This article confers 1.0 ARRT Category A Continuing Education credit, which will be awarded upon completion of an online post test. The entire text of this supplement, learning objectives, and the posttest are available at appliedradiology.org/aici.
Gadolinium-based contrast agents (GBCAs) were first approved for magnetic resonance imaging (MRI) in 1988 based on their ability to improve soft-tissue contrast relative to noncontrast exams.1 Since then, GBCA use has greatly expanded, and GBCAs are now used routinely in approximately 45% of MRI exams in the United States.2,3 Until recently, available general-use GBCAs consisted of the conventional-relaxivity simple linear agents Magnevist® (gadopentetate dimeglumine), Omniscan™ (gadodiamide), and OptiMARK™ (gadoversetamide); the high-relaxivity substituted linear GBCA MultiHance® (gadobenate dimeglumine), and three macrocyclic agents: ProHance® (gadoteridol), Gadavist® (gadobutrol), and Dotarem® (gadoterate meglumine).4-10 (Tables 1, 2) OptiMARK and Magnevist were removed from the market in 2018 and 2019, respectively, while GE introduced Clariscan™ (gadoterate meglumine), a generic version of Dotarem, in 2019.11
Until approximately 2006, GBCAs were deemed safe for patients, particularly those with renal impairment. Moreover, their safety profiles were considered comparable. Because substantial evidence demonstrated that higher doses provided additional diagnostic yield with few associated safety concerns, doses were frequently higher than the standard (approved) 0.1 mmol/kg dose. Double or even triple doses of some agents were routinely used for MR angiography and for specific central nervous system (CNS) applications.13,14 Today only one GBCA, ProHance, retains a triple-dose (0.3 mmol/kg) indication for MRI of the CNS in adults.8 Unfortunately, several events have occurred since 2006 that have altered the perception that all GBCAs are equally safe. These include the identification of nephrogenic systemic fibrosis (NSF) as a serious potential adverse event (AE),15 and an appreciation that gadolinium (Gd) is retained in brain, bone, and soft-tissue organs,16-19 with unknown long-term clinical consequences.
In this article, we present a summary of discussions from a recent Expert Panel Forum on Advancing Clinical Practice in MR Imaging. The primary goal of the forum was to explore the real and perceived risks associated with GBCA use and to elucidate the radiologist’s role in performing the necessary risk–benefit assessment when considering contrast-enhanced MRI for specific indications. We also address two fundamental issues: 1) when is contrast necessary, and 2) what are the best practices for its use in selected applications, specifically neuroradiology/stroke imaging, breast screening, and abdominal imaging? Finally, we discuss practical administrative and cost considerations related to contrast use in MRI.
Although the risks associated with GBCA use can be viewed as the probability of harm, the focus should center on evaluating real risk vs perceived risk. Real risk is evidence-based and determined by the severity and likelihood of harm. GBCAs vary with respect to their structure, stability, ionicity, osmolality, and viscosity, all of which can contribute to the severity and likelihood of harm. (Table 1)
Based on data provided in package inserts, most AEs associated with GBCAs are mild and non-serious. Serious AEs and allergic reactions requiring hospitalization are extremely rare, occurring in about 0.008% of GBCA administrations.21 Importantly, the rates and types of AEs are comparable across agents.4-11 Although contrast extravasation is an indirect safety issue associated with GBCAs, potentially causing pain and swelling at the injection site, this can be limited with GBCAs that have lower viscosity and osmolality that enable greater ease of injection.20 (Table 1)
Perhaps the most clinically impactful property of GBCAs relates to their stability and potential for dechelation; ie, release of the Gd ion from the chelate structure. Macrocyclic chelates encircle the Gd ion in a molecular cage, and both in vitro and in vivo data support the greater stability of these GBCAs relative to standard relaxivity linear GBCAs, particularly relative to the nonionic linear GBCAs.16-18 Nevertheless, there was little recognition of the potential clinical consequences of dechelation until 2006, when it was shown that the combination of severe renal insufficiency/failure, exposure to higher/repeat doses of GBCAs (primarily the least stable simple linear agents), and other less understood factors could lead to NSF in some patients.15
