Abstract
Poster 1
A Pilot RCT to Test the Effect of Lokomat-Applied Force Fields on Functional Walking Skills in People With Motor-Incomplete Spinal Cord Injury
1University of British Columbia, Vancouver, BC, Canada
2International Collaboration on Repair Discoveries, Vancouver, BC, Canada
3G F Strong Rehabilitation Centre, Vancouver, BC, Canada
Poster 2
Day-Long Movement Monitoring to Differentiate Healthy From Impaired Infant Development
1Oregon Health & Science University, Portland, OR, USA
2APDM, Inc, Portland, OR, USA
Poster 3
Quantifying the Feasibility of Pairing tDCS With Standard Physical Therapy Following Neurological Insult, Using Clinical and Robotic Assessment Tools
University of South Carolina, Columbia, SC, USA
Poster 4
Activity-Dependent Mechanisms of Neural Circuit Plasticity During Limb-Overuse Following Ischemic Stroke
UCLA, Los Angeles, CA, USA
Stroke is the number one cause of adult disability in the United States, but there are no current medical treatments. Previously published work in the lab identified a molecular growth program that is triggered in the peri-infarct cortex after stroke, a process that promotes axonal sprouting in the surviving brain. The current studies indicate that new neuronal connections form after stroke in an activity-dependent manner. This finding is especially germane in light of the 2006 EXCITE clinical trial, which found that stroke patients who engage in constraint-induced movement therapy demonstrate significant and lasting motor improvements. We have developed a novel limb-overuse paradigm that is analogous to human constraint-induced movement therapy. Using this model, we hypothesize that a specific ensemble of genes drives circuit rewiring during poststroke limb overuse, a clinically relevant rehabilitative therapy that converges injury and activity-dependent molecular processes. Preliminary neuronal tracing data suggest that recovery from forelimb motor stroke involves the formation of new circuits between the premotor areas and cortical areas located posterior and medial to the caudal forelimb region, including trunk motor areas and retrosplenial cortex. The cells that comprise these stroke repair circuits have been labeled with a fluorescent neuronal tracer, specifically isolated by FACS, and are undergoing RNA sequencing to generate an “activity-dependent transcriptome.” In parallel ongoing studies, a number of stroke-induced and activity-regulated candidate gene systems are being screened in vitro for potential roles in axonal sprouting. Selected candidates that increase neuronal sprouting will advance to in vivo genetic manipulation within the premotor circuits mapped during poststroke limb overuse.
Poster 5
Frontal-Subcortical Function and Use of Assistive Speech Devices for Aphasia
1UMDNJ, Newark, NJ, USA
2Kessler Foundation, West Orange, NJ, USA
3The Stroke Comeback Center, Vienna, VA, USA
4Kessler Institute of Rehabilitation, Saddlebrook, NJ, USA
Assistive speech devices (ASDs) may augment communication efficacy in aphasia. However, patients may have difficulty using a mobile device, which requires diverse mental abilities, including cognitive motor planning, concentration, and sequencing. We wished to evaluate whether performance on specific neuropsychological tests could predict device use success. To this end, 20 people with aphasia (60.6 ± 14.3 years old) completed 9 tasks, including Finger Tapping Test (FTT), aphasia-adapted Frontal Assessment Battery (FAB), Test of Oral and Limb Apraxia (TOLA), Western Aphasia Battery (WAB), Behavioral Inattention Test (BIT), Communicative Effectiveness Index, Catherine Bergego Scale, Naturalistic Action Test, and Neuropsychological Assessment Battery (NAB). All patients were trained to use an ASD (O’Brien Technologies Survivor Speech Companion System) followed by 7 days’ unrestricted home use. Then, each participant completed 3 device-based communication tasks (tell marital status, give directions to Kessler Foundation, and ask for water). We performed a Stepwise Discriminant Function Analysis to evaluate how pretrial tests performed in grouping high (>80%) versus low (<80%) scorers. FAB was the only significant predictor of device use success (Wilks’s λ = 0.713, F = 6.856, P = .018, 28.7%variance), correctly classifying 80% of high/low scores. A factor analysis indicated that a factor that included FAB, WAB, BIT, NAB, and TOLA explained 45% of variance (eigenvalues > 1). Identifying people with aphasia who can benefit from an ASD is vital to prescribing an ASD. Here, aphasia severity was not predictive, but an aphasia-adapted version of the FAB predicted device use success. These results show that frontal cognitive assessment may be needed in standard ASD assessment. Further research to identify the relation between skills assessed by FAB and neuropsychological deficits in aphasia is indicated.
Poster 6
Potential Therapeutic Effects of Sensory Tongue Stimulation Combined With Task-Specific Therapy in People With Spinal Cord Injury
1University of British Columbia, Vancouver, BC, Canada,
2International Collaboration on Repair Discoveries, Vancouver, BC, Canada,
3British Columbia Institute of Technology, Vancouver, BC, Canada
Poster 7
Does Paired Stimulation of Cutaneous and Proprioceptive Receptors Enhance Motor Performance of a Skilled Walking Task?
1University of British Columbia, Vancouver, BC, Canada,
2International Collaborations On Repair Discoveries, Vancouver, BC, Canada
Poster 8
A Novel Approach to Quantifying Changes in Locomotor EMG Patterns in Incomplete Spinal Cord Injury
1Shepherd Center, Atlanta, GA, USA
2Emory University School of Medicine, Atlanta, GA, USA
In incomplete spinal cord injury (SCI) it has been shown that robotic-assisted body-weight-supported treadmill training can improve stepping. Presumably, this is partly because of improving muscle activation patterns. Methods to quantify those patterns, their degree of abnormality, and their change over time have been an area of ongoing research. The goal of this project was to determine the normalcy of the SCI motor patterns by first developing a prototypical motor pattern for the uninjured population and then comparing the SCI motor patterns to this prototype before and over the course of locomotor training. To develop the prototypical muscle activation pattern, electromyography (EMG) from the right and left quadriceps, hamstrings, tibialis anterior, and soleus muscles were collected from 10 neurologically intact individuals while stepping in the Lokomat (for comparison to SCI). These muscle activation patterns were studied under various combinations of loading (40%, 60%, 95% body weight support; BWS) and treadmill speed (0.5, 0.7, 0.9 m/s). The EMG patterns for each gait cycle were divided into 10 equal bins and averaged within participants over at least 15 steps. Within a participant, the average EMG amplitude for each muscle within a bin was used to create a vector such that each muscle made an equal contribution to its value. The vector was then normalized to the magnitude of the vector, thus, creating a response vector. The response vectors (ie, the relative activation of each of the 8 muscles in 1 bin) for each bin were then averaged across participants to create the prototype pattern. For the injury group, EMG from 4 participants with incomplete SCI were recorded at 2 speeds (0.5 and 0.7 m/s) and 2 loads (40% and 60% BWS) while stepping in the Lokomat. EMG patterns were binned and compared with the uninjured prototype pattern using a similarity index (ie, the cosine of the angle between the prototype vector and individual SCI participant’s vector). The results for the uninjured participants indicated that the lowest BWS conditions (40% and 60% BWS) produce similar EMG patterns across uninjured individuals, whereas the highest BWS (95% BWS) produces atypical motor patterns. For the SCI group, initial results indicate that the similarity index can differentiate between muscle weakness (where muscles activate appropriately but with low magnitude) and abnormal muscle activation patterns (ie, increased dysynergias). Currently, the similarity index is being used to track changes in stepping EMG patterns in individuals with SCI that are related to locomotor training and transcutaneous spinal cord stimulation. In the future, this approach may be used as a metric to characterize injury profile and guide therapy.
Poster 9
Measurement of Volitional Muscle Activity Initiation and Cessation After Spinal Cord Injury
1Shepherd Center, Atlanta, GA, USA
2Emory University, Atlanta, GA, USA
Following spinal cord injury (SCI), control over muscle activation and relaxation is usually impaired, leading to slowed movement and hypertonia. Most efforts to date have measured and described impairment of the ability to activate paralyzed and/or paretic muscles. Equally important to restoring motor control, however, is the ability to deactivate muscles that have been inappropriately recruited or to accurately end contraction at the conclusion of a voluntary task or movement. Surface electromyographic (sEMG) recordings were made in 11 noninjured participants and 10 with incomplete SCI while attempting volitional ankle dorsiflexion in response to a 5-s audible cue. They were instructed to perform voluntary ankle dorsiflexion as quickly as possible on hearing the cuing tone and then to relax as completely and quickly as possible when it ended. sEMG from the tibialis anterior (TA) and triceps surae (TS) muscles was recorded with a 2-kHz sampling rate and processed into root mean square (RMS) envelopes with a 20-Hz smoothing filter. The TA envelope provided measurement of time from the cue to sEMG onset and to peak amplitude along with the time from cue cessation and beginning of RMS amplitude decline to the end of muscle activity. The antagonist envelope, from the TS, provided measurement of the time after cue cessation and the amplitude and duration of an antagonistic burst of muscle activation. No difference was seen between the 2 groups for the time to first motor unit fired, which is cognitive reaction time. From the first unit fired to peak firing, noninjured participants recruited TA motor units more quickly than did SCI participants: 245 ± 66 and 974 ± 674 ms, respectively. Cessation of firing was also faster in the noninjured group, with the time from beginning of firing decrease to the end of motor unit firing taking 399 ± 202 ms, whereas in the SCI group, cessation required 544 ± 343 ms. Finally, 86% of trials in noninjured participants included a burst of antagonist muscle firing, with a mean peak RMS amplitude of 36 ± 23 µV and duration of 237 ± 149 ms, that was seen in only 1 participant in the SCI group. These results suggest that reduced supraspinal control over motor unit firing in SCI affect both the initiation and cessation of contractions that can be quantified using the measurement techniques presented. Furthermore, it suggests that there are at least 2 functional mechanisms for normally ending a volitional contraction: cessation of descending excitation of the agonist and transient excitation of antagonistic muscle motor units. Both these seem to be altered by SCI and may explain abnormalities of muscle relaxation.
