Identifying carotid plaque inflammation using high and low molecular weight contrast agents
Jason Kraig Mendes1, Scott McNally1, Seong-Eun Kim1, Bradley D. Bolster2, Gerald S. Treiman3, and Dennis L. Parker1

1Radiology, University of Utah, SLC, UT, United States, 2Siemens Healthcare, SLC, UT, United States, 3Department of Veterans Affairs, SLC, UT, United States


Carotid plaque inflammation can be measured with dynamic contrast enhanced (DCE) MRI and is a marker for plaque instability. Despite this, DCE has not become a clinically viable tool in diagnosing carotid plaque instability and the corresponding stroke risk. The barrier to progress is a DCE protocol meeting requirements for clinical use to monitor medical treatment effect or failure. This project overcomes this barrier by developing a reliable and inclusive dual contrast DCE protocol to identify carotid plaque inflammation.


Dynamic contrast enhanced (DCE) imaging is useful in evaluating the functional status of a vascular system1. Gadofosveset is a clinically approved blood pool agent that binds to albumin resulting in a high effective molecular weight, only exiting the vessel lumen in areas of leaky neovessels and damaged endothelium2. Studies have shown that gadofosveset is better able to discriminate atherosclerotic vessel wall in rabbits than Gd-DTPA3 and symptomatic patients showed increased contrast agent uptake into carotid plaque as characterized by gadofosveset4. However, changes in adventitial vasa vasorum density and permeability may not be characterized by gadofosveset if the neovessels have not yet disrupted. As a result, Gd-DTPA may be better suited to track initial disease progression and therapy outcomes5 whereas gadofosveset may detect highly inflamed, end-stage atherosclerotic plaque. In this work we propose the use of dual contrast agent injections in conjunction with a 3D DCE radial stack of stars sequence6.


All imaging was performed on Siemens Trio and Prisma scanners with a 3D radial stack of stars sequence. Acquisition matrix was 192x304x12, resolution 0.8mmx0.8mmx1.5mm, TE/TR=2ms/5ms, 23 total measurements over 372s (16s per measurement). Data was reconstructed twice using the KWIC algorithm7. First the data is reconstructed into 16 temporal frames per measurement (1.3s effective temporal resolution) with the corresponding images being used to measure the AIF. The same data is then reconstructed a second time with a lower temporal resolution (4 temporal frames per measurement) to find the tissue contrast uptake curves. Following one minute of baseline measurement, 0.03 mmol/kg of Ablavar (Gadofosveset Trisodium, Lantheus Medical Imaging) was injected at 1.2 ml/s with a 20ml saline flush. Following another three minutes of measurements 0.1 mmol/kg of MultiHance (Gadobenate Dimeglumine, Bracco Diagnostics) was injected at 2ml/s with a 20 ml saline flush (Figure 1). All injections were performed with dual Medrad power injectors (Radiology Solutions, Bayer). Data was analyzed with Olea Sphere post processing software (Olea Medical). Image analysis


Evaluation of sequence improvements to DCE kinetic parameter estimation requires a reproducible set of tracer kinetic conditions. This was accomplished with an appropriate DCE phantom (Figure 3). Within the carotid plaque in Figure 4, there are areas of high ktrans as measured by low molecular weight Multihance along the juxtaluminal fibrous cap (red) and the adventitia (green). Ktrans measurements with the high molecular weight Ablavar are decreased in comparison, reflecting differences in permeability between the two agents. Figure 5 compares contrast agent uptake with intraplaque hemorrhage as identified with a MPRAGE scan on the bottom right. In this case, there is increased uptake of MultiHance but not of Ablavar near the center of the hemorrhage indicating inflammation but not necessarily neovessel rupture.


We have demonstrated feasibility of performing dual contrast agent injection for clinical evaluation of carotid disease. Two different molecular weight contrast agents may allow further characterization of plaque permeability in vulnerable carotid plaque. The long term effect of this project will be a more reliable and inclusive DCE protocol to identify carotid plaque inflammation.


This work was supported with resources from the George E. Wahlen Department of Veterans Affairs Medical Center (Salt Lake City, Utah) as well as funding from Ben B. and Iris M. Margolis Foundation, Cumming Foundation, Mark H. Huntsman Endowed Chair and Siemens Medical Solutions.


1. O'Connor et al. Br J Radiol. 2011; 84:S112-S120.

2. Caravan et al. J Am Chem Soc. 2002; 124:3152-3162.

3. Lobbes et al. Radiology. 2009;250:682-691.

4. Lobbes et al. Invest Radiol. 2010; 45:275-281.

5. Dong et al. Radiology. 2011; 260:224-231.

6. Block et al. JKSMRM. 2014; 18:87-106.

7. Song et al. Magn Reson Med. 2000; 44:825-32.


Figure 1: The dual injection protocol used two different molecular weight contrast agents with injections separated by a time to allow the system to reach steady state.

Figure 2: Radial stack of stars DCE data can be retrospectively reconstructed with different temporal resolutions. A short temporal resolution reconstruction is required to accurately determine the AIF while a longer temporal resolution reconstruction results in fewer streak artifacts for the tissue volume.

Figure 3: Two-compartment phantom with exchange compartment shown in (a), the perforated inner tube shown in (b) with an MRI cross-section of the complete phantom in (c). A representative AIF measured inside the perforated tube (blue line) and the corresponding exchange chamber contrast agent concentration (red line) are shown (d).

Figure 4: Ktrans calculated from dual contrast agents. A time-of-flight image is shown on the left with the yellow line indicating the position of the axial image and ktrans maps shown on the right.

Figure 5: Comparison of dual contrast agent uptake with the location of intraplaque hemorrhage. Difference in dual contrast agent uptake provides more complete information about carotid inflammation and disease progression.

Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)