John Oshinski1, Soroush Heidari Pahlavian2, Jordan Schuster 1, Xiaodong Zhong1,3, Rouzbeh Amini2, Francis Loth2, and Daniel Barrow1
1Radiology and Imaging Sciences, Emory University, Atlanta, GA, United States, 2Mechanical Engineering, University of Akron, Akron, OH, United States, 3Research and Development, Siemens Healthineers, Atlanta, GA, United States
Synopsis
The goal of
this study was to acquire mid-sagittal cine DENSE images to quantify
displacement and strain over the cardiac cycle in patients with Chiari
malformation and age-matched controls. The
major finding of this study was that both tissue displacement and tissue strain
in the parenchyma of the cerebellum was significantly greater in patients with
Chiari malformation than healthy controls.
Introduction
Brain tissue
undergoes pulsatile motion due to cardiac-driven pressure/flow variations over
each heartbeat. Altered brain motion has
been documented in certain pathologies, including intracranial brain tumors and
Chiari malformation (1,2). Recently,
cardiac-gated, cine, balanced steady state free procession (bSSFP) has been
used to examine cerebellum motion in subjects with Chiari malformation (3). Measurement
and examination of cerebellar tonsillar motion using conventional bSSFP cine imaging
is limited to tracking pre-defined anatomical landmarks over the cardiac cycle
and cannot quantify displacement information over the brain parenchyma. Therefore,
bSSFP is unable to quantify tissue strain
which has been shown to be important in cellular signaling and tissue
remodeling. Spiral cine imaging coupled
with a Displacement Encoding using Stimulated Echoes (DENSE) sequence offers a
method to quantify very small levels of brain tissue displacement on a
pixel-by-pixel basis over the entire imaging slice (4,5). The goal of this study was to acquire mid-sagittal
cine DENSE images to quantify displacement and
strain over the cardiac cycle in patients with Chiari malformation and
age-matched controls. We hypothesized that peak displacement and
peak strain in the cerebellum and cerebellar tonsils would be higher in patients
with Chiari malformation than in controls.Methods
We examined 9 adult healthy volunteers (age 28+/-4) and 12 patients (age range 31+/-6) with Chiari malformation. The Chiari diagnosis was confirmed by previous
clinically-ordered imaging study (> 5m mm of tonsillar decent below the
foreman magnum) and by clinical symptoms. All patients were referred by a single
neurosurgeon (DB) and were under consideration for decompression surgery. All patients
gave informed consent for a separate imaging study that included a cardiac-gated,
spiral cine DENSE sequence performed at the mid-sagittal plane on a 3T MRI scanner
(Siemens Prisma, Erlangen, Germany). Acquisition
parameter included: two directions of
motion encoding, encoding frequency, ke = 0.6-1.2 cycles/mm, two spiral
interleaves per heartbeat, pixel size = 1.2 × 1.2 mm2, slice
thickness = 7 mm, averages = 4, frames = 16-27, depending on the heart rate.
Using
an in-house written MATLAB program (DENSEpro), regions of interest (ROIs) were traced
in multiple brain regions (including the cerebellum) on the magnitude images
and copied to the phase images, figure 1.
For each ROI on each participant, displacement magnitudes were spatially averaged
for each cardiac frame. The frame
with the largest average displacement was defined as the peak displacement
in each subject. Principal strain was
calculated using a structured quadrilateral grid on the two-dimensional displacement
field. Peak principal strain was determined for each subject in same
manner outlined for peak displacement. Peak displacement
and peak strain were compared between the healthy controls and the Chiari
patients using a student’s t-test. Spatial
variation of tissue displacement over various brain regions was assessed using Lagrangian
pixel-wise trajectories (6). Results
Spiral
cine DENSE images were successfully acquired in all volunteers and all Chiari
patients. Peak tissue displacements in the cerebellum
were significantly higher in the Chiari
patients than in the healthy controls (0.19 ± 0.08 mm vs 0.09 ± 0.04 mm,
p=0.004). Similarly, Peak principal strain
values in the cerebellum were significantly
higher in the cerebellum in the Chiari patient compared to the health subjects
(0.77% ± 0.33% vs 0.35% ± 0.10%, p=0.002), figure
2. Spatial variation of displacement
and principal strains with the cerebellum were assessed using pixel-wise
trajectories and strain maps, respectively. Figure 3 shows this analysis for a
representative healthy control and a Chiari patient. The cerebellar tonsils had
noticeably more anterior-posterior motion compared to more superior tissues. The
spatial gradient of displacement resulted in localized elevated strain values
within the cerebellum in both healthy and patient cases, but strain values were
markedly larger in Chiari patients than in controls.Discussion
The
major finding of this study was that displacement and strain of brain
parenchyma in the cerebellum was significantly greater in patients with Chiari
malformation than in healthy controls. The
increased strain value was often characterized by localized increase in strain values
in the inferior/posterior regions of the cerebellum. This finding is noteworthy as this sub-occipital
anatomical region is often cited by Chiari patients as a region of pain during
coughing, sneezing, or laughing. Further
studies will examine temporal characteristics of motion, the motion and strain
in other brain regions, and examine motion patterns in patients after
decompression surgery.Conclusion
We
have demonstrated that cardiac-gated, spiral cine DENSE imaging can measure spatially
and temporally detailed variation of motion patterns in the brain and these
motion patterns can be used to determine tissue strain values. In patients with Chiari malformation, both
strain and motion were greater than in healthy controls.Acknowledgements
No acknowledgement found.References
1.
Hofmann, E., et al. Amer J of Neuroradiology, 2000. 21(1): p. 151-158
2.
Leung V, et al. J Neurosurg Spine. 2016
Apr;24(4):546-55.
3.
Radmanesh A, et al. Neuroradiology, 2015 Apr;57(4):387-93.
4.
Soellinger, M., et al. Magn Reson Med, 2009. 61(1): p. 153-62.
5.
Zhong, X., et al. Medical physics, 2009. 36(8): p. 3413-3419.
6. Spottiswoode
BS, et al. J Magn Reson Imaging. 2008 May;27(5):1019-27.