Silvia Mangia1, Alena Svatkova2,3, Peter Bednarik1,3, Igor Nestrasil2, Lynn E. Eberly4, Adam Carpenter5, and Shalom Michaeli1
1Radiology, CMRR, University of Minnesota, Minneapolis, MN, United States, 2Department of Pediatrics, University of Minnesota, Minneapolis, MN, United States, 3Central European Institute of Technology, Masaryk University, Brno, Czech Republic, 4Division of Biostatistics, University of Minnesota, Minneapolis, MN, United States, 5Neurology, University of Minnesota, Minneapolis, MN, United States
Synopsis
Rotating frame MRI methods including adiabatic
T1ρ, T2ρ, and RAFF4 were here
employed for characterizing the white matter (WM) of relapsing-remitting Multiple
Sclerosis (MS) patients. We calculated relaxograms from subcortical WM of MS
patients (excluding lesions) and age-matched controls, and compared them with
histograms of DTI outcomes. T1ρ, T2ρ and TRAFF4 were significantly different in the WM of MS
patients vs controls, while DTI outcomes did not detect group differences. These
findings are supported by recent validation studies using demyelination/dysmyelination
animal models, where RAFF4 exhibited exceptional ability to probe myelin
content/integrity which we attribute to enhanced sensitivity to slow/ultra-slow
motion.Introduction
Whereas
quantification of myelin content is crucial for both diagnostic and monitoring
purposes in a variety of diseases, a non-invasive tool for myelin quantification in
vivo is underdeveloped. Our group has pioneered novel relaxation methods
for generating unique types of MRI contrast based on frequency swept
radiofrequency (RF) pulses, which have been shown to detect tissue pathologies
in various diseases including stroke, Parkinson’s disease and multiple sclerosis.
In particular, we have developed a novel rotating frame relaxation
method, entitled RAFFn (Relaxation Along a Fictitious Field
in the rotating frame of rank n) [1], which produces flexible
MRI contrast with low specific absorption rate, and could be particularly
sensitive to slow/ultra slow motion in the millisecond time scale. Using animal
models in vivo and ex
vivo, we have demonstrated that relaxation
time constants measured with RAFF4 and RAFF5 are exceptionally sensitive to myelin
content and integrity [2, 3], which we attribute
to the ability to detect conformational exchange of the methylene groups in myelin
sheaths. It had been shown that the high rotating frame configuration of the RAFFn
technique is uniquely sensitive to slow motional processes, and provides a better marker of myelin than other MRI
methods such as magnetization transfer (MT), T1, T2, T1ρ or
T2ρ [3]. In this work, using clinical
3 Tesla Prisma platform we have extended our previous findings where we have
shown that the T1r measurements in relapsing–remitting
MS (RRMS) patients at 4T [4] could detect abnormalities
in the normal appearing white matter (NAWM) of MS patients. We thus embarked on
the study of RRMS patients using our novel relaxometry modalities, and we
investigated how pathological abnormalities in the white matter of MS patients are
detected using T1ρ, T2ρ
and RAFF4. We have also compared these MRI metrics with the outcomes of DTI, i.e.,
mean diffusivity (MD) and fractional anisotropy (FA).
Methods
All studies were conducted on a 3 T/ 90 cm bore, Siemens Prisma console,
64-channel receive system. Four RRMS individuals (49±10 y/o) and four control subjects
(54 ± 5 y/o) participated in this study. For T1ρ, T2ρ and RAFF4
acquisitions, 30 slices were acquired
with segmented GRE readout (4 segments), voxel size: 1.6x1.6x3.6 mm3,
GRAPPA=3, TE=3.18 ms; TR=2 s. The parameters of the preparation pulses were used
as previously described [1, 5-8]. For DTI, parameters
were: 128 directions, with 5
additional non-diffusion weighed (b0) images, b-value=1500s/mm2,
voxel size 1.8x1.8x1.8mm3, TR=2820 ms, TE=72.6 ms; multi band (MB)=4,
80 slices. Brain segmentation of the high resolution T1-weighted images was
carried out with FreeSurfer (FS) version 5.3.0. Rotating frame MRI and DTI
metrics were then co-registered to the T1-weighted images, and histograms were
extracted from the segmented subcortical WM. Modes of the histograms were then estimated
per subject, and compared among subject groups with unpaired two-tailed
t-test.
Results and
Discussion
Representative parametric maps are shown in Fig. 1, while the results of
relaxation and DTI measurements are summarized in Table 1. The T1ρ, T2ρ and RAFF4 relaxation
time constants of the WM exhibited significant differences between MS patients and
control subjects. Greater differences between MS and controls were observed with
T1ρ (p=0.003) and T2ρ (p=0.001) as compared to RAFF4 (p=0.022). Interestingly,
no statistically significant differences in the WM between MS patients and
controls were detected with MD and FA. Our recent animal studies using healthy
rats [3] and mice with mucopolysaccharidosis I (MPS I) [2] have demonstrated that RAFF4 is sensitive
predominantly to myelin content and integrity. This was attributed to its enhanced
sensitivity to slow / ultra slow motion. Because MS is characterized by several
processes such as demyelination, axonal degeneration, gliosis and neuroinflammation,
each of which are characterized by different motional regimes, the rotating
frame relaxation contrasts have the potential to be sensitive to these
pathological processes by “tuning” to the relevant regime of motion.
Conclusion
We have used several MRI
modalities, i.e., adiabatic T1ρ and T2ρ, RAFF4 and DTI, to investigate their ability to detect abnormalities
of the white matter in MS. As opposed to DTI outcomes, each of the rotating
frame MRI techniques detected group differences in the white matter of MS
patients vs controls, possibly providing sensitivity to different
aspects of the MS pathology. In agreement with our previous findings in animal
models, RAFFn detected abnormalities in the WM of MS patients which we
attribute to the sensitivity of RAFFn to myelin.
Acknowledgements
NIH grants: P41 EB015894, P30
NS076408, UL1TR000114References
[1]
Liimatainen et al. Magn Reson Med 2015;73:254-62. [2] Satzer et al. PLoS One
2015;10:e0116788. [3] Hakkarainen et al. Magn Reson Med 2015, in print. [4]
Mangia et al. Mult Scler 2013; 20:1066-73. [5] Liimatainen et al. J Magn Reson
2011; 209:269-76. [6] Michaeli et al. Magn Reson Med 2005; 53:823-9. [7]
Michaeli et al. J Magn Reson 2004;169:293-9. [8] Michaeli et al. J Magn Reson
2006;181:135-47.