Poster 10
Acute and Postacute Assessment of Postural Control and Cognitive Efficiency Following Concussion
1Oregon Health and Science University, Portland, OR, USA
2Veterans Affairs Medical Center, Washington DC, USA
3Georgetown University, Washington DC, USA
Poster 11
A Quantitative Method for Measuring Bilateral Arm Activity via Accelerometry
Washington University in St Louis, MO, USA
Poster 12
Spatial Cognitive Function of the Right Insula: A Study With Stroke Survivors
1Kessler Foundation, West Orange, NJ, USA
2Rutgers University, Newark, NJ, USA
Brain lesions involving the right insular cortex may induce spatial neglect, a neurocognitive disorder manifested as a failure to attend or act on stimuli on the contralesional (ie, left) hemispace. To investigate specific functional deficits following insular lesions, we studied individuals with lesions centered at the right insular region and examined whether an insular damage critically predicts impairment in a specific everyday activity. In all, 12 participants (6 men and 6 women; mean age = 72.3, standard deviation [SD] = 10.9) were included for their lesions restricted in the right insular region, with more than 90% of the insular cortex and less than 10% of the surrounding areas involved. All the participants met the criterion of having spatial neglect, assessed with a neuropsychological test, the Behavioral Inattention Test (BIT), using the standard cutoff score of 129/146; mean BIT score = 95.2; SD = 44.3. Participants’ performance of everyday activities was evaluated with the Functional Independence Measure (FIM) and the Catherine Bergego Scale (CBS). With 18 evaluation items (score = 1-7 per item; higher value implies better function), FIM is the most common clinical measure for independence following stroke. The 10-item CBS (score = 0-3 per item, higher value implies greater severity) is an observational assessment for spatially asymmetrical action deficits in daily life. Participants’ clinical brain images were acquired on average 4.8 days (SD = 7.9) poststroke, and their behavioral performance was evaluated 18.6 days (SD = 7.4) poststroke. Overall, participants had difficulty performing everyday activities, indicated by their total FIM (mean = 102.4, SD = 30.0) and CBS scores (mean = 15.7, SD = 6.9). To investigate whether a lesion location in and around the right insula was critically associated with a specific FIM or CBS item, we performed a series of voxel-based lesion behavior mapping analysis, with a 5% false detection rate. The only result that reached statistical significance was from a CBS item for evaluating difficulties in finding personal belongings in a familiar environment. The critical region (z = 2.826-3.891) was at the anterior insula, with its adjacent extreme capsule and inferior medial temporal areas. This finding suggests the function of the anterior insula in visual search or mental representation of familiar objects in the context of a familiar environment.
Poster 13
AMES + Brain Stimulation for Treatment of Hand Plegia: A Proof-of-Concept Study
Oregon Health and Science University, Portland, OR, USA
It has been estimated that half of all stroke survivors never regain functional use of the upper limb (UL). We investigated, in a pilot study with 6 participants (average age = 54 years; average time since stroke = 39 months), the effects of AMES (ie, assisted movement with enhanced sensation) treatment with concurrent brain stimulation. Inclusion criteria were >1 year postinjury, ≤3 score on the modified Ashworth scale, and no measurable extension of the thumb, fingers, or wrist, but with some residual proprioception in the affected UL. The Study Physician screened out study candidates with comorbidities that would interfere with full study compliance. On enrollment, each participant was randomized to 1 of 2 treatment groups: one in which AMES was combined with transcranial direct current stimulation (tDCS) and the other in which AMES was combined with repetitive transcranial magnetic stimulation (rTMS). There was no control group in this study. The participants were treated for 20 minutes during each of 30 sessions, typically conducted 3 times per week over a period of 10 to 12 weeks. On each treatment day, the participant sat with the affected UL in the AMES device while being prepared for either tDCS or rTMS stimulation. Both methods of brain stimulation were intended to stimulate the motor cortex on the injured side of the brain. During AMES treatment, the AMES robotic device ranged the thumb and 4 fingers at the MCP joints (±15 degrees, 5 degrees/s), opening and closing the hand. During hand opening, a muscle vibrator stimulated (60 pulses/s, 2 mm) the long flexor tendons of the thumb and fingers; during hand closing, a different vibrator stimulated the corresponding extensors. The participant faced a computer screen that provided real-time EMG biofeedback from the finger flexor and extensor muscles. The participant was instructed to assist the movement imposed on the hand by increasing the agonist EMG to a target level while minimizing the antagonist active. The target EMG was initially set, and then maintained, at a level of 25% of the participant’s maximum volitional agonist EMG. All 6 participants regained some volitional movement at 1 or more of 6 locations: thumb, fingers (n = 4), and wrist, at some point during the 10- to 12-week treatment period. On average, participants exhibited some volitional extension on 13/30 treatment days, and on each day there was volitional movement, it was observed at an average of 1.9 of the 6 possible locations. Scores improved slightly on the Chedoke-McMaster Assessment but not significantly. Similarly, a strength test showed no significant strength gains. However, 1 participant recovered considerable functional use of the affected hand, and this recovery has persisted for 1½ years postparticipation. These results demonstrate the feasibility of conducting a clinical trial, without significant risk to participants, and further suggest some potential to reverse chronic hand plegia after stroke.
Poster 14
Reorganization of the Locomotor Network in Parkinson Patients With Freezing of Gait
OHSU, Portland, OR, USA
Freezing of gait (FoG) is a unique and disabling clinical phenomenon of advanced Parkinson’s disease (PD) characterized by brief episodes of inability to step that often occurs on initiation of gait. Recent work implicates compromised structural integrity and transient functional disconnection between subcortical and cortical regions of the locomotor network in individuals who experience FoG. In the current study, we use a novel multimodal neuroimaging approach to assess differences in functional and structural connectivity of the locomotor network between PD patients with (FoG+) and without (FoG−) FoG along with age-matched controls. For the current study, 15 patients with mild to moderate PD and 14 age-matched healthy controls (HCs) were recruited. Parkinson’s patients were tested in the morning off medication and were classified as either FoG+ (n = 8) or FOG− (n = 7) based on their new FoG questionnaire score (NFOGQ). Resting state functional MRI data were collected, and functional connectivity of the locomotor network was assessed between seed regions of the supplementary motor area (SMA) and bilateral subthalamic nuclei (STN) and cerebellar and mesencephalic locomotor regions (CLR and MLR, respectively). Diffusion-weighted images were also collected, and probabilistic tractography was performed between locomotor hubs where functional connectivity differences were observed and there are known anatomical connections. Finally, clinical testing to assess disease severity and gait function was performed. Principal results were as follows: FoG+ patients showed greater connectivity between bilateral MLR and SMA (P < 0.02) and between left CLR and the SMA compared with both FoG− patients and HCs (P < .03). We report a striking difference in the relationship between structural and functional connectivity between the right MLR and SMA, whereby more structural connections were associated with reduced functional communication in HCs (r = −0.37) but increased functional communication in FoG+ patients (r = 0.50). Additionally, more functional connectivity between the right MLR and SMA was strongly correlated with both clinical ratings of FoG (r = 0.71) and self-reported NFOGQ scores (r = 0.74) in FoG+ patients, suggesting that the observed increases in communication are not compensatory. The current findings demonstrate a reorganization of functional communication within the locomotor network in FoG+ patients, whereby the higher-order motor cortices responsible for gait initiation communicate with the MLR and CLR to a greater extent than the patterns observed in FoG− patients and HCs. The observed pattern of altered connectivity in FoG+ does not appear to serve a compensatory role, as evidenced by the positive association between ratings of FoG and increased functional connectivity in the MLR-SMA loop. Intervention tasks targeted at reducing communication in the disadvantageous locomotor loops identified here (MLR/CLR-SMA) may result in improved gait initiation and a reduction in freezing.
Poster 15
Targeted Exercises to Increase Cortical Influence Over Spinal Reflexes
1James J. Peters VA Medical Center, Bronx, NY, USA
2Icahn School of Medicine at Mount Sinai, New York, NY, USA
Poster 16
Damage and Recovery Following White Matter Stroke: A Key Role for Age
UCLA, Los Angeles, CA, USA
Subcortical white matter stroke (WMS) constitutes up to 30% of all stroke subtypes and has devastating clinical consequences. Mechanisms of damage and repair in WMS mainly involve oligodendrocyte and axon injury and regeneration and are distinctly different from the cellular and molecular events resulting from large artery, “gray matter” stroke. Diminished recovery from stroke in elderly patients implies that damage and repair processes are affected by advanced age, but such effects have not been studied in WMS. This study aimed to evaluate the effect of age on white matter damage and repair using a WMS mouse model and promote functional recovery in aged animals following this type of stroke. WMS was produced with focal microinjection of the vasoconstrictor L-NIO into the subcortical white matter ventral to the mouse forelimb motor cortex in young-adult (2 months), middle-aged (15 months), and old mice (24 months). A blocker of the neural growth inhibitor Nogo, NgROMNI, was administered to old mice following stroke to promote repair in a systemic delivery approach designed to model what would be practical in humans. Functional impairment and recovery were assessed using the pasta matrix reaching task, a behavioral test aimed at evaluating strength, reach capacity, and dexterity of the mouse forelimb as well as the grid walking task. WMS was found to produce inflammation, localized oligodendrocyte cell death, and white matter atrophy that were more pronounced in old animals compared to young adults. Behavioral testing revealed an age-dependent exacerbation of forelimb motor deficits caused by the stroke, with decreased long-term functional recovery in old animals. Treatment with NgROMNI resulted in gradual behavioral improvement in old mice and a return to control levels 1 month after WMS—a similar recovery profile to that observed in untreated young-adult mice. These findings demonstrate a profound effect of age on the outcome of WMS but suggest that targeted therapeutic interventions can effectively restore function in aged animals.
Supported by: NIH R01NS071481 and the Dr Miriam and Sheldon G. Adelson Medical Research Foundation.
