Improving Diagnosis and Safety: An Inside View of Recent Advances and Safety Developments in High-resolution Vascular and Cardiac MRI

This report is based on medical evidence presented at sanctioned medical congress, from peer reviewed literature or opinion provided by a qualified healthcare practitioner. The consumption of the information contained within this report is intended for qualified Canadian healthcare practitioners only.

MEDICAL FRONTIERS - 16th Scientific Meeting and Exhibition of the International Society for Magnetic Resonance in Medicine

Toronto, Ontario / May 3-9, 2008

Figure 1.

In recent years, magnetic resonance angiography (MRA) has become a reliable diagnostic tool used in a variety of indications, including imaging of cerebral, thoracic, abdominal, renal and peripheral arteries. Several different techniques can be used to obtain MR images, but by far the most common approach is contrast-enhanced (CE)-MRA. Currently, CE-MRA is the preferred modality to image various kinds of vascular disease, especially peripheral artery disease (PAD) and its pre-interventional work-up.

Gadolinium (Gd) chelates are commonly used to improve signal and tissue contrast in MR imaging (MRI), producing relatively high-quality images, although not as precise as digital subtraction angiography (DSA). This is because until now, all Gd-based contrast media injected to enhance the vasculature rapidly diffused into the extracellular space following injection, allowing for only a brief first-pass or dynamic imaging during the injection itself. Unlike extracellular contrast media, as the first high-relaxivity blood-pool contrast agent developed, gadofosveset trisodium remains in the intravascular space for an extended period of time, allowing radiologists to image the vasculature not only on first-pass but also during an extended phase or steady-state that can last for more than 60 minutes
e 1).

Table 1.


Differences between the high-relaxivity blood-pool contrast medium and extracellular contrast agents have potentially important clinical implications, as was demonstrated during this year’s ISMRM meeting. “As clinical radiologists, we have to make a diagnosis first before we treat the patient and that diagnosis of course must be accurate, so it is really a consequence of clearly seeing what is on the image and then deciding what to do later on,” stated Dr. Winfried Willinek, Department of Radiology, University of Bonn, Germany. Single Injection, Steady-state Imaging

Images obtained with gadofosveset in the steady-state allow radiologists to see clearly far more than they can on first-pass imaging and help them make the diagnosis with the same degree of confidence as they might have using conventional angiographic techniques. For example, with a blood-pool contrast agent, radiologists are able to perform ultra-high-resolution, steady-state scans that allow them to visualize distal outflow vessels in the legs and confirm that a distal vein is suitable for bypass surgery, sparing the patient with critical limb ischemia the prospect of amputation. “The long intravascular distribution also means we are not limited to just one bed, but can image the arterial and the venous system at the same time,” Dr. Willinek told delegates.

The new contrast agent lights up the vessels for an extended period of time, allowing radiologists to image more than one area of the body following a single injection. If images obtained on first-pass contrast enhancement are unsatisfactory, gadofosveset provides a “salvage” opportunity to go back and image different parts of the vasculature without requi
, as would be necessary with other Gd-based agents.

Figure 2.


“Any area where you need to look at very small vessels—whether it is a large vessel that has become narrowed or some of the smaller vessels more peripherally—steady-state imaging allows you to get clarity of imaging,” confirmed Dr. Alan Moody, Professor of Medical Imaging, University of Toronto, Ontario, in an interview. This was amply demonstrated by Dr. Willinek where images obtained during the “add-on” or steady-state phase of gadofosveset-enhanced imaging were of significantly better quality than first-pass images. Radiologists could, for example, track narrow lumen in the centre of the vein that first-pass imaging would not visualize at all.

The positive predicted value on first-pass imaging based on a prospective analysis of 27 subjects with PAD was approximately 75% whereas it was 100% on steady-state imaging with gadofosveset, as confirmed by DSA carried out in all 27 patients. “Steady-state imaging allows physicians to plan treatment more accurately and identify target lesions for intervention,” Dr. Willinek remarked.

