Gadolinium based contrast agents (GBCAs) are T1-shortening agents that result in marked elevation of signal on T1-weighted (T1W) images. GBCA enhancement is crucial in the detection and characterization of abdominal diseases.
It provides at least two additional imaging properties that facilitate diagnosis: the pattern of blood delivery (i.e., capillary enhancement) and the size and/or rapidity of drainage of the interstitial space (i.e., interstitial enhancement) [1,2].
This review describes important aspects for the achievement of high-quality diagnostic GBCA enhanced images and discusses critical issues related to their optimized use.

MOTION-FREE IMAGE QUALITY

The first important aspect to a confident analysis of GBCA enhancement is to obtain high quality and reproducible MR studies. In keeping with this, we need motion-free images, devoid of substantial artifacts, especially those due to motion. Respiration is the most important and most troubling source of artifacts in abdominal MR imaging. Control of these artifacts is most critical in GBCA enhanced studies in which dynamic T1W imaging is employed. Motion can also adversely affect fat suppression, reducing difference in signal intensities between enhanced tissue and background tissue. Motion can even simulate GBCA enhancement or obscure it [3]. Furthermore, motion impairs the comparison between the different phases of post-contrast imaging and pre-contrast T1W images. However, using breathing-independent sequences and breath-hold sequences we can obtain high-quality diagnostic MR images in the great majority of patients. For evaluation of GBCA enhancement, we obtain T1W gradient echo sequences (instead of spin-echo sequences): spoiled gradient echo (2D-SGE) or three dimensional gradient echo (3D-GE). Current “state-of-the-art” MR systems use 3D-GE as the primary technique for dynamic contrast-enhanced imaging of the abdomen. The addition of parallel imaging technique, especially with multi-channel coils, can significantly shorten imaging times, rendering improved comfort and as a result compliance for patients on breath-hold sequences.

In noncooperative agitated patients, we can use magnetization-prepared rapid-acquisition gradient echo (MPRAGE), which is a single shot sequence, with image acquisition duration of 1 to 2 seconds, rendering it relatively breathing independent [4]. However, adequate quality MPRAGE sequences (with 180° non-slice selective technique, which is required for dynamic GBCA-enhanced acquisitions) are currently available only in “state-of-the-art” MR systems. In this regard, the higher intrinsic signal to noise of 3T system allow for good quality image with MPRAGE especially when water excitation is employed [4].

CORRECT SUBPHASE OF ENHANCEMENT

Following GBCA injection we routinely obtain three phases to evaluate tissue perfusion (hepatic arterial dominant phase), blood pool venous withdrawal (early hepatic venous phase), and interstitial space size (interstitial phase) (Fig. 1).

The most important set of images are those obtained in the hepatic arterial dominant phase (HADP), in which the contrast must be present in the hepatic arteries and portal veins prior to appearing in the hepatic veins [5-9] (Fig. 1). There is a relatively short time window for this and correct timing is critical. Three different techniques including empirical timing, test bolus and bolus tracking methods have been employed [10]. It has been recently reported that arterial-phase bolus-track liver examination (ABLE) technique is a successful method for the acquisition of hepatic arterial dominant phase [11]. A bolus track sequence, which produces an image approximately every second, is used to detect the arrival of GBCA to the level of celiac axis. After 8 seconds, the liver is scanned with standard ordered k space for a 16-20 second sequence. During the 8 second-period, the patient is given breathing instructions. This is also the technique that we routinely use now at our institution.

All these techniques depend on the empirical estimation of circulation time of the contrast material from the site of injection or abdominal aorta to the liver [5,10]. Because the circulation time of the contrast material to the liver shows variations depending on various factors, different subphases of enhancement, other than the HADP, may be detected in the “early post-contrast hepatic imaging”, based on the vessel enhancement patterns as well as the enhancement of abdominal parenchymal organs [5]. A recent report of our group on this subject, employing an 18-s set time-delay empiric timing scheme for the initiation of scanning allowed us to observe five subphases of early contrast enhancement of the liver: (a) early hepatic arterial phase (EHAP), (b) mid-hepatic arterial phase (MHAP), (c) late hepatic arterial phase (LHAP), (d) splenic vein only HADP (SVHADP) and (e) HADP [5] (Table 1).