Poster 17
Feasibility of Bilateral Arm Training in the Subacute Setting
University of Maryland School of Medicine, Baltimore, MD, USA
Poster 18
Bilateral Reach Training Following a Unilateral Ischemic Injury to the Caudal Forelimb Area (CFA) of the Motor Cortex in Rats
1University of Texas, Austin, TX, USA, 2Institute for Neuroscience, Austin, TX, USA
We have previously found that following a unilateral ischemic injury to the CFA of the motor cortex in rats, skill learning with the paretic limb or alternating skill training between the paretic and nonparetic limb facilitates paretic forelimb recovery. Similarly, clinical literature indicates that rehabilitative training with only the paretic arm and both the paretic and nonparetic arm in individuals affected by upper-extremity paresis is effective in restoring some function in the affected body side. Currently, no behavioral metric exists for rodents requiring the use of both limbs concurrently as a rehabilitative training paradigm following unilateral focal motor cortical ischemic damage. In this study we sought to (1) develop a behavioral metric for bilateral rehabilitative training in a rodent model of focal motor cortical ischemia and (2) determine whether this behavioral metric benefits recovery in the paretic limb of rats compared with rats receiving either control procedures or rats receiving paretic limb training in a task that requires less skill. Animals were trained preoperatively on a unilateral reaching task (single-pellet retrieval task) and were then given an ischemic injury. They were then trained for 33 days on either (1) the bilateral reaching task, (2) a unilateral reaching task (tray task), or (3) control procedures. The rats were probed weekly using the single-pellet retrieval task. The initial findings of this study indicate that skillful concurrent use of both forelimbs facilitates paretic forelimb recovery better than control procedures and a skillful task requiring the use of only the paretic forelimb. Ongoing studies are examining the link between cortical reorganization and the behavioral effects of unilateral versus bilateral training.
Poster 19
Impact of Motor Cortical Stimulation Timing During Planar Robotic Training on Neuroplasticity in Older Adults
1University of Maryland School of Medicine, Baltimore, MD, USA
2Moss Rehabilitation Research Institute, Elkins Park, PA, USA
3Geriatrics Research, Education and Clinical Center, Veterans Affairs Medical Center, Baltimore, MD, USA
Robotic technology is increasingly used for rehabilitation, and the potential exists to enhance use-dependent plasticity with noninvasive motor cortex stimulation specifically timed with the practiced reaching movements. The objective was to determine how the timing of stimulation influenced neuroplasticity associated with robotic reaching. During a repetitive reaching intervention using a planar robot with 480 trials of movement, 16 participants completed 3 separate sessions with different stimulation timings (premovement, EMG triggered, random). Subthreshold, single-pulse transcranial magnetic stimulation (TMS) was delivered at premovement (120-150 ms prior to movement onset), EMG triggered (when muscle activity exceeded threshold), or delivered randomly (anytime between visual cue and movement completion). There were 5 assessments, which included the amplitude and direction of TMS-evoked movements with corresponding motor-evoked potential (MEP) recordings from the biceps, triceps, and anterior and posterior deltoid muscles. Each assessment included 10 averaged trials, and MEP data were organized into elbow and shoulder agonist and antagonist groups. Change scores from the initial assessment were calculated as follows: percentage change in amplitude, absolute change in direction (0°-180°), and change in MEP amplitude. Data were analyzed using a general linear model with repeated measures for time and a between-subjects effect for stimulation conditions. The evoked movements significantly increased in amplitude following the premovement and random conditions but not following the EMG-triggered condition (P < .05). The TMS-evoked directions significantly changed following all conditions (P < .05), and no differences between groups were observed. There was a significant difference between conditions for shoulder and elbow agonist MEPs, with a general increase in amplitude following premovement and a significant decrease following the EMG-triggered condition (P < .05). These findings demonstrated that stimulation delivered during premovement and EMG-triggered conditions influenced neuroplasticity with a robotic reaching intervention; yet the mechanism by which these changes occurred appeared to differ. The direction of plasticity may change rapidly and selectively during the planning and production of movement. This dynamic change in plasticity shapes the way in which practice results in motor system changes.
Poster 20
Transcallosal Effects of Chronic Below-Elbow Arm Amputation: A Pilot Study
1Georgetown University, Washington, DC, USA
2Medstar National Rehabilitation Hospital, Washington, DC, USA
3DC Veterans Affairs Medical Center, Washington, DC, USA
Poster 21
Graph Theoretical Analysis of Resting State EEG in Postacute Stroke Recovery
1Maryland Exercise and Robotics Center of Excellence, Baltimore Veterans Affairs Medical Center, Baltimore, MD, USA
2Department of Physical Therapy and Rehabilitation Science, University of Maryland School of Medicine, Baltimore, MD, USA
3University of Maryland Rehabilitation and Orthopaedics Institute, Baltimore, MD, USA
4Department of Neurology, University of Maryland School of Medicine, Baltimore, MD, USA
Poster 22
Reliability of Negative BOLD in the Primary Motor Cortex
1Atlanta VAMC, Center for Visual and Neurocognitive Rehabilitation, Atlanta, GA, USA
2Emory University, Atlanta, GA, USA
For more than 10 years, the existence of the negative BOLD signal in functional magnetic resonance imaging (fMRI) has been demonstrated across multiple domains. There is strong evidence indicating that the negative BOLD response in the primary motor cortex represents an active inhibition of areas in which it is found. This has important implications for rehabilitation research because many disorders of the motor system are characterized by a loss of cortical inhibition otherwise found in healthy adults. This loss of inhibition is also seen during the aging process and is correlated with decreases in motor performance also associated with aging. The present study investigated the reliability of the evoked negative BOLD response in 7 healthy young adults during a finger tapping paradigm in fMRI over 3 sessions. The results were compared with a cohort of 7 healthy older adults again assessed over 3 imaging sessions. The findings indicate that negative BOLD can reliably be evoked in the ipsilateral motor cortex across multiple sessions in younger adults. However, the spatial locus and magnitude of BOLD activity in the ipsilateral primary motor cortex in older adults is greatly variable.
Poster 23
Mechanisms of Rhythm and Pattern Generation of the Human Lumbar Spinal Cord Tested by Epidural Stimulation
1Medical University Vienna, Vienna, Austria
2Vienna University of Technology, Vienna, Austria
3Baylor College of Medicine, Houston, TX, USA
4Foundation for Movement Recovery, Oslo, Norway
The lumbar spinal cord has the capability to produce rhythmic motor outputs in the absence of descending modulations across mammals, including humans.1,2 The human lumbar cord can generate a rich repertoire of rhythmic locomotor-like and non-locomotor-like patterns in response to a nonpatterned multisegmental drive provided by epidural spinal cord stimulation (SCS).2,3 In the present work, we hypothesized that the variety of these rhythmic motor patterns can be explained by common basic control principles of central rhythm and pattern generation. We studied 10 motor-complete spinal cord injured patients with epidural electrodes over the lumbar spinal cord for spasticity control. SCS was applied in the supine position that reduced the influence of afferent information essential for rhythm generation (hip extension, limb load). Then, 10-s segments of stable rhythmic patterns of EMG activities in the quadriceps, hamstrings, tibialis anterior, and triceps surae unilaterally were extracted from the EMG data. A nonnegative matrix factorization (NMF) algorithm was applied to the pooled EMG profile data across patients, muscles, and samples of stable EMG patterns to test whether decomposition would identify underlying basic activation patterns. A total of 39 samples of 10-s segments of rhythmic EMG patterns were identified in 7 patients (effective SCS frequency = 29.5 ± 4.85 Hz). The rhythm frequency was identical across all muscles in a given sample. Despite the variety of rhythmic EMG patterns generated, NMF of the pooled data demonstrated that they could be closely reproduced by a linear combination of a small number of basic activation patterns with appropriate weights. According to the Akaike Information Criterion,4 a model with 3 basic activation patterns had the highest relative probability to best describe the EMG profile data. Two basic activation patterns had sinusoidal-like shapes, one peaking during an extension-like and the other during a flexion-like phase. There was a highly significant negative correlation of their weights required to construct the various EMG profiles. The third basic pattern peaked in the early flexion phase and was not related to the other basic activation patterns. The constant phase relation of rhythmic outputs to 1 lower limb suggests plurisegmentally organized rhythm generation. The 2 elementary basic activation patterns closely resemble the 2-phase motor pattern of fictive locomotion of experimentally reduced animal models that is generated by central pattern generating networks.1 The third basic pattern confirms neural processes beyond half-center oscillations. Understanding the processing and pattern-shaping capabilities of the human lumbar cord networks is clinically significant for the modification of altered motor control in neurological conditions.
Grillner S. Science. 2011;334:912-913.
Dimitrijevic MR, et al. Ann N Y Acad Sci. 1998;860:360-376.
Minassian K, et al. Hum Mov Sci. 2007;26:275-295.
Akaike H. IEEE Trans Automat Contr. 1974;19:716-723.
Poster 24
Alternating Reflex Modulations by Spinal Neural Circuits “Outside” the Human Lumbar Locomotor Pattern Generator
1Medical University Vienna, Vienna, Austria
2Vienna University of Technology, Vienna, Austria
3Baylor College of Medicine, Houston, TX, USA
4Foundation for Movement Recovery, Oslo, Norway
Epidural lumbosacral spinal cord stimulation (SCS) generates a variety of motor outputs to the lower limbs of motor-complete spinal cord injured (SCI) individuals, depending on the applied stimulation parameters. Specifically, coordinated rhythmic flexion-extension movements can be evoked at 25 to 50 Hz.1 The motor activities produced by SCS comprise a series of stimulus-triggered posterior root-muscle (PRM) reflexes that are modulated to form the burst-like electromyographic (EMG) activities.2 Periodic, alternating modulations of successive PRM reflexes are another consistent EMG pattern generated with SCS frequencies and intensities below the ones leading to the rhythmic flexion-extension movements. These patterns, characterized by an oscillation period covering 2 consecutive responses and the attenuation of every other response magnitude, will be elaborated in the following. PRM reflexes of quadriceps (Q), hamstrings (Ham), tibialis anterior (TA), and triceps surae (TS) bilaterally to SCS at 2 to 26 Hz derived from 8 motor-complete SCI individuals were analyzed for their EMG characteristics for 50 ms poststimulus. Stimulation at 16 Hz and intensities of 1 to 1.5 times the respective motor thresholds were most effective in generating the alternating EMG patterns across all muscles, and there was no specifically effective segmental stimulation site. Generally, the alternating EMG patterns were established with the first few SCS pulses. Periodic reflex alternations occurred more frequently in the thigh (found in 29% of all data sets) than the leg muscles (20%). In between Q and Ham, the EMG patterns were either modulated in phase or had a reciprocal relation. In TA and TS, in-phase modulations of the EMG patterns in the antagonists were the common finding. In between the left and right lower limb, a series of PRM reflexes of a given muscle group were generally modulated in phase. The amplitude modulations of the successive PRM reflexes reflected the excitability of the spinal motor neural circuitry at the time each impulse arrived from the depolarized posterior root fibers. The short period of SCS required to establish the alternating reflex modulations along with the lower effective SCS frequencies and intensities as compared with those generating the rhythmic flexion-extension movements suggest the activity of relatively simple neuronal networks as the underlying mechanism. These networks may incorporate time-dependent processes that enhance or reduce neuronal activities and may include recurrent inhibition and reciprocal interconnection between circuits controlling different motor pools. Gaining insight into spinal mechanisms of inhibition and excitation as well as the frequency-specificity of their engagement in the processing of afferent volleys will be crucial for advancing and developing novel rehabilitation strategies.
