This study was approved by the Research Ethics Committee of the Reuth Rehabilitation Medical Center, Tel-Aviv, Israel and by the Research Ethics Committee of Soroka University Medical Center, Beer Sheva, Israel. All participants in this study signed an Informed Consent Form when they agreed to participate, and all methods were performed in accordance with the relevant guidelines and regulations.
The ClinicalTrials.gov identifier number is NCT02215590, 13/08/2014.
Subjects:
1. Participants post-ABI:
A total of 36 (7 females, 29 males; mean age 60.44±12.07 years) were enrolled in the study. Participants were approached using a database of hospitalized patients in Reuth Rehabilitation Medical Center, Tel-Aviv, Israel and using an add that was published in a local newspaper. After signing an informed consent, participants went through a screening process that included detailed neurological assessment of muscle strength and tone, sensation, reflexes, balance and coordination assessments, in addition to a cognitive assessment and medical history review by a neurologist. Inclusion criteria were: within an age range of 18-80 years; Residual dynamic balance impairment due to ABI; At least a year post-ABI (TBI or ischemic stroke) before recruitment; Ability to walk at least 10 meters with or without an assistive device; Drug therapy unchanged for one month prior to trial and during the entire trial period. Lastly, with a score above 19 on the Montreal Cognitive Assessment test (MoCA) [32]. Exclusion criteria were: Presence of an acute progressive neurological; systemic, or musculoskeletal disorder affecting gait and balance; Severe visual or hearing impairment; Pulmonary or cardiac condition impairing exercise endurance; Psychiatric disorders; Alcoholism or drug use.
2. Structural MRI Control groups:
Data of 36 aged-matched healthy participants, and 35 elderly participants with no neurological or psychiatric disorders and no brain damage was obtained with permission from the Cambridge Centre for Ageing Neuroscience (Cambridge Centre for Ageing and Neuroscience, 2019).
Experimental procedure:
The study included behavioral and MRI assessments. Each assessment was conducted twice, pre- and post-intervention. Between the assessments, participants underwent a rehabilitation program using the Re-stepTM technology (mechatronic shoes) [25, 26].
For the behavioral assessment we used the community balance and mobility scale (CB&M) in order to assess dynamic balance. This scale assesses difficulties in ambulation and balance skills which are needed for community integration in individuals with stroke and adults with TBI [33, 34]. The scale includes 13 tasks requiring multitasking and complex motor tasks (e.g., Unilateral stance, forward to backward walking, descending stairs, crouch and walk). Higher scores indicate better balance and mobility skills (maximum possible score=96). Furthermore, we used the 10-Meter Walking Test (10MWT) for participants post-stroke in order to assess self-paced gait velocity [35]. Participants were asked to walk a 14-meter track at a comfortable speed and then at a fast speed. The middle ten meters were timed. Time was measured by a handheld stopwatch. Velocity was computed based on the track length and time parameters. The 10MWT was added to the behavioral assessments after the begging of the experiment, and was therefore performed only on the stroke participants.
Outcome measures:
Resting-state functional connectivity, which is sensitive to brain network changes following brain injury [36], and brain volume measure for detecting structural brain changes following brain injury [37].
Interventional procedure:
22 sessions, given twice a week. Each session began with several warm-up exercises like mobilization and strengthening for 10 minutes, that were followed by training using the mechatronic shoes [25], for up to 40 minutes and ended with 10 minutes of cool-down exercises of stretching and relaxation walking (see, [26] for details about the task and the perturbation protocol).
MRI sessions and MRI acquisition:
Participants underwent two identical MRI sessions, pre- and post-intervention, each session included a 3D anatomical scan, a Rs-fMRI scan, fMRI-Localizer scan and a DTI scan (not reported here).
The time length of the 3D anatomy scan was four minutes and 50 seconds. This scan was acquired using a high resolution T1-weighted anatomical protocol with fast spoiled gradient-echo (FSPGR) sequence, with a voxel size of 1×1×1 mm, (Repetition Time (TR) = 8165 ms, Echo Time (TE) = 3.74 ms, 256×256 acquisition matrix). The field of view (FOV=192 mm) covered the entire cerebrum and the cerebellum.
The time length of the fMRI-Localizer scans was nine minutes and 50 seconds. The fMRI data was acquired using a gradient echo EPI with voxel size of 3×3×3 millimetres (mm), TR=2000 millisecond (ms), TE=35ms, flip angle=77°, 35 slices, with a 0.6 mm gap.
The fMRI-Localizer scan was a block design experiment containing six conditions (five movement blocks and one rest block); left or right limbs movement of dorsi/palmar flexion for palms and dorsi/plantar flexion for the ankles and bipedal ankle movements, (the limb moving frequency was equal to 1 hertz (Hz) as demonstrated by the experimenter before the scan). Movement blocks of 12 seconds were separated by resting periods of 10 seconds, which were cued by a fixation cross on a black screen. Visual cues for instructing hand and ankle movements were displayed on a screen during the experiment. Participants trained on the task before the scan using a dedicated apparatus located outside the scanner. The order of the blocks was random. In total, each localizer session included 25 movement blocks (five repetitions of each of the five movement conditions).
The time length of the Resting state scan was equal to seven minutes and 26 seconds. The resting state data acquisition parameters were similar to the fMRI-Localizer data scan. At the resting state session, a cross “+” was displayed in the middle of the screen, and participants were instructed to fixate on it during the scan.
The magnetic resonance imaging and data acquisition were performed at the imaging center of Soroka Medical Center using a 3-Tesla Philips Ingenia whole-body MRI scanner (Philips Ingenia, Amsterdam, Holland).