On the EHAP, the renal cortex, spleen, pancreas and liver similarly demonstrated no or slight enhancement, which significantly increased on the MHAP. The enhancement of the pancreas on two first subphases was significantly lower, both qualitatively and quantitatively, compared to the other later phases (LHAP, SVHADP and HADP), where it was not significantly different. The liver demonstrated significantly higher enhancement on HADP and SVHADP compared to LHAP. Our results suggest that it is too early to evaluate liver enhancement during EHAP and MHAP (Fig. 2). In addition, EHAP is essentially identical to non-contrast images in demonstrating liver lesions, which is thus unable to provide distinctive enhancement of focal liver lesions. The LHAP, SVHADP and HADP can be used to evaluate liver lesion enhancement and may be considered as the optimal range of phases of enhancement. The pancreatic enhancement was maximal in LHAP, and therefore pancreatic capillary blush can be a useful surrogate for optimal timing to evaluate liver disease (Fig. 1). These impressions are also supported by previous reported findings in the literature [6,7,8,12,13,14]. The enhancement of renal veins can help to determine the adequacy of liver enhancement, since we consider an important finding of true LHAP is contrast in these veins, in addition to visual pancreatic enhancement (Fig. 1). On the subsequent SVHAD phase there is also the enhancement of suprarenal IVC, splenic vein and portal vein. On HADP, contrast also appears in the superior mesenteric vein (Table 1).

TABLE 1 - The vessel and organ enhancement patterns according to subphases

*Reprinted with permission from reference 5

The combination of these vessel enhancement patterns with the enhancement patterns of the renal cortex and spleen may also serve as adequate surrogates and a guide for the optimal phase of enhancement in the case of chronic pancreatitis or other pancreatic diseases, where pancreatic enhancement may be minimal [5].

Using these landmarks, we can determine with confidence if the “first pass” or capillary bed enhancement of tissues has been captured. Optimal timing of contrast enhancement is essential not only for the detection of liver lesions, especially hypervascular tumors such as HCC, hypervascular liver metastases and benign hepatocellular tumors, but also for evaluation of response to treatment (Fig. 3) [5-9,12-15]. The vascularity of tumors, as evidenced by extent of early enhancement, has been shown to predict the likelihood of response to certain treatment methods [14].

It is essential to know if the contrast-enhanced images have optimal timing when analyzing other abdominal parenchymal organs as well. For instance, too little pancreatic enhancement is consistent with pancreatic fibrosis or chronic pancreatitis, and too little enhancement of renal cortex may imply ischemic nephropathy or acute cortical necrosis. This can be reliably judged on HADP images, based on the fixed vessels landmarks cited above. In the EHAP, minimal enhancement of pancreas or renal cortex may reflect an early image acquisition rather than disease process (Fig. 2).

The early hepatic venous phase images are obtained between 45 and 90 seconds post-contrast injection and can be recognized by the presence of GBCAs in portal and hepatic veins (Fig. 1). Thus, the time window is relatively wide and timing is not so critical. This phase shows maximal enhancement of the hepatic parenchyma and is especially useful for the detection of hypo/isovascular HCC and hypovascular metastases. It is also important in lesion characterization, specifically hypervascular lesions such as focal nodular hyperplasia, which shows fading and HCC which shows washout in this phase [6,8,9,14,15] (Fig. 3).

The Late hepatic-venous or interstitial phase has a broad time range, which is from approximately 90 seconds to 5 minutes after initiation of contrast injection, with no exact timing requirement. During this phase, hepatic parenchymal enhancement persists (Fig. 1). This phase also provides additional information to characterize focal hepatic lesions, by demonstrating their late-phase temporal handling of contrast. Hemangiomas reveal progressive enhancement, persistent enhancement is observed in small-sized hemangiomas, and washout of hypervascular metastases and HCC is also apparent during this phase [6,8,9,14,15] (Fig. 3).