Dimitrijevic MR, et al. Ann N Y Acad Sci. 1998;860:360-376.
Minassian K, et al. Spinal Cord. 2004;42:401-416.
Poster 25
Educate, Train, Treat, Track: Bringing State-of-the-Art Care to Our Military With TBI
1US Army Office of the Surgeon General, Falls Church, VA, USA
2US Army Medical Research and Material Command Combat Casualty Care Research Program, Fort Detrick, MD, USA
3Oak Ridge Institute for Science and Education, Bellcamp, MD, USA
4University of North Carolina at Chapel Hill, NC, USA
5Sister Kenny Research Center, Minneapolis, MN, USA
Here, we describe the US Army traumatic brain injury (TBI) program and display progress from the US Army TBI Task Force. The capabilities and services in the deployed and garrison environments within the context of Department of Defense (DoD) policy for TBI care and existing gaps within the system are discussed as well as the evolution of, and current, policies and clinical algorithms in the deployed and garrison environments and DoD clinical recommendations. We elaborate on the neurocognitive assessment tool and role of neurocognitive assessment in return to duty decision making, present the DoD TBI coding procedures, and discuss challenges in analyzing coded data. Army TBI research initiatives related to TBI are delineated, and finally, we discuss the Army TBI education and training strategies used to educate a widely dispersed population of medical staff and providers as well as specific tools and resources developed to support the TBI mission to include patient education handouts, educational videos and slide decks, and the TBI Rehabilitation ToolKit.
Poster 26
Premotor Cortical Stimulation in Stroke Rehabilitation: Neural Mechanisms of Recovery
1Cleveland Clinic, Cleveland, OH, USA
2Kessler Foundation, West Orange, NJ, USA
In recent clinical trials, adjuvant cortical stimulation has shown no advantage in stroke rehabilitation, unlike in prior work. Variable success may be a result of the fact that the usual target, the primary motor cortex (M1), is spared only in a few. Targeting higher motor areas, such as the premotor cortex (PMC), may be more effective because it assumes the role of the damaged M1 and contributes more significantly to corticospinal and interhemispheric connections. Knowing individual mechanisms of recovery would help understand variability of effect. In a pilot randomized, double-blinded clinical study, we examine the efficacy and underlying mechanisms of stimulating the PMC in rehabilitation of the hand in chronic stroke. Patients were assigned to rehab+stim or rehab+sham groups. Rehabilitation was delivered for 1 hour, 3 times/wk for 5 weeks. Anodal transcranial direct current stimulation was applied at 1 mA to the PMC in the stroke hemisphere using MRI-based stereotaxy. We measured impairments and force of the paretic hand. Diffusion tensor imaging (DTI) defined integrity, and transcranial magnetic stimulation (TMS) noted the physiological efficiency of corticospinal tracts. TMS also noted transcallosal inhibition from stroke on the intact hemisphere. Functional magnetic resonance imaging (fMRI) noted interhemispheric balance for M1 and PMC. Age-matched controls were also studied. Impairments were alleviated, and the force of the paretic hand improved across all patients; however, changes in the rehab+stim group alone approached values in healthy individuals. The rehab+stim group also had improved corticospinal conduction, transcallosal inhibition exerted from stroke on intact motor areas, and interhemispheric balance that shifted back to M1 and PMC in the stroke hemisphere. These mechanisms were less prominent in the rehab+sham group. Patients with greater integrity of corticospinal tracts from M1 and PMC in the stroke hemisphere and stronger changes in corticospinal conduction showed significantly better recovery of force of the paretic hand: r = 0.9, P = .01; r = 0.78, P = .05, and r = 0.88, P = .02. Those who showed greater shift of interhemispheric control to the PMC in the stroke hemisphere also exhibited higher recovery (r = 0.79, P = .05). Thus, the PMC may be a potential surrogate target for stimulation in concurrent chronic stroke rehabilitation. Success of pairing may depend on integrity and functionality of the corticospinal output from primary and premotor areas and the potential of the targeted region (such as PMC) in the stroke hemisphere to regain focus of movement control. Variability of adjuvant cortical stimulation in rehabilitation can be mitigated by addressing prognostic indicators with DTI, TMS, and fMRI.
Poster 27
Diffusion Tensor Imaging Exhibits More Proximate and Reliable Transcranial Magnetic Stimulation Measures Compared With fMRI in Stroke
1Cleveland Clinic, Cleveland, OH, USA
2Kessler Foundation, West Orange, NJ, USA
In stroke, functional MRI (fMRI) serves as a poor guide to direct transcranial magnetic stimulation (TMS). Localization with fMRI is variable because its hemodynamic contrast is contorted in and around infarcted tissue. Furthermore, fMRI reflects activity in the gray matter, whereas TMS examines conduction via white matter, such as corticospinal tracts (CSTs). To develop a better guide for TMS in stroke, we examined whether imaging CST terminations in cortices using diffusion tensor imaging (DTI) is more reliable and proximate to sites of TMS than fMRI. Four patients with stroke and 4 aged-matched healthy controls underwent fMRI during hand movement and DTI. Stereotactic TMS was delivered to motor cortical sites in a 7 × 5 grid in patients’ affected hemispheres while we recorded 5 motor-evoked potentials (MEPs) in the paretic first dorsal interosseous muscle. We compared the distances between and reliability of MEPs at the following sites: site of highest fMRI activation, cortical sites with best myelin (transverse diffusion) and overall (fractional anisotropy) integrity of CST terminations, and optimal site of TMS (weighted center of gravity of MEPs). Overall, MEPs were reliable when TMS was applied to sites with best integrity of CST terminations; however, when TMS was applied to the site of maximum fMRI activation, MEPs were absent in 50% of patients and 25% of controls. At sites of best myelin integrity of CST, trial-to-trial reliability of MEPs (measured with coefficient of variation) trended toward being better than that at the site of maximum fMRI activation (55.58% ± 44.74% vs 70.32% ± 13.63%; P = .07). In the affected hemisphere, the cortical sites with best myelin integrity and overall integrity of CST terminations were closer to the optimal TMS site than the site of maximum fMRI activation (4.1 ± 5.0 mm and 5.7 ± 6.7 mm vs 14.9 ± 3.9 mm, respectively; P < .05). Preliminary results of this ongoing work suggest that imaging CSTs may be reliable and accurate in localizing TMS in stroke. Future studies can explore whether DTI guidance maximizes outcomes of therapeutic TMS in stroke rehabilitation. Exploring reliable navigation for TMS will allow us to customize therapy to the patient’s own neural substrates.
Poster 28
Individually Targeted Transcranial Direct Current Stimulation Enhances Fluency in Patients With Chronic Nonfluent Aphasia
1Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA
2Center for Cognitive Neuroscience, University of Pennsylvania, Philadelphia, PA, USA
3Department of Neurology, University of Pennsylvania, Philadelphia, PA, USA
Poster 29
Activation Changes Induced by Mnemonic Strategy Training During Memory Encoding and Retrieval in Patients With Mild Cognitive Impairment
1Atlanta VAMC RR&D Center of Excellence, Decatur, GA, USA
2Emory University, Atlanta, GA, USA
The diagnosis of mild cognitive impairment (MCI) is widely considered a precursor to Alzheimer’s disease. Patients with MCI demonstrate significant learning and memory deficits that are often accompanied by reduced prefrontal and medial temporal lobe activation during functional magnetic resonance imaging (fMRI) tasks. Given the known interactions between the prefrontal cortex and medial temporal memory system, cognitive rehabilitation methods that re-engage the prefrontal cortex may ultimately maximize residual hippocampal functioning and improve learning and memory as a result. However, it is currently unknown whether and the extent to which any such changes affect the encoding and/or retrieval phase. Therefore, we examined the overlap in activation changes (posttraining minus pretraining) during the successful encoding and subsequent retrieval of object location associations in a group of 9 MCI patients who underwent mnemonic strategy training as part of a randomized controlled trial.1 These patients underwent 2 separate fMRI sessions and practiced using mnemonic strategies during 3 intervening training sessions. In addition to significantly improved memory for the stimuli learned during training, there were robust increases in activation during both encoding and retrieval. Areas of overlap were generally constrained to those commonly associated with the so-called default mode network (ie, lateral temporal, inferior parietal, and midline areas), which is consistent with the role these regions play in episodic memory recall. Increased activation in the lateral and anterior aspects of the prefrontal cortex and associated subcortical circuits were typically only observed during encoding, a finding that suggests that these regions play a unique role in the encoding process. Similar results emerged as participants encoded and retrieved novel associations. It is important to note that many of these effects were specific to those receiving mnemonic strategy training because they were not evident in a matched exposure control group of 9 additional MCI patients. These results demonstrate that cognitive rehabilitation (re)engages the prefrontal cortex and associated subcortical structures in relatively distinct ways during encoding and retrieval.
