesion detection and characterization with GD-EOB-DTPA. Clin Radiol 2011;66(7):673–84. [21] Kanematsu M, Semelka RC, Matsuo M, et al. Gadolinium-enhanced MR imaging of the liver: optimizing imaging delay for hepatic arterial and portal venous phases—a prospective randomized study in patients with chronic liver damage. Radiology 2002;225(2):407–15. [22] Materne R, Horsmans Y, Jamart J, Smith AM, Gigot JF, Van Beers BE. Gadolinium-enhanced arterial-phase MR imaging of hypervascular liver tumors: comparison between tailored and fixed scanning delays in the same patients. J Magn Reson Imaging: JMRI 2000;11(3):244–9. [23] Campos JT, Sirlin CB, Choi JY. Focal hepatic lesions in Gd-EOB-DTPA enhanced MRI: the atlas. Insights Imaging 2012;3(5):451–74.