Chengcheng Zhu1, Thomas Hope1, Henrik Haraldsson1, Farshid Faraji1, David Saloner1, and Michael Hope1
1Radiology, University of California, San Francisco, San Francisco, CA, United States
Synopsis
Abdominal aortic aneurysms (AAAs) with focal inflammation
(identified by USPIO uptake) have been reported to predict faster growth. Previous
2D T2* mapping method is limited by spatial resolution. This study evaluated 3D
high-resolution techniques (up to 1.3mm isotropic) for inflammation imaging of
AAAs. Experiments were preformed using both USPIO phantoms and in vivo patient
studies. We found the signal characteristics of 3D DANTE-SPACE images had good
agreement with T2* value drop, and it provided higher resolution and possible
information on USPIO concentration. Therefore, 3D high resolution methods may
help risk stratify patients with AAA disease by characterizing and quantifying
inflammation.Purpose
Abdominal
aortic aneurysms (AAAs) with focal inflammation have been reported to grow 3
times faster than those without
1, while their diameters were similar. Ultrasmall
superparamagnetic iron oxide (USPIO) particles can identify inflammation
(macrophages) in vivo using MRI. USPIOs induce both T1 and T2* shortening.
Previous studies have focused on the T2* effects, looking for signal voids or
quantitative mapping with T2*-weighted images. Such techniques, however, have
limited slice resolution (5mm)
1. We aim to evaluate higher resolution techniques for USPIO
imaging.
Methods
Sequences: Four MRI
sequences (Table 1) were evaluated at 3T: 1) Conventional 2D T2* mapping with
10 echoes; 5mm slice thickness; 2) T1 weighted 3D blood suppressed fast-spin
echo with variable flip angle train (DANTE-SPACE)
2: 1.3mm isotropic resolution, it is predominately T1 weighted
but also has T2* weighting given its long echo train; 3) T1 weighted 3D Volume
Interpolated Breath-hold Examination (VIBE, a gradient echo sequence):
1.3x1.3x2.6mm, very small T2* weighting; and 4) T1 weighted Ultra short TE (UTE,
3D radial acquisition): 1.3mm isotropic resolution, nearly ‘pure’ T1 weighting.
Phantom Study: A saline phantom with
a range of USPIO concentrations (0.94mg Fe/L to 600mg Fe/L) was made to
represent the possible concentrations of USPIO in vivo. Images were acquired
with four sequences and the T2* values and signal intensities were measured and
compared.
Patient Studies: 24
patients with AAA disease (all male, age 69±4.7, maximal diameter 4.3±1.0cm)
were imaged on a Siemens 3T Skyra scanner before and 2-3 days after USPIO
injection. T2* mapping and DANTE-SPACE were used in all patients. VIBE and UTE
were preliminarily evaluated in a few patients. Changes of T2* values and
signal characteristics in different sequences were compared.
Results
Phantom Study: MRI
signal strength is plotted relative to USPIO concentration for the four
sequences (Figure 1). Both SPACE and
VIBE show signal increase-and-drop curves, however with different peaks (7.5
versus 60mg Fe/L). UTE signal continually rises with increasing USPIO
concentrations, whereas T2* values drop. T2* mapping is not very sensitive at
either very low or very high concentrations, while 3D sequences provide
additional information.
Patient study:
Patients were classified as four types based on SPACE signal change (I: no
change; II: signal drop; III: signal enhance; IV: signal drop and enhance. Figure
2-4). Among all 24 patients, 16 were unchanged; 8 patients were observed T2*
drop, and all showed SPACE signal drop or enhance (drop only: 3; enhance only:
3; drop and enhance: 2). Of patients with USPIO uptake, 5/8 (62.5%)
subsequently had repair for progressive AAA disease, and only 2/16 (12.5%) had
surgery in the patients group without sign of USPIO uptake.
Discussion
There was good agreement between SPACE
signal increase/decrease and a decrease in T2* value. SPACE signal drop likely indicates
high or intra-cellular concentration of USPIO, while signal enhancement likely indicates
relatively low concentration of USPIO. These findings suggest that SPACE can be used for
intracellular USPIO imaging, which has been associated with fast AAA growth
1 , with the advantage of higher spatial resolution than conventional T2*
imaging. A
larger scale longitudinal study with histology validation is needed to confirm
these findings. Signal change characterization can also be readily implemented
in the clinical routine. The UTE technique is particularly appealing given its
linear relationship with increased USPIO concentration permitting quantification
of inflammation severity (Figure 5). Positive contrast methods are more robust
than negative contrast methods, and are less sensitive to imaging artefacts. 3D
radial acquisition of UTE sequence is also insensitive to motion. The potential
of UTE sequence will be evaluated in future patient studies. Additional
patients will be recruited and all patients will be followed with serial
imaging and a relationship to clinical outcome will be sought.
Conclusion
3D
high resolution MRI can be used to identify USPIO uptake within AAA, with good
agreement with traditional 2D T2* mapping method. These methods may help risk
stratify patients with AAA disease by characterizing and quantifying
inflammation.
Acknowledgements
This study is supported by NIH grants R01HL114118, R01NS059944 and R01HL123759.References
1. Richards, J.
M. et al. Abdominal aortic aneurysm
growth predicted by uptake of ultrasmall superparamagnetic particles of iron
oxide: a pilot study. Circ Cardiovasc
Imaging 4, 274-281,
(2011).
2. Zhu, C. et al. Isotropic 3D Black Blood MRI of
Abdominal Aortic Aneurysm Wall and Intraluminal Thrombus. Magn Reson Imaging, doi:
10.1016/j.mri.2015.10.002 (2015).