Hampstead BM, Sathian K, Phillips PA, et al. Mnemonic strategy training improves memory for object location associations in both healthy elderly and patients with amnestic mild cognitive impairment: a randomized, single blind study. Neuropsychology. 2012;26:385-399.
Poster 30
Neurochemical Predictors of Motor Recovery in Stroke
1University of Kansas Medical Center, Kansas City, KS, USA
2University of Missouri, Columbia, MO, USA
Poster 31
Effects of Feedback to Enhance Self-efficacy on Paretic Hand Use in Chronic Stroke
1University of Southern California, Los Angeles, CA, USA
2Rancho Los Amigos National Rehabilitation Center, Downey, CA, USA
3China Medical University, Taichung City, Taiwan
4University of Maryland Baltimore, Baltimore, MD, USA
Poster 32
Home-Based Robot-Assisted Ankle Rehabilitation for Chronic Stroke Survivors
A T Still University, Mesa, AZ, USA
Stroke is the leading cause of disability in the United States. Foot drop, a major sequela associated with stroke, contributes to locomotor impairments. Robot-assisted repetitive task practice is one approach that has been shown to improve lower-extremity function and locomotion in stroke survivors. Robotic training, however, is typically confined to large clinics or research laboratories that few patients have access to. The purpose of the current study was to investigate the effects of home-based robot-assisted ankle rehabilitation on strength, locomotion, and quality of life in chronic stroke survivors. Four individuals participated in this single-group repeated-measures design study. Isometric dorsiflexion strength, locomotor function, balance, and quality of life were assessed 3 times during a 2-week baseline period, 3 times during a 12-week intervention period, and once after a 4-week follow-up period. The intervention consisted of three 60-minute home-based robot-assisted training sessions (Foot Mentor, Kinetic Muscles Inc, Tempe, AZ) per week for 12 weeks. Use and performance data from the robotic device were monitored remotely, and feedback was given weekly via telephone. All 4 participants adhered to the intervention protocol, with no reports of adverse events. All 4 participants demonstrated increases in strength, gait speed, gait distance, and quality of life over time. At week 12, maximal isometric dorsiflexion force increased by an average of 37%, gait speed increased by an average of 0.2 m/s, distance on the 6-minute walk test increased by an average of 34.6 m, and the physical function composite score of the stroke impact scale increased by an average 7.5 points. Limited carryover was observed 4 weeks after the cessation of treatment. No consistent improvements in balance as measured by the limits of stability test on the Balance Master (NeuroCom, Clackamas, OR) were observed. These results suggest that home-based robot-assisted ankle rehabilitation improves strength, locomotor function, and quality of life in chronic stroke survivors. Continued treatment, however, may be required maintain the improvements. In addition, balance did not appear to be affected by this intervention. In the changing landscape of health care, it is important to investigate alternative methods for delivering physical therapy. Home-based robotic interventions are one such methodology falling under the heading of telerehabilitation. The results presented here provide preliminary evidence supporting the use of home-based robotics for the treatment of distal lower-extremity dysfunction in chronic stroke survivors.
Poster 33
Central Plasticity of Segmental Cutaneous Nociceptive Primary Afferents Is Associated With Evoked Nociceptive Hyperreflexia After Hemisection Spinal Cord Injury in Rats
1Neurology, Physiology and Biomedical Engineering, Emory University School of Medicine, Atlanta, GA, USA
2Biomedical Engineering, Georgia Institute of Technology, Atlanta, GA, USA
3Hulse SCI Lab, Spinal Cord Injury Research Program, Shepherd Center, Atlanta, GA, USA
The cutaneous trunci muscle (CTM) nociceptive intersegmental spinal reflex results from stimulation of segmental dorsal cutaneous nerves (DCNs) and demonstrates complex neurophysiological plasticity after hemisection spinal cord injury (SCI). The central projection patterns of A and C fibers in DCNs were analyzed 6 weeks after a unilateral T10 spinal cord hemisection. Retrograde axonal tracers, cholera toxin subunit B for myelinated A fibers, or isolectin B4 for nonpeptidergic, unmyelinated C fibers were injected into bilateral T7 and T13 DCNs, one tracer on one side and the other on the other side. Immunohistochemistry was performed on serial transverse sections of the spinal cord at T7 and T13 to measure the projection fields of labeled A and C fibers quantitatively as immunoreactive areas. Synaptophysin, a synaptic vesicle protein, was used to identify the synaptic terminations of the A and C fibers. Neurophysiological changes of both Ad and C fiber–evoked reflex responses have been shown after spinal cord hemisection injury, both above and below the level of injury, on the side of hemisection, and on the uninjured side. In this study, we investigated the central projection profiles of bilateral A and C fibers of T7 and T13 DCNs 6 weeks after a unilateral T10 hemisection injury and in uninjured animals. Both A and C fiber projection areas increased overall in T7 and T13 after injury compared with normal. However, the greatest mean increase was seen in C fibers at T7, whereas increases of A fibers were the next greatest at T13. Only C fibers at T7 showed a significant difference between the injured and uninjured sides, with those on the injury side being greater. The rostrocaudal distributions of both A and C fiber projections expanded in both rostral and caudal directions after injury. Not only was there greater area of labeled A and C fiber neurites, but they also covered a greater territory of the dorsal horn after injury. After hemisection SCI, the altered projection patterns of DCN A and C fibers in the dorsal horn demonstrate central plasticity of the nociceptive inputs of the CTM reflex. These changes are consistent with the neurophysiological plasticity seen in this reflex after injury. Preliminary work with synaptophysin has shown that these afferent projection area changes also represent overall synaptogenesis following injury. These data demonstrate that anatomical changes of sensory fibers in the CTM pain reflex probably contribute to the nociceptive hyperreflexia observed neurophysiologically after hemisection injury. This supports the idea that the CTM reflex could be a good animal model in which to study the neuroplasticity associated with neuropathic pain after SCI that can be quantitatively analyzed, at least for the altered spinal processing of nociceptive signals.
Poster 34
Central Plasticity of Segmental Cutaneous Nociceptive Primary Afferents Is Associated With Evoked Dysautonomia After Cervical Spinal Cord Injury in Rats
1Neurology, Physiology and Biomedical Engineering, Emory University School of Medicine, Atlanta, GA, USA
2Hulse SCI Lab, Spinal Cord Injury Research Program, Shepherd Center, Atlanta, GA, USA
3Biomedical Engineering, Georgia Institute of Technology, Atlanta, GA, USA
The cutaneus trunci muscle nociceptive intersegmental spinal reflex results from stimulation of segmental dorsal cutaneous nerves (DCNs). DCN stimulation also generates a depressor blood pressure response via the autonomic nervous system in normal animals but can generate a pressor response after severe cervical spinal cord crush injury (SCI). The central projection patterns of A and C fibers in DCNs were analyzed 2 and 4 weeks after a C7 bilateral crush SCI. Retrograde axonal tracers, cholera toxin subunit B for myelinated A fibers, or isolectin B4 for nonpeptidergic, unmyelinated C fibers were injected into bilateral T7 and T13 DCNs, one tracer on one side and the other on the other side. Immunohistochemistry was performed on serial transverse sections of the spinal cord at T7 and T13 to measure the projection fields of labeled A and C fibers quantitatively as immunoreactive areas. Synaptophysin, a synaptic vesicle protein, was used to identify the synaptic terminations of the A and C fibers. Responses of the autonomic system to electrical DCN stimulations were assessed by cannulation of the carotid artery to measure blood pressure. Blood pressure changes to DCN stimulation in cervical injured rats range from mild dysautonomia to frank autonomic dysreflexia and were usually more pathological with rostral DCN stimulation relative to caudal DCN stimulation. In this study, we investigated the central projection profiles of bilateral A and C fibers of T7 and T13 DCNs 2 and 4 weeks after injury and in uninjured animals. C fiber projection areas, more so than A fiber projection areas, increased in T7 and T13 after C7 bilateral crush injury compared with normal. The rostrocaudal distributions of both A and C fiber projections expanded in both rostral and caudal directions after injury. Not only was there a greater area of labeled A and C fiber neurites, but they also covered a greater territory of the dorsal horn after injury. Current measurements are under way to compare the extent of afferent sprouting at T7 versus T13 to see if the rostral/caudal physiological observations are paralleled by anatomical changes.
Poster 35
Motor Control Changes in Incomplete SCI Generated by Adding Tonic Transcutaneous Spinal Cord Stimulation to Robotic Locomotor Training: A Test-of-Concept Study
1Hulse SCI Lab, Spinal Cord Injury Research Program, Shepherd Center, Atlanta, GA, USA
2Neurology and Physiology, Emory University School of Medicine, Atlanta, GA, USA
Following spinal cord injury (SCI), there is an altered combination of afferent and supraspinal input to the lumbar neural circuitry for stepping. Tonic transcutaneous spinal cord (dorsal root) stimulation (tSCS) at the lumbar level has been shown to augment locomotor output after SCI in a frequency-dependent manner, suggesting that increased afferent input may be able to substitute, to a degree, for decreased supraspinal input. This suggests that tSCS can alter spinal cord motor control. In this test-of-concept study, we investigated both the acute effect of tSCS on voluntary motor control and the effect of adding tSCS to robotic locomotor training. We studied the change in motor control over the course of locomotor training in terms of reflex modulation and muscle activation patterns during stepping and using clinical scales for gait function and spasticity. A patient with a chronic C5 motor-incomplete SCI underwent a brain motor control assessment (BMCA) both with and without tSCS (at 50 Hz) to determine if tSCS could alter motor control during voluntary tasks. The patient then participated in a series of 36 robotic body-weight-supported treadmill training sessions to improve gait function, and overground measures (Timed Up and Go, Berg Balance, 6-minute walk, and 10-m walk over a GAITRite) were serially recorded. The patient then underwent an additional 36 robotic and 24 overground gait training sessions combined with sub–motor threshold tSCS (450 µs biphasic pulses, 50-70 mA, 30 Hz), and the same overground gait assessments were done again. After training, the patient again underwent a BMCA. H- and posterior root motor reflex modulation and muscle patterns during gait, lower-extremity motor scores, and the Spinal Cord Assessment Tool for Spastic Reflexes were evaluated to further characterize changes in motor control. With tSCS, the patient demonstrated better muscle activation patterns and speed of muscle activations during voluntary tasks and also showed improvements in gait function with locomotor training that were then augmented following locomotor training combined with tSCS. Following the addition of tSCS, reflex modulation became more normal, with H reflexes being larger in stance and smaller in swing. Further analyses of motor control measures are ongoing. This test-of-concept study shows that traditional rehabilitation strategies to improve locomotion and voluntary motor control can be augmented by tSCS and justifies a larger study of combined tSCS and locomotor training.