 
FIELD STRENGTH

The current worldwide proliferation of 3.0T MR imaging systems beyond academic and research centers has added other important consideration to the routine clinical practice, especially regarding GBCA-enhanced images: 3.0T versus 1.5T.

The major advantage of MR imaging at 3T compared to 1.5T is the theoretical twofold increase in signal-to-noise ratio (SNR), which can be translated into higher spatial resolution and/or temporal resolution, particularly with the use of parallel imaging techniques [4,16-25]. At 3.0T, post-gadolinium T1W 3D-GE sequence can be obtained with higher quality than at 1.5T, primarily because of the thinner section acquisition, and is relatively resistant to the drawbacks of 3.0T MRI including specific absorption rate constraints, prolonged T1 relaxation times and the increase in imaging artifacts [16-25].

The ability of GBCAs to reduce T1 (known as the relaxivity) is slightly lower at 3T (5% to 10%) compared to 1.5T. However, the T1 relaxation times of the tissues are prolonged on the order of 40% or more at 3.0T compared to 1.5T. Therefore, an equivalent dose of GBCA at 3.0T causes an increased contrast difference compared to 1.5T [16,18,21,23-25]. This increased effect of GBCAs contributes to better SNR and contrast-to-noise ratio (CNR) at 3.0T. Corroborating this and supporting previous descriptions in the literature, one recent report of our group has demonstrated consistent differences in the extent of enhancement of abdominal organs between 1.5T and 3.0T on the subphases of hepatic arterial enhancement, with achievement of higher relative enhancement at 3.0T [17]. The benefits of this higher extent of enhancement at 3.0T may potentially translate into better detection of lesions that have a blood supply greater or lesser than background abdominal organs. This is particularly important to detect hypervascular lesions in the cirrhotic liver (such as HCC) (Fig. 3), and hypervascular metastases [6,9,16,17,21].

The addition of the newly available 32-channel coil to MRI systems will permit significant improvement in parallel imaging, which will accelerate 3D GE and will aid in achieving ultra-short, high-quality, dynamic 3D imaging following bolus contrast administration. These aspects will probably render much superior detection of lesions on systems equipped with this functionality, raising an important consideration to the daily clinical practice of comparing lesions between different MR studies. For example, a MR exam for control of a probably high-grade dysplastic nodule versus small HCC in a cirrhotic patient or for treatment control of small liver metastases in a breast cancer patient obtained at a 3.0T MR system. The lesions are better seen at the present exam and seem to be enhanced more in the HADP post-contrast imaging, when compared to the previous exam obtained at 1.5T. Is the difference a result of better lesions display at 3.0T compared to 1.5T or disease progression?

TYPES OF GBCAs

Gadolinium is a rare earth metal atom and has seven unpaired electrons, which renders it highly paramagnetic. Free gadolinium is toxic in vivo and forms colloid particles that are phagocytized by the reticuloendothelial system. The chelation of gadolinium to organic ligands (chelate complex) is necessary for the atom to be used as an in vivo contrast agent in humans. This process makes the ion chemically inert [1,2,26,27]. There are several formulations available with different ligands, constituting the GBCAs (Table 2). Depending on the ligand, GBCAs are classified into linear or macrocyclic (according to the backbone structure of their amine group) and may be further subclassified according to their charges as ionic or non-ionic. Depending on these characteristics, GBCAs dissociate to varying extent in solution. Following intravenous injection, it is possible to detect gadolinium-chelate complexes, free gadolinium ions and free ligands in human tissues. This dissociation can be defined by the thermodynamic stability constant and dissociation constant [28,29]. Thermodynamic stability constant determines the concentration at which gadolinium ions will dissociate from gadolinium-chelate complexes. The rate of this dissociation reaction is dependent on the dissociation constant [28,29]. Taken together, these two constants define the affinity of ligands for gadolinium ions at the physiological pH [28,29]. Each GBCA has a different thermodynamic stability constant and dissociation rate. Macrocyclic GBCAs create tighter bonds with gadolinium and therefore have higher thermodynamic stability constants and lower dissociation rates [28]. The electrostatic charges present in ionic GBCAs render tighter bonding than nonionic GBCAs, and therefore ionic GBCAs are more stable than non-ionic GBCAs. These latter pharmacological characteristics have become relevant in the radiological literature since October 2006, when it became recognized that GBCAs could result in the condition nephrogenic systemic fibrosis (NSF) in patients with renal impairment [30]. High thermodynamic stability constants and lower dissociation rates (greater affinity of ligands for gadolinium ions) are important qualities of GBCAs to minimize risks of NSF.