Most unconfounded cases of NSF were associated with exposure to the standard-relaxivity simple linear GBCAs, Omniscan and Magnevist; as a result, the American College of Radiology (ACR) designated these agents, along with OptiMARK, as Group I, or high-risk, agents.21 (Table 3) These agents have since been contraindicated by the US Food and Drug Administration (FDA) for use in patients with acute kidney injury (AKI) or chronic, severe kidney disease (ie, an estimated glomerular filtration rate [eGFR] <30 mL/min/1.73 m2).4-6 Group II, or low-risk, agents include the higher-relaxivity substituted linear agent MultiHance, and the macrocyclic agents Dotarem, Gadavist, and ProHance.21
In 2017, the ACR made renal function screening (either by questioning the patient or measuring serum creatinine) optional for Group II agents in inpatients and outpatients.21 Elimination of eGFR screening without increasing risk to patients is potentially beneficial in terms of both cost and time efficiency. Conversely, if Group I agents are used, an eGFR should be obtained for inpatients, while outpatients should be asked about their history of conditions that may be associated with reduced renal function.21 Note that there is a single Group III agent, Eovist® (gadoxetate disodium), a liver-specific MR contrast agent, for which there are limited data regarding NSF risk, but for which few, if any, unconfounded cases of NSF have been reported.21 Changes in clinical practice such as screening high-risk patients for renal dysfunction, use of the lowest effective contrast dose, and use of a Group II GBCA appears to have eliminated NSF risk in patients referred for contrast-enhanced MRI.
A second change with regard to the perceived safety profile of linear vs macrocyclic GBCAs occurred in 2014 with a publication that demonstrated increased signal in the dentate nucleus and globus pallidus on noncontrast T1 MRI scans in patients with a history of having received gadolinium-based contrast.19 Although this initial study investigated only patients who received Magnevist or Omniscan, subsequent studies found that T1 hypersignal on noncontrast T1 MR images occurred more frequently with linear GBCAs than with macrocyclic agents. This caused many to speculate that, as with NSF, the linear GBCAs dechelate more readily, leading to long-term Gd retention in the brain and body.23-29 Gd retention has since been identified even after exposure to low GBCA doses in patients with normal renal function,30-31 and in the presence of an intact blood-brain barrier.32 This indicates that Gd retention likely occurs to a greater or lesser extent in all patients who receive a GBCA. Among macrocyclic GBCAs, visible hypersignal has been seen so far only after multiple injections of Gadavist and Dotarem.33-39 Among linear GBCAs, greater Gd retention has been observed after Omniscan than after Magnevist, MultiHance, or Eovist.22,29,31,32,40-43 Notably, knowledge that Gd is retained in body tissues is not new; several studies have long shown that Gd is retained in bone and body tissues.16-18
Although frequently considered a marker for Gd retention, visible T1 hypersignal is actually a very poor indicator. Although many disease conditions unrelated to Gd can cause hyperintensity in the dentate nucleus and globus pallidus, the brain has been shown to retain Gd even in the absence of visible hyperintensity or demonstrable T1 shortening. Unequivocal demonstration of Gd retention comes from tissue sampling studies using inductively coupled plasma mass spectrometry (ICP-MS). Such studies provide far superior sensitivity for quantifying Gd than studies based on analysis of MRI signal intensity.22
Using ICP-MS, Murata et al demonstrated the presence of Gd in the brain following administration of all GBCAs available in the U.S. at the time, including macrocyclic agents.29 Gd was also detected in bone at much higher levels than in the brain.29 Unfortunately, the study was performed on tissue samples from a relatively small number of human decedents (N=9) who, in some cases, passed away only days after contrast administration. Hence, further confirmatory studies are needed in a larger population. Gd retention studies are performed more easily in animals. Two recent such studies comparing Gd retention among macrocyclic agents 28 days after exposure confirmed that differences do exist, with significantly lower levels of Gd noted with ProHance than with Dotarem and Gadavist in the cerebellum, cerebrum, kidneys, liver, and skin.44,45 A third study performed in rats demonstrated Gd clearance differences among the macrocyclic agents in the first weeks and months after administration, with ProHance clearing much more rapidly than Gadavist and Dotarem.43 This might be clinically significant given that one rat year equates to roughly 30 human years, and might be of particular relevance when imaging very young patients, whose brain and cognitive function are still developing.
In May 2017, the FDA stated that although Gd retention has been observed, no adverse health effects have been identified and no restrictions on GBCA use are warranted. However, the agency cautioned that GBCA use should be limited wherever possible.46 The FDA also stated that, based on the literature, Gd retention levels appear to be greater with linear agents than with macrocyclic agents. At about the same time, a summit meeting of drug industry representatives, medical professionals, and FDA personnel met to discuss Gd retention and the possible implications and future actions.47 It was decided that new language would be added to the prescribing information for GBCAs, warning of the potential for Gd retention, particularly with linear agents, and recommending that risk minimization measures be practiced in at-risk populations. Notably, the European Medicines Agency (EMA) position differs markedly from the FDA’s, with all simple linear agents removed from the market and leaving only macrocyclic GBCAs available for general use, along with the substituted linear agents MultiHance and Primovist (Eovist) for liver imaging.48
With no unequivocal evidence of clinical sequelae from retained Gd, assessing the real risk of Gd retention is difficult. GBCAs have been approved and used routinely in clinical practice for more than 30 years, with some 450 million doses administered worldwide with little acute or long-term adverse reactions. However, until further long-term studies confirm that no adverse clinical outcomes are associated with Gd retention, careful GBCA selection and use is warranted in at-risk populations.
Care should be taken in patients requiring serial MRI examinations, particularly those with seizures or multiple sclerosis (MS), and in young women with dense breasts who require MRI for breast cancer screening or monitoring. Contrast should be used primarily in cases where it has the greatest potential to impact patient management. Moreover, macrocyclic GBCA use is considered prudent in such cases to minimize the potential risks of Gd retention.
Most MRI examinations in children require contrast. These include workups of neoplasia, infections, and inflammatory processes, particularly at initial diagnosis, with exceptions for most congenital and/or structural disorders. The potential impact of Gd retention in pediatric patients remains unknown, as long-term data are unavailable. However, logic would dictate, and trends support, that macrocyclic agent use in children would be prudent. In this regard, animal studies have shown ProHance to be cleared more rapidly than Gadavist and Dotarem, resulting in lower levels of Gd retention in the first weeks and months.43-45 If the human situation reflects that of animals, then the first 5 weeks of rat life would correspond to approximately 3 human years. In these patients, less Gd would be retained if ProHance were used.
On the other hand, whereas all the macrocyclic agents (Dotarem, Gadavist, and ProHance) and the substituted linear agent MultiHance, are approved for use in children older than 2 years of age, only Dotarem, Gadavist, and MultiHance have a neonate indication.7-10 Among these agents, MultiHance is unique in possessing higher r1 relaxivity, meaning that greater signal is obtained at equivalent dose or equivalent signal at lower dose compared with all other GBCAs. (Table 2) MultiHance has a flexible dosing indication in neonates and infants, allowing for doses of 0.05 – 0.1 mmol/kg.7 Thus, a lower dose and, potentially, lower levels of retained Gd are possible with MultiHance.
The Expert Panel Forum on Advancing Clinical Practice in MR Imaging highlighted contrast use in several applications. “Why Gadolinium Matters Today,” a presentation delivered by David Enterline, MD, provided an overview of similarities and differences among linear and macrocyclic GBCAs. A summary of select efficacy data is shown in Figure 1, along with a link to the full video presentation.
In “Contrast Enhanced MR Acute Stroke Protocol in the ED,” John Karis, MD, discussed the goals of acute stroke MR imaging, as well as some of the technical and practice considerations in imaging this condition. (See Figure 2 for the conclusions of his presentation and a link to the full video.)
James Sancrant, DO, spoke on various aspects of breast MRI screening, including indications, performance, and protocols. Dr. Sancrant presented a case in which breast MRI was used to image a patient who underwent 3D mammogram and ultrasound. While, both modalities generated equivocal results, breast MRI clearly showed a brightly-enhancing lesion. (Figure 3) A link to the video presentation is provided in the figure legend.
“Contrast-enhanced MR of the Abdomen and Pelvis,” by Kristin Porter, MD, PhD, outlined the ACR’s appropriateness criteria for abdominal and pelvic MR imaging and discussed a number of cases highlighting the use of contrast-enhanced MRI of the liver and urologic system, and pelvic imaging in women. Figure 4 shows an example of a hepatic application for MRI, the differentiation of mass-forming intrahepatic cholangiocarcinoma (ICC) from hepatocellular carcinoma (HCC). (See the figure legend for a link to the full video presentation.)
Contrast-enhanced MRI of the bowel was the focus of a presentation by Jorge Soto, MD, who reviewed indications and shared his institution’s protocols for the procedure. Dr. Soto also showed several examples of patients undergoing MRI for Crohn’s disease, rectal carcinoma, perianal fistulas, and other conditions. See Figure 5 for a summary of Dr. Soto’s findings in the follow-up of a patient with recurrent Crohn’s disease who had undergone resection of the terminal ileum.
Economic considerations are vital in today’s healthcare environment. Practice considerations including the selection and utilization of MR contrast can impact workflow, which in turn affects costs. For example, GBCA-enhanced imaging can increase scan time, but also potentially reduces the need for additional and/or unnecessary imaging, which may lead to overall cost savings. Decisions related to which GBCA(s) a practice uses can have economic implications: one may choose to stock several GBCAs to allow for selection of an optimal agent based on an individualized risk-benefit assessment or, alternatively, only 1 or 2 agents based on pricing or contractual constraints. In either case, an increasing number of practices are choosing to use a macrocyclic agent based on their safety profile with respect to NSF risk and Gd retention, particularly if the applications include pediatric applications and/or serial imaging.
Radiologists are not always directly involved in GBCA selection at their institution. In such cases, communication between the Radiology Department, hospital administration and/or contrast decision making committees, is important. An evidence-based case can be made by the Radiologists when macrocyclic or higher-relaxivity agents offer the possibility of improved diagnostic confidence, help to address potential medico-legal issues, or simply provide physicians and patients with greater peace of mind.
Contrast-enhanced MRI is fundamental to clinical practice and is faster and more efficient than many other diagnostic procedures. To perform an effective, individualized risk–benefit analysis, radiologists should be familiar with the differences in properties and associated risks among GBCAs and how these relate to agent selection for specific applications.
Enterline D, Karis J, Malinzak M, Porter KK, Sancrant J, Soto JA. Why Gadolinium Matters Today. Appl Radiol. 2021;50(1):supplement.
David Enterline, MD, FACR, Associate Professor of Radiology, Division of Neuroradiology, Duke University School of Medicine; John Karis, MD, Director of MRI & Brain Imaging, Barrow Neurological Institute; Michael Malinzak, MD, PhD, Assistant Professor, Duke University School of Medicine; Kristin K. Porter, MD, MR Modality Chief, University of Alabama at Birmingham; PhD; James Sancrant, DO, Chair, Clinical Operations, Triad Radiology Associates; Jorge A. Soto, MD, Chief of Radiology, Boston Medical Center.