Poster 36
Real-Time Optimization of Sensory Stimulation to Improve Walking Metrics After Spinal Cord Injury
1Hulse SCI Lab, Spinal Cord Injury Research Program, Shepherd Center, Atlanta, GA, USA
2Spinal Cord Injury Clinic, Atlanta Veterans Administration Medical Center, Atlanta, GA, USA
3Biomedical and Electrical Engineering, Georgia Institute of Technology, Atlanta, GA, USA
In normal individuals, spinal neurons integrate sensory feedback from the legs with descending commands from the brain to produce walking. After severe spinal cord injury (SCI), spinal neurons are mostly disconnected from the brain, leading to the loss of voluntary movement below the injury, including the ability to walk. However, spinal neurons associated with walking—collectively called the locomotor central pattern generator (CPG)—remain connected to the peripheral sensory nervous system from the legs. Sensory stimulation can recruit the CPG even after SCI, at least to some degree. Our research explores how well the best sensory stimulation patterns can recruit the CPG to assist with walking in humans after severe SCI. Toward that end, we have developed a real-time, closed-loop optimization approach to explore the stimulation space. Participants were placed in a robotic gait orthosis, the Lokomat (Hocoma Inc, Switzerland), to record force output and lower-extremity electromyography (EMG) during standardized stepping. Multiple sensory stimulation sites were tested, including: the fibular nerve (both common and superficial), the tibial nerve (both proximal and distal), sartorius muscle afferents, and the posterior roots, using transcutaneous spinal cord stimulation. These sites were tested both individually and in combination. At each site, a stimulation pattern varying in both frequency and pulse amplitude across the gait cycle was optimized in real time to generate the most normal EMG activity pattern and the best muscle force profile (the least robotic assistance). So far, the optimization algorithm has shown a posterior root stimulation-frequency-dependent effect on the EMG pattern during walking, and the algorithm was able to identify the stimulation pattern that generated the best EMG activity and force profile. Ongoing work is exploring the results from the other sensory stimulation sites with other activation patterns. In summary, optimization techniques may be a powerful tool for studies where outcomes can be measured quantitatively and input can be changed on rapid time scales.
Poster 37
Study Protocol: Rationale and Methods for a Pilot Study of Brain Imaging to Predict Response to Robotic Rehabilitation During Inpatient Rehabilitation
1University of Maryland School of Medicine, Baltimore, MD, USA
2University of Maryland Rehabilitation and Orthopedic Institute, Baltimore, MD, USA
3Maryland Psychiatric Research Center, Catonsville, MD, USA
4University of Maryland, Baltimore County, Baltimore, MD, USA
Robotic therapy has shown benefit for poststroke rehabilitation of arm hemiparesis. However, most research on robotic rehabilitation has been applied to patients irrespective of the anatomical and functional characterization of their brain lesions. Given the evidence from basic human motor sciences of a dynamic interplay between different brain regions for the control of movement, we hypothesize that lesion characterization by multimodal brain imaging could lead to better prediction of who will benefit most from robotic rehabilitation. This knowledge would inform health policy guidelines for the use of robotic therapy poststroke. To this end, we are enrolling patients within 8 weeks of first hemiparetic stroke who are admitted for inpatient rehabilitation. In addition to standard multidisciplinary therapy, patients undergo 5 days of additional robot-assisted therapy with the InMotion2 shoulder-elbow planar robot. Prior to robot-assisted therapy, patients are evaluated via MRI imaging for resting state connectivity and q-space diffusion-weighted imaging. Arm impairment is measured via the Fugl-Meyer Assessment and health-related quality of-life via the Neuro-QOL computer adaptive test. Preliminary data will be presented.
Poster 38
Improved Motor Performance in Chronic Spinal Cord Injury Following Upper-Limb Robotic Training
1Burke Medical Research Institute, White Plains, NY, USA
2Cornell University, New York, NY, USA
3Massachusetts Institute of Technology (MIT), Boston, MA, USA
4Harvard Medical School, Boston, MA, USA
Poster 39
Initiating Word-Finding Trials With Left-Hand Movement During Anomia Treatment Remaps Frontal Language and Executive Mechanisms
1VA Center of Excellence for Visual and Neurocognitive Rehabilitation, Atlanta, GA, USA
2Emory University, Atlanta, GA, USA
3University of Alabama, Birmingham, AL, USA
4Ohio State University, Columbus, OH, USA
5University of Florida, Gainesville, FL, USA
6University of California, San Diego, La Jolla, CA, USA
7Brooks Rehabilitation Clinical Research Center, Jacksonville, FL, USA
Chronic aphasias suggest limits to left-hemisphere compensation for damaged language substrates. To facilitate reorganization of these substrates, we developed an intention treatment (IT) to shift lateral frontal mechanisms rightward during anomia treatment by initiating word-finding trials with complex left-hand movements. We previously showed that IT successfully does so and that treatment effects generalize to untrained items and discourse. Preliminary data indicate that better improvement during treatment is associated with rightward lateral frontal laterality shifts in nonfluent patients but with rightward posterior perisylvian shifts in fluent aphasias. This work addresses the effects of IT specifically on frontal cortices associated with executive control and language output.
Poster 40
Facilitating Neuronal Recovery After Pediatric Traumatic Brain Injury Using Optogenetics
1F M Kirby Research Center for Functional Brain Imaging, Kennedy Krieger Institute, Baltimore, MD, USA
2The Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, Baltimore, MD, USA
3Department of Anesthesiology and Critical Care Medicine, Johns Hopkins University School of Medicine, Baltimore, MD, USA
4Department of Pediatrics, Johns Hopkins University School of Medicine, Baltimore, MD, USA
The robustness of plasticity mechanisms during brain development is essential for synaptic formation and has a beneficial outcome following sensory deprivation. However, the role of plasticity on the recovery following acute brain injury in children has not been well defined. Traumatic brain injury (TBI) is the leading cause of death and disability among children, and long-term disability from pediatric TBI can be particularly devastating. We investigated the altered cortical plasticity 2 to 3 weeks after injury in a pediatric rat model of TBI. Significant decreases in the neurophysiological responses across the depth of the noninjured, primary somatosensory cortex (S1) in TBI rats, compared with age-matched controls, were detected with electrophysiological measurements of multiunit activity (86.4% decrease), local field potential (75.3% decrease), and functional magnetic resonance imaging (77.6% decrease). Because the corpus callosum is a clinically important white matter tract that was shown to be consistently involved in posttraumatic axonal injury, we investigate its anatomical and functional characteristics after TBI. Indeed, corpus callosum abnormalities in TBI rats were detected with diffusion tensor imaging (9.3% decrease in fractional anisotropy) and histopathological analysis (14% myelination volume decreases). Whole-cell patch clamp recordings further revealed that TBI results in significant decreases in spontaneous firing rate (57% decrease) and the potential to induce long-term potentiation in neurons located in layer V of the noninjured S1 by stimulation of the corpus callosum (82% decrease). These results suggest that neurons located in layer V of the noninjured hemisphere are particularly vulnerable to cortical injury. Our goal was to rescue the hypoactivity of layer V neurons associated with post-TBI using channelrhodopsin-2 (ChR2) activation. For this purpose, via lentivirus infection, we genetically engineered rats to express ChR2 in layer V neurons. We facilitated layer V neuronal excitability in the weeks following the injury in order to recover brain functions. Our preliminary results suggest that optogenetics manipulation of layer V neurons is a promising approach for reversing the adverse neuronal mechanisms activated post-TBI.
Poster 41
Implementation of a Multicenter, International, Randomized Clinical Trial in Subacute Stroke Patients Using Wireless Health Technology
UCLA Wireless Health Institute, Los Angeles, CA, USA
Poster 42
Development of Virtual Games With Functional Electrical Stimulation for Poststroke Hemiplegia
1Case Western Reserve University, Cleveland, OH, USA
2MetroHealth Rehabilitation Institute of Ohio, Cleveland, OH, USA
Virtual reality (VR) games were developed to enhance contralaterally controlled functional electrical stimulation (CCFES) therapy for restoring hand function to hemiparetic stroke survivors. CCFES enables hemiplegic patients to open their paretic hand by surface stimulating hand extensors. Stimulation is proportional to the degree of unimpaired hand opening as detected by a sensor glove. Thus, volitional nonparetic hand opening produces stimulated paretic hand opening. CCFES features important for motor recovery are the following: motor intent linked with execution, bilateral movement, and functional task practice. Pilot CCFES studies demonstrated efficacy in chronic (>6 months) and subacute (<6 months) hemiplegic patients. VR games may optimize CCFES. Existing CCFES includes repetitive hand opening practice cued by audio (at home, 5 d/wk, 2 h/d). Although this exercise facilitates repetition, it is not goal oriented and does not require much motor planning or attention. VR games from the literature are independently promising but have not been combined with electrical stimulation. VR+CCFES games were designed in collaboration with clinicians, followed by iterative refinement by hemiplegic patients. Participants were hemiplegics and able-bodied personnel involved in 2 ongoing CCFES trials. Games were first evaluated by 3 physiatrists, 2 occupational therapists (OTs), and a biomedical engineer to ensure that the following criteria were satisfied: (1) game was fun and intuitive; (2) difficulty was appropriate; (3) goals were achieved using targeted (not compensatory) movements; (4) performance feedback helped motivate improvement; (5) graphics were not distracting; (6) CCFES helps performance. Evaluations were repeated until criteria were met—first by clinicians, then hemiplegic patients. These games were developed using Unity (Unity Technologies, CA) to run on PCs and used bend sensors to track hand motion (Phidgets, Alberta, Canada).