Simultaneous Imaging of Arterial and Venous Beds

Another advantage of this particular contrast agent is that it allows radiologists to look at the venous syste
ial system, where it potentially could replace ultrasound to diagnose deep-vein thrombosis (DVT).

Figure 3.


In a prospective study involving 139 subjects, Dr. Guido Kukuk, University of Bonn, and colleagues assessed MR phlebography findings as determined during the steady-state phase of gadofosveset-enhanced MRA. All patients had suspected or known PAD. In 46 out of 139 subjects (33%), MR phlebography provided additional information which was confirmed on colour duplex sonography.

As was reported here, incidental DVT was found in three patients, varicosis in 42 (30% of the group), and an arteriovenous fistula was diagnosed in one individual. MR phlebography allowed for mapping of veins suitable for bypass surgery in all 43 eligible patients or 100%, inves
add-on to peripheral CE-MRA, MR phlebography with a blood-pool contrast agent allowed identification of incidental but relevant venous disease.

Figure 4.



3D Phase Contrast MRA

Dr. Jelena Bock, University of Freiburg, Germany, presented promising findings on the utility of 3D phase contrast (PC)-MRA of the thoracic aorta. As Dr. Bock told delegates, time-resolved PC-MRA allows radiologists to measure three directional velocities within vascular systems of interest. These data can then be used to analyze blood flow as well as vascular geometry by PC-MRA.

Relaxivity and Injection Rate

As discussed by Dr. Henrick Jacob Michaely, Section Chief, Vascular MRI, University of Heidelberg, Mannheim, Germany, one way to increase the signal-to-noise ratio without decreasing image quality is to use a contrast agent with a high relaxivity such as gadofosveset (relaxivity is the ability of a contrast agent to generate signal, which is necessary to obtain contrast between two different tissues, such as blood vessels vs. the background). “Once injected into the vessels, gadofosveset has four to five times higher relative relaxivity than standard extracellular contrast agents,” he indicated.

Pre-clinical data indicate that approximately 75% to 85% of the agent binds to albumin once injected, but it is not clear how quickly this binding takes place, Dr. Michaely noted. What is known is that non-bound molecules have similar relaxivity as conventional extracellular contrast agents, “so there can be a maximal enhancing effect if sufficiently high amounts of gadofosveset bind to serum albumin in a sufficiently fast time,” he added.

Investigators therefore compared three injection rates to determine if there were any differences in signal enhancement among 21 volunteers. Group 1 received the contrast agent at a rate of 1 mL/sec, group 2 at a rate of 2 mL/sec and group 3 at a rate of 4 mL/sec. All subjects were given a standard dose of gadofosveset at 0.03 mmol/kg body weight.

As Dr. Michaely reported, 20 out of the 21 MRA scans yielded at least a “good” image quality with a median score of four in all three groups (in one subject, the sequence was started too late so the image was not captured). The minimum score in exams which were diagnostic was at least three, he added, and no differences in image quality were observed between the three injection rate groups. Image intensity was “slightly higher” with the fastest injection and there was a trend towards faster enhancement with the highest injection rate, but overall there were no statistically significant differences in the quality of the image obtained using the three different injection rates. “We can conclude that first-pass MRA is feasible with a good image quality regardless of the injection speed, so if you have a patient with limited breath-hold time, the faster injection rate may be the more appropriate rate,” Dr. Michaely concluded.

Economic Analysis of Imaging Modalities

Mr. Bastian Hass, IMS Health, Nuremberg, Germany, and colleagues investigated the cost-effectiveness of a diagnostic imaging strategy using MRA enhanced with a blood-pool agent vs. strategies starting with either MRA enhanced with extracellular agents or standard DSA. He presented his findings at the International Society for Pharmacoeconomics and Outcomes Research (ISPOR) 13th Annual International Meeting held concurrently in Toronto.

All diagnostic modalities were analyzed in the context of severe peripheral arterial occlusive disease in Canada. As investigators noted, the aggregated mean costs per initial diagnostic modality as well as incremental costs per quality-adjusted life-year (QALY) gained were evaluated.