In clinical use GBCAs can further be classified into three types according to their distribution in the body following intravenous injection: a) extracellular agents, b) combined extracellular and intracellular agents, c) blood pool agents [26,27].

Nonspecific Extracellular Contrast Agents

The great majority of the GBCAs in clinical use are nonspecific extracellular contrast agents, sharing similar pharmacokinetics with iodinated contrasts in the abdomen and throughout the body. After intravenous injection, they follow the route of blood circulation and in the abdomen first reach the aorta and its branches; enter the splanchnic and splenic circulation by the celiac axis, superior and inferior mesenteric arteries, and then into their companion veins and subsequently into the portal vein. Contrast agents enter the venous system after the passage through the sinusoids in the liver, and after passing the capillaries in the peripheral circulation. They are freely redistributed from the vascular to the interstitial space [1,2,26,27]. Whereas the iodine molecule is directly imaged at computed tomography (CT), in MR imaging it is the effect of gadolinium that is evaluated rather than the agent itself. Gadolinium exhibits an amplification effect, in which many adjacent water protons are relaxed by a single gadolinium atom, shortening T1 of the tissues (effect known as relaxivity, r1). As a result, MR imaging is orders of magnitude more sensitive to the effect of gadolinium than is CT to the effect of iodine [27]. The recommended dose of the extracellular contrast agents is 0.1 mmol/kg of body weight. The recommended injection rate is 2-3 ml/sec. All extracellular agents are eliminated by the kidneys and are not excreted by the hepatobiliary system in patients with normal renal function. They do not exhibit protein binding (Table 2).

Extracellular agents can be used for the acquisition of hepatic arterial dominant, early hepatic venous and interstitial phases of standard gadolinium-enhanced MRI studies. Contrast agents with greater T1-relaxivity may show better enhancement effects. In this regard, Gadovist [gadobutrol, (Bayer Schering Pharma AG, Germany)], which is a nonionic macrocylic GBCA, has higher T1 relaxivity and can be theoretically administered in a lower dose to achieve the same imaging effect. Due to its macrocylic structure Gadovist has higher stability in solution and there are no cases of NSF related to its use [1].

Dotarem [gadoterate meglumine, (Guerbet, Aulnay-sous-Bois, France)] is an ionic macrocylic agent, and therefore combines both the predominant stability factor of macrocyclic design with the ancillary factor of ionicity. Hence, from a theoretical standpoint of stability, Dotarem should be one of the best of all the GBCAs. Until the present time, there are no cases of NSF related to its use [1].