Paddle tennis: Players control a paddle to bounce a ball past the opponent’s paddle. Graded hand opening is targeted, so vertical paddle position is controlled by the paretic hand opening angle. Difficulty is adjusted manually or automatically by changing the paddle size, ball speed, and opponent speed. Automatic difficulty increases if the hit rate is below 50% and decreases if the rate is above 70%. Performance is quantified by score, hit rate, and motor repetitions.
Skee ball: Players launch balls toward rings at speeds proportional to hand opening rate. Rings move, requiring control of hand opening speed. Difficulty is adjusted automatically or manually by changing ring size. Performance is quantified by score, accuracy, and motor repetitions.
Marble maze: Players rotate mazes to guide balls out into a bucket. Maze rotation is proportional to hand opening angle, targeting posture maintenance. Difficulty is adjusted by bucket size and adding multiple balls within a maze. Performance is quantified by completion time and score.
Future work will involve clinical trials. Treatment will include lab sessions (45 minutes VR+CCFES and 45 minutes OT) and home sessions (VR+CCFES for 2 hours daily).
Poster 43
Influence of Elbow Angular Spasticity Zones on One-Trial Motor Learning in Chronic Stroke
1McGill University, Montréal, Canada
2Feil and Oberfeld JRH/CRIR Research Centre, Jewish Rehabilitation Hospital Site of the Centre for Interdisciplinary Research in Rehabilitation of Greater Montréal, Montréal, Canada
3University of Montréal, Montréal, Canada
Motor learning studies in individuals with stroke have mainly focused on the less-impaired arm to separate confounding influences of motor deficits from motor learning capabilities. One previous study assessing motor learning (error correction) in the hemiparetic arm using a 1-trial learning paradigm showed that error correction strategies were related to both cognitive and arm motor impairment. Individuals with mild paresis used correction strategies resulting in successful 1-trial learning, whereas those with moderate to severe paresis used different strategies or were unable to completely correct movement errors. Zones in shoulder-elbow angular space in which muscles have spasticity and abnormal agonist/antagonist activation during voluntary movements have been identified by the measurement of joint angles corresponding to stretch reflex thresholds (SRTs). Incomplete correction strategies observed previously may be related to movements made beyond angular SRTs that elicit spastic resistance of muscles. This mechanism may confound the interpretation of motor learning studies. Our objective was to assess the influence of spasticity zones on motor learning in stroke. Participants with stroke (Fugl-Meyer scores: 32-61/66) made rapid 35° to 50° horizontal elbow extension movements from an initial 3° to a final 6° target in 16 blocks of 6 to 10 trials each. In session 1, movements were made in midrange, and in session 2, movements were restricted to a joint range that did not surpass the flexor SRT (outside of the spasticity zone). For each block, movements were alternatively not loaded or loaded by a position-dependent load (30% MVC). Participants were instructed to extend the elbow to the final target in a single fast and accurate movement and correct movement errors as soon as possible. Visual feedback of elbow position and movement speed was provided. Angular positions before correction were used to identify error correction strategies. Changes in load condition from load to no load and vice versa resulted in overshoot and undershoot errors, respectively. In session 1, participants corrected errors in 1 to 4 trials. When movements occurred outside of the spasticity zone, the number of trials needed to correct errors fell to 1 to 3, with the majority needing only 1 to 2 trials. The presence of spasticity may confound the results of motor learning studies and should be accounted for while interpreting study results.
Poster 44
Altered Obstacle Avoidance Behavior in Individuals With Good Arm Recovery After Stroke
1School of Physical and Occupational Therapy, McGill University, Montréal, Canada
2Feil-Oberfeld Research Centre, Jewish Rehabilitation Hospital site of Centre for Interdisciplinary Research in Rehabilitation of Greater Montréal, Montréal, Canada
3Centre Interdisciplinaire de Recherche en Réadaptation et Intégration Sociale, Québec City, Canada
4Département de Réadaptation, Université Laval, Québec City, Canada
5Centre de Recherche, Centre Hospitalier de l’Université de Montréal, Montréal, Canada
After stroke, individuals with good sensorimotor recovery of their affected arm report decreased use of the arm in activities of daily living. Decreased use of the affected arm may be associated with undetected motor deficits that are only identifiable when individuals attempt higher-order tasks that require complex planning and adaptation to produce appropriate interjoint and intersegment coordination. One higher-order motor task—the ability to avoid obstacles while reaching—commonly occurs in everyday environments but is not routinely assessed by clinical scales. We hypothesized that well-recovered people after stroke would be less successful in avoiding an obstacle in the reaching path compared with age-equivalent healthy controls. Obstacle avoidance ability during reaching in a virtual environment (VE) was compared between well-recovered stroke participants and healthy controls. A VE was developed simulating a grocery store aisle and a commercial refrigerator with sliding doors stocked with bottles on 2 shelves. Participants reached as fast as possible with their affected/dominant arm for a bottle on one shelf (nonobstructed reaching—template [T]). In random trials (RAND, 30% of 60 trials), the door ipsilateral to the reaching arm closed and partially obstructed the bottle at reach initiation. Participants were instructed to touch and retrieve the bottle without the hand or arm hitting the door. Arm and trunk movements were recorded with 24 active markers by an Optotrak system. Outcome variables were overall success rates, movement performance, and movement quality variables for T, successful (Succ), and failed (Fail) avoidance as well as Succ/Fail divergence points of the end point trajectory from the template profile (DP = percentage of reach distance). In T trials, stroke participants used less wrist flexion, wrist abduction, and shoulder rotation compared with controls. In RAND, 36% of controls and 12% of stroke participants were successful >65% of the time (z = 2.248; P < .05). For both groups, successful door avoidance was characterized by DP occurring closer to the starting position (Control: DPSucc = 11.2% ± 7.0%, DPFail = 34.1% ± 37.3%, P < .05; stroke: DPSucc = 20.5% ± 16.1%, DPFail = 60.4% ± 33.7%, P < .05). However, the margin of error in the stroke group was about half that of the controls. In addition, stroke participants had to significantly increase end point trajectory length compared with controls to successfully avoid the door. Stroke participants had residual movement deficits that were revealed through a challenging motor task. The potential of using challenging UL tasks to identify higher-order motor control deficits should be considered when assessing poststroke motor recovery.
Poster 45
Does Accelerometer-Based Feedback Increase Walking Activity During Inpatient Rehabilitation Poststroke? Preliminary Results From a Randomized Controlled Trial
1Toronto Rehabilitation Institute, UHN, Toronto, ON, Canada
2Heart and Stroke Foundation Centre for Stroke Recovery, Toronto, ON, Canada
3University of Toronto, Toronto, ON, Canada
Poster 46
Intensity-Dependent Effects of tDCS on Corticospinal Excitability in Chronic SCI
1Burke Medical Research Institute, White Plains, NY, USA
2Starlab Barcelona SL, Barcelona, Spain
3Burke Rehabilitation Hospital, White Plains, NY, USA
4Cornell University, New York, NY, USA
5University of Western Australia, WA, Australia
6Harvard Medical School, Boston, MA, USA
Poster 47
Neurophysiological Signatures of Proximal and Distal Recovery After Stroke
1Columbia University, New York, NY, USA
2Johns Hopkins University, Baltimore, MD, USA
3University of Zurich, Zurich, Switzerland
It is generally believed that the recovery of upper-extremity paresis in stroke patients is proximal to distal: arm segments regain strength before the forearm and hand.1,2 This observation suggests that distal arm strength has greater reorganizational demands, perhaps because of lower redundancy, before it can manifest. If this clinical assumption is true, then we would predict that the corticospinal integrity of a proximal muscle (biceps, BIC) would be preserved or emerge before that of a distal muscle (first dorsal interosseus, FDI). Here, we sought to identify the time courses of recovery of corticospinal integrity in proximal and distal cortical representations. We probed FDI and BIC corticospinal integrity in 14 first-time ischemic stroke patients at 1, 4, 12, 24, and 52 weeks after stroke. Corticospinal integrity in the lesioned and nonlesioned hemispheres was assessed with single-pulse transcranial magnetic stimulation at (1) 130% resting motor threshold (r130) and (2) 100% maximal stimulator output while patients produced approximately 20% maximum voluntary contraction (aCST). Our primary outcome was time until emergence of an identifiable motor-evoked potential (MEP; >40 µV) in FDI and BIC, arising from the lesioned and nonlesioned hemispheres. We performed a time-to-event analysis with α = .05. We categorized patients according to whether they had a simultaneous or nonsimultaneous FDI-BIC event. We examined if there was a simultaneous or nonsimultaneous emergence of FDI-BIC MEP for each test, comparing across hemispheres. We found no significant difference between hemispheres in time of FDI and BIC MEP onset for both r130 and aCST. These preliminary data suggest that in many cases, there may be no neurophysiological recovery gradient—that is, that the hand’s cortical integrity returns simultaneously with that of the biceps. We may find that a gradient of strength and/or control recovery exists behaviorally despite this simultaneity in MEP emergence, which would suggest that other neurophysiological processes may be responsible for recovery or that plasticity that we cannot directly measure (eg, in the rubrospinal tract) may mediate proximal recovery. Alternatively, it is possible that the chosen time intervals are too coarse to detect subtle longitudinal differences in MEP emergence between proximal and distal cortical representations.
Twitchell. 1951.
Colebatch et al. 1989.