MRI data acquisition-healthy control group:
The control group data was taken, with permission, from the Cambridge Centre for Ageing Neuroscience (CamCAN) dataset
(Cambridge Centre for Ageing and Neuroscience, 10 January 2019, Cam-CAN Data Repository Cambridge, accessed 30 August 2019, <http://www.http://www.cam can.org/index.php?content=dataset>).
This data set was acquired on a 3T Siemens whole body MRI scanner (Siemens MAGNETOM TrioTim syngo MR B17). 3D anatomical scans were acquired using a high resolution T1-weighted anatomical protocol with a magnetization-prepared rapid acquisition with gradient echo (MPRAGE), (TR=2250 ms, TE=2.99 ms, FOV=256x249x192 mm, TI=900 ms, Flip angle (FA=9 deg); 1mm isotropic; GRAPPA=2; TA=4 min 32s).
Imaging analysis:
Functional, resting-state, and structural data were analysed by Brain Voyager 20.6 (Brain Innovation, Maastricht, The Netherlands).
Pre-processing of the localizer scan data:
Pre-processing included removal of the first two functional images of each run series to allow stabilization of the BOLD signal; correction of the slice scan time acquisition (ascending-interleaved; using a cubic-spline interpolation algorithm); head motion correction (using a trilinear/sinc interpolation) and a temporal high-pass filtering using a cut-off frequency of 2 sine/cosine cycles.
Pre-processing of the Resting-state scan data:
The pre-processing included removal of the first two functional images of each run series to allow stabilization of the BOLD signal; correction for slice scan time acquisition (ascending-interleaved; using a cubic-spline interpolation algorithm); a trilinear interpolation approach in order to remove head motions; a high-pass (GLM-Fourier) frequency filter with a cut-off value of 2 sine/cosine cycles and a low-pass Gaussian-Full Width at Half Maximum (FWHM) of 1.9 data points [38]. Further, we used band pass filtering with an 8th order Butterworth filter with cut-off frequencies of 0.009<f>0.08Hz [38]. Lastly, we projected out averaged signals of the white matter, the cerebro-spinal fluid, and head motion parameters. This step was conducted by running a multiple GLM regression analysis. The residuals of this analysis were free from the unwanted components and were used as inputs in the resting state functional connectivity (FC) Region of interest (ROI) analysis. Functional connectivity was computed using a correlation analysis (Pearson correlation coefficient) between the time courses of pre-defined ROIs (M1, cerebellum, thalamus, putamen, superior frontal, and superior parietal).
Both the fMRI-task and fMRI-rest data sets of each participant were spatially aligned onto the corresponding anatomical scan (T1 weighted structural scan) by an automatic alignment procedure (implemented in Brain Voyager 20.6). The results of the automatic alignment were inspected during the processing and manually adjusted if needed. Subsequently, the co-aligned images were transformed into Talairach space [39].
Regions of interest definition:
12 ROIs were examined: leg areas in M1 and cerebellum, thalamus, putamen, superior frontal, and superior parietal bilaterally. The M1 and the cerebellum were identified using the localizer scan (contralateral ankle movement vs baseline). The coordinates of each ROI were selected based on activation peaks of the above contrasts. The thalamus, putamen, superior frontal, and superior parietal were defined for each participant anatomically using FreeSurfer software V 5.0 (developed by the Laboratory for Computational Neuroimaging at the Athinoula A. Martinos Center for Biomedical Imaging at Massachusetts General Hospital in Charlestown in Boston, MA). The size of each ROI was defined as the number of functional voxels (3mm isovoxel). (Detailed characteristics (mean, SD, SEM) of each ROI are presented in Table 1).
Independent component analysis (ICA):
resting state data was analysed using the ICA method in Brain Voyager 20.6. This method allows the detection of a set of statistically independent spatial maps (networks) on a subject-by-subject basis during the resting-state scans and subsequently measures changes in the strength of these spatial maps after the intervention [40]. The analysis was composed of two stages: In the first, 30 ICA components were defined for each participant based on the individual resting-state scans. In the second, consistent components across participants were selected. We compared the averaged related activation across the networks pre- and post-perturbation training at the voxel level using the “self-organizing groups” feature in Brain Voyager. We minimized detection of false positives (type I errors) by using cluster-corrected familywise error rate correction at p < 0.05. We decided to focus our analysis on the sensorimotor network and the cerebellar networks that were shown to be sensitive to balance training [27].
Processing of the anatomical data for the volume analysis:
We analysed the 3D anatomical scan for each participant using Free-Surfer software 5.0 [41]. To quantify the differences in volumes, we computed the percent change with respect to the control group, according to the following formula:
Statistical analysis:
for statistical calculations, we used SPSS statistics (SPSS for Windows, Version 16.0; SPSS Inc., Chicago). The significance level was set to p<0.05 for all statistical tests. Normality assumption was tested by using the Kolmogorov-Smirnova test (p>0.05). Paired t-test was used for within-subject analyses of the behavioral data (pre-intervention vs. post-intervention). The ES (Cohen’s d) for the within-subject design was calculated by dividing the mean difference between pre and post-intervention by the pooled SD.
For the resting state functional connectivity analysis, we used z-Fisher transformation in order to normalize the distribution of the correlation coefficients. To check for the significance of the correlations between functional connectivity measures, brain volume parameters and functional behavioral measures, we used multivariate linear regression models. Whole-brain multiple comparison concerns were addressed by a cluster-correction for family-wise error rate at p<0.05. Lastly, for the brain volume analysis, we ran an independent t-test for data that did not deviate from a normal distribution and Mann-Whitney U test for data that deviated from a normal distribution.