Combined Extracellular and Intracellular Contrast Agents

This class of GBCAs is distributed into the extracellular space including vascular and interstitial spaces, and intracellular spaces of hepatocytes (hepatocyte phase). Therefore, these agents can also be termed as combined extracellular and hepatocyte specific agents. They can be used for the acquisition of the hepatic arterial dominant, early hepatic venous, interstitial and hepatocyte phases of GBCA-enhanced MRI studies. These agents are Multihance [gadobenate dimeglumine, (Bracco Diagnostics, Milan, Italy)] and Eovist [gadoxetic acid, (Bayer HealthCare Pharmaceuticals, Wayne, NJ, USA), in the US] / Primovist [gadoxetic acid, (Bayer Schering Pharma AG, Germany), outside the US]. They are taken up by hepatocytes and excreted into bile ducts. Consequently, they have dual elimination including both renal and biliary eliminations (Table 2). The hepatocyte phase is useful to characterize lesions with biliary structures, especially focal nodular hyperplasia (FNH) [27,31,32]. This phase is particularly helpful for the differentiation of FNH from adenoma. FNH contains hepatocytes and biliary canaliculi; therefore, hepatocyte specific agents can be uptaken and excreted into the bile ducts in FNH. However, hepatic adenomas do not contain normal hepatocytes and biliary canaliculi; therefore, hepatocyte specific agents do not show uptake. Thus, while FNH enhances on the hepatocyte phase, hepatic adenomas do not [27,31]. The hepatocyte phase is also helpful for the detection of lesions which do not contain hepatocytes, as signal intensity difference is expanded between enhanced liver parenchyma and non-enhanced lesions. This includes metastases, adenomas or poorly differentiated hepatocellular carcinomas (HCCs).

Multihance has been used in Europe for number years and in the United States, it was approved for use in December 2004 [26,27]. The agent has shown good patient tolerance, and to date with over three million doses administered, no cases of NSF have been associated with its use alone (no confounded cases). It demonstrates weak and transient binding with serum albumin in the intravascular space, remaining for a longer time than do other gadolinium chelates. In addition, this protein-binding characteristic results in increased T1 relaxivity compared with that achieved with other GBCAs. Increased T1 shortening results in increased signal intensity, which is useful for MR angiography and may yield improvements in tumor imaging [31,32]. Serial contrast-enhanced liver imaging can be performed with the use of Multihance after bolus injection, in the same fashion as with other nonspecific extracellular contrast agents [5,17,31,32]. The results are comparable with other conventional extracellular contrast agents, particularly for the improved visualization of hypervascular lesions [5,17,31,32]. Multihance is approved for use at 0.1 mmol/kg; however, it has been shown that 0.05 mmol/kg (half dose) of this agent has comparable diagnostic efficacy compared to the full dose of standard extracellular GBCAs [33]. We therefore routinely use half dose Multihance. One of our major considerations is to minimize the volume of a GBCA. Multihance shows hepatobiliary contrast enhancement peak at 60 to 120 minutes after intravenous injection. This requires two separate imaging sessions for the patient.

Eovist/Primovist has greater degree of protein binding than Multihance and a much greater proportion of biliary excretion (Table 2). Consequently, the hepatocyte phase is earlier, occurring at about 20 minutes (lasting till about 4 h). This permits ready acquisition of an entire post-contrast study, including hepatic arterial dominant, early hepatic venous, and interstitial phases, with hepatocyte-phase, in one imaging session. Eovist/Primovist is used at ¼ dose (0.025 mmol/kg) of the standard dose of extracellular GBCAs. However, there is not sufficient data demonstrating the diagnostic efficacy of this dose of Eovist compared to the standard doses of extracellular GBCAs. A recent report on this subject [34] using the quantitative assessment has demonstrated that the enhancement effect of abdominal solid organs and aorta in the arterial phase and major in the portal phase with Gd-EOB-DTPA (Primovist) was significantly lower compared with that of Gd-DTPA (Magnevist) in the same subjects (healthy male volunteers), but significantly higher in the latter phases. The authors suggest the reassessment of Gd-EOB-DTPA dose [34].

TABLE 2 - Gadolinium based contrast agents

*Gadobenate dimeglumine, gadoxetic acid, gadofosveset and gadobutrol have higher T1 relaxivity compared to the other agents.

†Gadobenate dimeglumine has been reported to be diagnostically effective at its half dose (0.05 mmol/kg).