Poster 48
Instrumented Gait Analysis to Differentiate Patients With Mild Cognitive Impairment
1California State University, Fresno, Fresno, CA, USA
2California Headache and Balance Center, Fresno, CA, USA
Poster 49
Turning Stability in People With Parkinson’s Disease
1Department of Electrical, Electronic, and Information Engineering—Guglielmo Marconi (DEI), University of Bologna, Bologna, Italy
2Department of Neurology, OHSU, Portland, OR, USA
Poster 50
Alterations in Upper-Limb Muscle Synergies With Level of Motor Impairment in Chronic Stroke Survivors
1Rehabilitation Institute of Chicago, Chicago, IL, USA
2Northwestern University, Chicago, IL, USA
Previous studies have shown that motor coordination may be accomplished by assembling task-dependent combinations of a few muscle synergies, defined here as a fixed pattern of activation across a set of muscles. Our recent study of chronic stroke survivors with severe motor impairment showed that some muscle synergies underlying isometric force generation at the hand are altered in the affected arm. To test the hypothesis that alterations in synergy structure vary with the level of motor impairment, we examined spatial patterns of elbow and shoulder muscle activation recorded during an isometric force target matching protocol performed by 24 chronic stroke survivors, evenly distributed across mildly, moderately, and severely impaired groups (Fugl-Meyer score >50, between 26 and 50, and <26 out of 66, respectively). Six neurologically intact, age-matched individuals served as the control group. We applied nonnegative matrix factorization to identify the underlying muscle synergies and compared their structure across groups. For all groups, spatial patterns of muscle activation could be explained by task-dependent combinations of only a few (typically 4) muscle synergies. Broadly speaking, the 4 synergies in the control group as well as the mildly and moderately impaired stroke groups were composed of elbow flexors, elbow extensors, shoulder abductors/extensors, and shoulder adductors/flexors, respectively. In contrast, the composition of muscle synergies exhibited consistent alterations in the group of severely impaired stroke survivors. Specifically, the anterior deltoid was coactivated with medial and posterior deltoids within the shoulder abductor/extensor synergy (deltoid synergy). Additionally, the shoulder adductor/flexor synergy was dominated by activation of the pectoralis major, with limited anterior deltoid activation. Overall, our results suggest that alterations in the muscle synergies underlying isometric force generation are confined mainly to stroke survivors with severe motor impairment.
Poster 51
Feasibility of Relating Continuous Monitoring of Turning Mobility to Fall Risk and Cognitive Function
1OHSU, Department of Neurology, Portland, OR, USA
2APDM Inc, Portland, OR, USA
Poster 52
Learning Strategies of Young Preterm and Term Infants
University of Southern California, Los Angeles, CA, USA
Poster 53
Measuring Cortical Physiology in Stroke Patients With Severe Arm Impairment
1Medstar National Rehabilitation Hospital, Washington, DC, USA
2Georgetown University, Washington, DC, USA
3Washington DC Veterans Affairs Medical Center, Washington, DC, USA
Poster 54
Time Course and Magnitude of Motor Learning During Gait Rehabilitation: A Preliminary Study
Emory University, Atlanta, GA, USA
Poster 55
From BCI to AAC and Back: Toward Practical Augmentative and Alternative Communication Systems Using Brain-Computer Interfaces
1Engineering, Brown University, Providence, RI, USA
2SpeakYourMind Foundation, Providence, RI, USA
3Brown Institute for Brain Science, Providence, RI, USA
4Center for Neurorestoration and Neurotechnology, Rehab R&D Service, Department of Veterans Affairs, Providence, RI, USA
5Neurology, Massachusetts General Hospital, Boston, MA, USA
6Neurology, Harvard Medical School, Boston, MA, USA
Augmentative and alternative communication (AAC) systems aim to restore communication for individuals with speech or language impairments using residual motion and adapted input devices. For individuals with severe motor impairments who lack the ability to use standard AAC access methods, brain-computer interfaces (BCIs) may offer an alternative by restoring effective communication using control signals extracted directly from the brain. Translating BCIs into useful, practical, and reliable AAC systems will require innovations in multiple areas. In addition to developing and testing investigational methods for intracortical control of external devices (eg, Hochberg et al1,2), we have also focused on creating customized user interfaces. The BrainGate Radial Keyboard is a novel typing interface composed of a radial key layout and a custom ambiguous text entry system. In a recent evaluation by 1 participant as part of the BrainGate Neural Interface System pilot clinical trials (IDE), the Radial Keyboard yielded higher accuracy and faster typing rates than a standard QWERTY keyboard. The participant, who had tetraplegia and anarthria (inability to speak) as a result of brainstem stroke, preferred to use the Radial Keyboard, indicating that it was “easier to use.” To assess the potential utility of BCIs for practical communication purposes, direct comparisons between BCIs and currently available AAC systems are needed. In addition to our ongoing work in creating reliable neural decoders for future use in AAC, we have also recognized an opportunity to leverage affordable off-the-shelf hardware to create improved non-BCI AAC devices, and our academic labs have spun off a nonprofit organization called the SpeakYourMind Foundation to create and support such personalized AAC solutions.
Hochberg et al. Nature. 2006.
Hochberg et al. Nature. 2012.
Poster 56
How Does the Brain Support the Recovery of Hand Function for Grasping Following Surgical Hand Replantation or Transplantation?
1University of Missouri, Columbia, MO, USA
2Christine M. Kleinert Institute, Louisville, KY, USA
Traumatic amputation of a hand can be reversed through advanced surgical replantation (own hand) or transplantation (other’s hand) procedures. Remarkably, with time and practice, these patients can recover function of the replanted/transplanted hand for grasping and manual interaction with objects. To date, all previous studies of hand replant/transplant patients have investigated whether reestablishment of afferent/efferent traffic between the hand and brain leads to the reversal of amputation-related reorganizational changes in the primary sensory and motor cortex. The goal of the current work is to better understand how the brain supports the functional recovery of manual grasping in these patients. We hypothesize that the recovery of grasp relies on the recruitment of the anterior intraparietal cortex, a region known to be critically involved in the sensorimotor transformations necessary for grasp. As patients regain function of their replanted/transplanted hand for grasping, brain activity associated with grasping with their replanted/transplanted hand is predicted to closely match activity observed for grasping in healthy controls. To test these predictions, we used functional MRI to characterize brain activity associated with grasping using a transplanted hand in patient DR, a 38-year old, right-hand dominant man who suffered traumatic amputation of his left hand in 1998 and underwent allogeneic hand transplantation 13 years later, and patient WH, a 60-year old, right-hand dominant man who had his own left hand surgically reattached hours after traumatic amputation in 2008. During object grasping with their replanted/transplanted hand (versus a control task involving object touching with the knuckles of the hand, without grasping), both patients showed robust activity within the right (contralateral) sensorimotor cortex, spanning the central and postcentral gyri/sulci. The location of this activity was inconsistent with our a priori hypothesis, positioned anterior to the location of the anterior intraparietal cortex known to be important for grasping. For both patients, grasp activity partially overlapped with activity defined for the replanted/transplanted hand using separate sensory and motor paradigms. Likewise, independent region-of-interest analyses showed that motor- and sensory-defined areas for the left hand were more strongly activated for grasping with the left replanted/transplanted hand versus the control task. To compare with these results, we are currently collecting data from healthy age-, gender-, and handedness-matched control participants using the same grasping, sensory, and motor paradigms. Our goal is to not only characterize grasp activity in healthy control participants at the group level but to also understand the range of individual-participant variation for comparison with individual patient results. Notably, both patients tested here also showed evidence of sensory dysfunction for their replanted/transplanted hand and, to compensate, relied heavily on visual feedback during grasping. It is possible that the current results reflect atypical grasp activity, indicative of incomplete functional recovery of the replanted/transplanted hand for grasping.
Poster 57
Wearable Audio-Biofeedback System for Gait Rehabilitation of Individuals With Parkinson’s Disease
1DEI, University of Bologna, Bologna, Italy
2Department of Rehabilitation Sciences, University of Leuven, Leuven, Belgium
Nieuwboer et al. 2007.
Ferrari et al. 2013.
EU-FP7/2007-2013 Grant Agreement No. 288516 (CuPiD project).
Poster 58
Correlation Between Gait Parameters and Balance in Very Young Children
1Department of Pediatrics, West Virginia University, Morgantown, WV, USA
2Biostatistics, West Virginia University, Morgantown, WV, USA
The purposes of this study were the following: (1) to identify gait parameters that particularly correlate with motor abilities in young children and (2) to establish standardized normative gait data for future analysis of children with pathological gait.
Poster 59
Changes in Resting-State Functional Connectivity Following BCI-EEG-Based Intervention in Subacute and Chronic Stroke Patients
University of Wisconsin–Madison, WI, USA
Poster 60
Peripheral Nerve Injury Induces Long-Term Synaptic Depression in the Rat Somatosensory Cortex
1Kennedy Krieger Institute, Baltimore, MD, USA
2Johns Hopkins University School of Medicine, Baltimore, MD, USA
Evidence suggests that peripheral nerve injury alters the function of the sensory-motor cortices and that these changes may account for sensory dysfunctions often occurring after the injury. We previously demonstrated that peripheral nerve injury results in immediate1 and long-term2-4 changes in the function of the primary somatosensory cortex (S1) contralateral to the injured limb. Specifically, we found that the injury is accompanied by increases in the activity of inhibitory interneurons, a phenomenon that appeared to be mediated via the transcallosal pathway. However, the cellular mechanisms leading to this plasticity remain unknown. The goal of this research was to investigate the mechanism by which the transcallosal fibers affect neuronal activity after injury. Because one of the main inputs into cortical layer V is transcallosal fibers, whole-cell patch clamp recordings from identified pyramidal (excitatory) neurons in layer V in S1 were collected with and without stimulation of the transcallosal projections. In control and limb-denervated rats, we studied long-term potentiation (LTP), which is one of the major cellular mechanisms that underlie learning and memory. The amplitude of the excitatory postsynaptic currents was measured for 30 minutes following high-frequency (100 Hz) stimulation of the transcallosal fibers. The results demonstrated that stimulation of the transcallosal fibers induced a 36.93% ± 1.76% amplitude increase in layer V neurons in control rats (n = 11 neurons) but a −16.97% ± 1.0% amplitude decrease in denervated rats (n = 9). Thus, the results suggest that injury decreased the probability of inducing LTP and increased the probability for long-term synaptic depression. The increases in inhibitory interneuron activity that we have identified previously might account for the increase in synaptic depression in S1. Increases in synaptic depression associated with peripheral nerve injury might be manifested in the postinjury sensory dysfunction that patients often suffer from.
Han et al. Neurorehabil Neural Repair. 2013.
Pelled et al. Neuroimage. 2007.
Pelled et al. PNAS. 2011.
Li et al. PNAS. 2011.