Results indicated that the mean overall cost of using gadofosveset-enhanced MRA was $7814 (CDN). In comparison, aggregated costs using eithe
DSA were $8637 or $9842, respectively (Table 2). “Imaging strategy with initial gadofosveset-MRA is less costly than strategies initially using standard MRA or DSA,” investigators concluded.

Table 2.


The group then calculated the additional cost per QALY gained by standard MRA vs. gadofosveset-enhanced MRA as well as vs. that for standard DSA. By using a threshold of $50,000—the current accepted cost of purchasing a QALY in Canadian funds—they found that standard MRA is above the $50,000 threshold over 90% of the time, while DSA was beyond the threshold over 96% of the time. “The focus of this study was on cost-effectiveness and if you want to focus on cost-effectiveness, results are very promising,” Mr. Hass said in an interview.

Cardiac MRI

According to Dr. Matthias Friedrich, Associate Professor of Medicine, University of Calgary, Alberta, the various imaging techniques all have their strengths and weaknesses, but the aim of all diagnostic imaging is to move the time of diagnosis to an early stage of disease and, ideally, prevent it. For example, computed tomography (CT) scans offer the best images from a morphological point of view, but they are not suitable for characterizing tissue pathology, function or metabolism. “Nuclear medicine has an opposite profile, as it’s good at looking at cardiac metabolism but bad with morphology,” he remarked. In contrast, cardiovascular magnetic resonance (CMR) is a “good global player,” its particular strength being tissue pathology.

Noting that the Gd-based contrast agents “are still among the safest contrast agents we have,” Dr. Friedrich presented multiple images obtained with several different Gd-based contrast media including Gd-DTPA and gadobutrol, a second-generation macrocyclic MR agent (Figure 4). The images presented affirmed the utility of CMR for assessing cardiac morphology, function, flow, vessels, lumen, tissue composition and metabolism—all in one imaging study. For example, “no one would argue that CMR is the gold standard for left ventricular function analysis,” Dr. Friedrich commented.

At the same time, CMR clearly identifies perfusion deficits, atrio-ventricular stenosis, myocarditis, cardiomyopathies, myocardial infarction (both extent of tissue damage and whether it is reversible), fibrosis and scar tissue. “The beauty of CMR is that you can use the same scanner a few seconds later to produce a totally different image with different imaging parameters or different scanner settings,” Dr. Friedrich stated, “so it is a very powerful tool to answer a lot of questions in a single patient with a single exam.”

Update on Gd Chelates and Nephrogenic Systemic Fibrosis

One of the other appealing features of the first commercially available blood-pool contrast agent is that the dose needed to achieve good images is about tenfold lower than the dose required when imaging with first-generation extracellular agents. This may reduce the risk of exposing patients to Gd ions that are thought to cause the rare but serious condition referred to as nephrogenic systemic fibrosis (NSF). As explained in an interview with Dr. Tim Leiner, Assistant Professor of Radiology, Maastricht University Hospital, The Netherlands, each of the currently available contrast agents have varying degrees of kinetic stability. This makes it easier for some Gd-based contrast agents to break free of the chelator in patients with poor renal function, who are less able to eliminate Gd, after which it is deposited in tissues.

The least stable of the Gd-containing agents appear to be those that are non-ionic linear chelates, such as gadodiamide and gadov
ich have lower stability values than ionic linear chelates such as Gd-DTPA. Macrocyclic chelates, such as gadobutrol, differ significantly from linear chelates and have the highest complex stability of the Gd-containing contrast media.

Figure 4.


That the chemical structure of the various Gd compounds appears to influence their subsequent deposition in tissues was demonstrated by diagnostic imaging researcher Dr. Martin Sieber, Berlin, Germany (Eur Radiol 2008;Epub ahead of print). In renally impaired rats, Gd concentration in the skin was measured after the animals received either gadodiamide, gadoversetamide, Gd-DTPA or gadobutrol. Skin concentrations of Gd were then compared to those in healthy rats.