‡Gadobutrol solution has double amount of gadolinium in its 1 molar (M) concentration compared to 0.5 M concentration of the other GBCAs’ solutions

Blood Pool Contrast Agents

Blood pool agents predominantly stay in the vascular space. Vasovist [gadofosveset, (Epix Pharmaceuticals, Lexington, MA, USA)] is the blood pool agent approved by the FDA (for body MRAs). 85% of Vasovist binds serum albumin transiently and reversibly. A small amount of Vasovist is also distributed into the extracellular space. Binding to serum albumin provides higher T1 relaxivity and extended intravascular enhancement, which is even higher and longer with Vasovist compared to the other protein binding GBCAs (Multihance and Eovist). This agent may be advantageous for vascular imaging. However, the diagnostic role and safety profile of this agent needs to be determined, and we have no clinical experience with this agent.

CONCLUSION

This review describes the important characteristics of GBCA-enhanced imaging. We have emphasized the importance of breath-holding and of understanding the significance of exact subphase of enhancement. We have described 1.5T and 3.0T imaging and advantages of 3T. The major concern with these agents is NSF. Finally, we have described a number of the agents in clinical use that we recommend using.

REFERENCES

1.      Semelka RC. Abdominal-Pelvic MRI, 3^rd edition. New York: Wiley-Liss, 2009 (in press).

2.      Semelka RC, Helmberger T. Contrast Agents for MR Imaging of the Liver. Radiology 2001; 218:27-38.

3.      Rescinito G, Sirlin C, Cittadini Jr G. Body MRI artefacts: from image degradation to diagnostic utility (Artefatti in RM body: dall’effetto degradante sulle immagini alle potenzialità diagnostiche). Radiol med 2009; 114:18–31.

4.      Altun E, Semelka RC, Dale BM, Elias J Jr. Water excitation MPRAGE: an alternative sequence for postcontrast imaging of the abdomen in noncooperative patients at 1.5 Tesla and 3.0 Tesla MRI. J Magn Reson Imaging. 2008; 27:1146-1154.

5.      Goncalves Neto JA, Altun E, Vaidean G, et al. Early contrast enhancement of the liver: exact description of subphases using MRI. Magn Reson Imaging 2009; 27:792-800.

6.      Willatt JM, Hussain HK, Adusumilli S, Marrero JA. MR imaging of hepatocellular carcinoma in the cirrhotic liver: challenges and controversies. Radiology 2008; 247:311–30.

7.      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; 225:407–15.

8.      Semelka RC, Martin DR, Balci NC. Magnetic resonance imaging of the liver: how I do it. J Gastroenterol Hepatol 2006; 21: 632–7.

9.      Silva AC, Evans JM, McCullough AE, Jatoi MA, Vargas HE, Hara AK. MR Imaging of Hypervascular Liver Masses: A Review of Current Techniques. RadioGraphics 2009; 29:385–402.

10.  Chu LL, Joe BN, Westphalen ACA, Webb EM, Coakley FV, Yeh BM. Patient-specific time to peak abdominal organ enhancement varies with time to peak aortic enhancement at MR imaging. Radiology 2007; 245: 779–87.

11.  Sharma P, De Becker J, Beck GM, Friel HM, Burrow B, Martin D. Arterial-phase Bolus-track Liver Examination (ABLE): Optimization of Liver Arterial Phase Gadolinium Enhanced MRI Using Centric Re-ordered 3D Gradient Echo and Bolus Track Real-Time Imaging. Proc. Intl. Soc. Mag. Reson. Med. 2006; 14:3318.

12.  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; 184:63–9.

13.  Danet IM, Semelka RC, Nagase LL,Woosley JT, Leonardou P, Armao D. Liver metastases from pancreatic adenocarcinoma: MR imaging characteristics. J Magn Reson Imaging 2003; 18:181–8.

14.  Braga L, Semelka RC, Pietrobon R, Martin D, de Barros N, Guller U. Does hypervascularity of liver metastases as detected on MRI predict disease progression in breast cancer patients? AJR Am J Roentgenol 2004; 182:1207–13.

15.  Kierans AS, Leonardou P, Shaikh F, Semelka RC. Body MR imaging: Sequences we use and why. www.appliedradiology.com , 2009.

16.  Ramalho M, Altun E, Heredia V, Zapparoli M, Semelka RC. Liver MR imaging: 1.5T versus 3T. Magn Reson Imaging Clin N Am 2007; 15: 321–47.

17.  Goncalves Neto JA, Altun E, Elazzazi M, Vaidean G, Chaney M, Semelka RC. Enhancement of abdominal organs on hepatic arterial phase: quantitative comparison between 1.5 T and 3.0 T MRI. Magn Reson Imaging 2009, In Press.

18.  Choi JY, Kim MJ, Chung YE, et al. Abdominal applications of 3.0-T MR imaging: comparative review versus a 1.5-T MR system. Radiographics 2008; 28:30.

19.  Merkle EM, Dale BM, Paulson EK. Abdominal MR imaging at 3.0T. Magn Reson Imaging Clin N Am 2006; 14:17–26.

20.  Merkle EM, Dale BM. Abdominal MRI at 3.0 T: the basics revisited. AJR Am J Roentgenol 2006; 186:1524–32.

21.  Zapparoli M, Semelka RC, Altun E, et al. 3.0-T MRI evaluation of patients with chronic liver diseases: initial observations. Magn Reson Imaging 2008; 26:650–60.

22.  Soher BJ, Dale BM, Merkle EM. A review of MR physics: 3T versus 1.5T. Magn Reson Imaging Clin N Am 2007; 15:277–90.

23.  Edelman RR, Salanitri G, Brand R, et al. Magnetic resonance imaging of the pancreas at 3.0 Tesla: qualitative and quantitative comparison with 1.5 Tesla. Invest Radiol 2006; 41:175–80.

24.  Merkle EM, Dale BM, Barboriak DP. Gain in signal-to-noise for firstpass contrast-enhanced abdominal MR angiography at 3 Tesla over standard 1.5 Tesla: prediction with a computer model. Acad Radiol 2007; 14:795–803.

25.  Michaely HJ, Kramer H, Dietrich O, et al. Intraindividual comparison of high-spatial-resolution abdominal MR angiography at 1.5 T and 3.0T: initial experience. Radiology 2007; 244:907–13.

26.  Balci NC, Semelka RC. Contrast agents for MR imaging of the liver. Radiol Clin North Am 2005; 43:887–898.

27.  Gandhi SN, Brown MA, Wong JG, Aguirre DA, Sirlin CB. MR Contrast Agents for Liver Imaging: What, When, How. RadioGraphics 2006; 26:1621–36.

28.  Rofsky NM, Sherry DA, Lenkinski LE. Nephrogenic systemic fibrosis: a chemical perspective. Radiology 2008; 247:608-612.

29.  Lin S-P, Brown JJ. MR contrast agents: physical and pharmacologic basics. J Magn Reson Imaging 2007; 25:884-899.

30.  Penfield JG, Reilly RF Jr, Medscape. What nephrologists need to know about gadolinium. Nat Clin Pract Nephrol 2007; 3:654-668.

31.  Reimer P, Schneider G, Schima W. Hepatobiliary contrast agents for contrast-enhanced MRI of the liver: properties, clinical development and applications. Eur Radiol 2004;14:559– 78.

32.  Kuwatsuru R, Kadoya M, Ohtomo K, et al. Comparison of gadobenate dimeglumine with gadopentetate dimeglumine for magnetic resonance imaging of liver tumors. Invest Radiol 2001; 36:632–641.

33.  Schneider G, Maas R, Schultze Kool L, et al. Low-dose gadobenate dimeglumine versus standard dose gadopentate dimeglumine for contrast-enhanced magnetic resonance imaging of the liver: an intra-individual crossover comparison. Invest Radiol 2003; 38:85-94.

34.  Tamada T, Ito K, Sone T, et al. Dynamic Contrast-Enhanced Magnetic Resonance Imaging of Abdominal Solid Organ and Major Vessel: Comparison of Enhancement Effect between Gd-EOB-DTPA and Gd-DTPA. J. Magn. Reson. Imag 2009; 29:636–640.