For non-ionic linear compounds, prolonged circulation times seen for all of the compounds in the renally impaired group led to higher concentrations of Gd in the skin compared with the linear ionic compounds, where the increase in concentrations in the skin was less pronounced. No long-term Gd retention in the skin was observed for any of the animal models following administration of the macrocyclic compounds.

In a related experiment, the same group evaluated possible long-term retention of Gd in the skin of rodents, again following administration of different Gd-based contrast agents. Here again, they observed statistically significant differences in Gd concentrations in the skin between the different classes of compounds. For linear non-ionic compounds, concentrations were detected in the skin for up to 250 days following administration, whereas for the linear ionic compounds, a relatively lower level of Gd retention was observed over time. Some 40 days after the last injection (all animals received five days of treatment with the contrast agent), values in the skin after administration of macrocyclic compounds approximated those in animals who received only saline or nothing at all.

Dr. Andrea Loewe, Berlin, pointed out that very limited numbers of NSF episodes have been reported with Gd-DTPA despite over 90 million applications of its use, and almost all of them have been reported in the U.S. A highly methodical search of the Northern California Kaiser Permanente database involving all patients with a creatinine level in excess of 1.8 mg/dL who had received gadopentetate as the sole Gd-based contrast agent (the only one used in the entire medical system) also found that the risk of developing NSF in patients with renal disease is “markedly lower” than has been previously reported.

Indeed, as noted by Dr. Thomas Hope, Kaiser Permanente, San Francisco, based on a thorough search of the Northern California database, the incidence of NSF was between 0% and 0.03% in a predialysis population and between 0.14% and 0.4% in dialysis patients and this included any Gd contrast MR using gadopentetate, not only MRA. Currently, no confirmed episodes of NSF have been seen in association with the sole and exclusive administration of gadobutrol, gadofosveset, gadoxetic acid or gadoteridol. Furthermore, as Dr. Leiner emphasized, “There has been a sharp drop in the number of reported cases in 2007 compared to 2006, probably reflecting altered practice patterns.” Still, “it is paramount that these agents be used with caution in patients with renal insufficiency,” he added.

It is also recommended that patients perform more frequent manual exchanges or additional automated peritoneal dialysis cycles for at least 48 hours after administration of a Gd contrast agent. “If you use CT and inject iodinated contrast agents, the risk of acute kidney failure is much greater than the risk of NSF from Gd-based contrast media,” Dr. Leiner cautioned. “So concerns about NSF should not lead to unjustified withholding of the procedure. MRA and MRI are still the procedures of first choice in many patients, even after having weighed the pros and cons of injecting contrast. If you use macrocyclic agents for patients with chronic kidney disease stage 4/5, there seems to be no problem.”


The utility of CMR is well established and well served by second-generation extracellular agents. CE-MRA with extracellular agents does not provide the same resolution as conventional digital angiography. With the introduction of blood-pool agents such as gadofosveset, ultra-high-resolution imaging is achievable. Because gadofosveset binds to albumin, it circulates much longer in the blood compared with extracellular agents. The high relaxivity and longer residence time of this intravascular agent is an important new advance that allows radiologists to obtain higher resolution images—tenfold to 20-fold better resolution than other comparable agents—during an extended steady-state phase. Images obtained using this novel contrast agent during MRA already rival the detail and accuracy of those obtained on standard angiography and radiologists are still exploring the opportunities provided by steady-state imaging with a blood-pool agent.

By providing radiologists with a highly accurate picture of pathologies in both vascular beds, it is certain that gadofosveset-enhanced MRA could become a very important non-invasive tool for the diagnosis and pre-interventional workup of patients with PAD, among other vascular anomalies, and significantly reduce the need for invasive and expensive diagnostic imaging modalities.

We Appreciate Your Feedback
Please take 30 seconds to help us better understand your educational needs.
Are you a healthcare professional practicing in Canada?
What type of healthcare professional best describes you?
Did you find this publication to be of educational value?
What is you preferred format for receiving and reading a report?
Would you like to have video available with the report?
Suggested future topics: