Abstract
Poster 1
Effect of Prism Adaptation Treatment on Reading Performance in Right-Brain Stroke Survivors
1Kessler Foundation, West Orange, NJ, USA
2Rutgers New Jersey Medical School, Newark, NJ, USA
3Hunter College, the City University of New York, New York, NY, USA
Spatial neglect is a common disorder following unilateral brain injury such as a stroke. After right-brain damage, patients with spatial neglect typically show symptoms in the egocentric frame of reference. Some patients also manifest bias in the allocentric frame of reference. A recent study (Gossmann et al., 2013, Neurorehabilitation and Neural Repair) suggests that prism adaptation treatment (PAT) selectively improves egocentric but not allocentric bias in a cancellation test. In the present study, we examined whether this selective effect can be replicated in a reading task.
We studied reading performance in 25 patients (6 females) with spatial neglect. Patients were on average 69.3 years old (SD = 10.7) and 23.1 days post stroke (SD = 12.2) at the time of baseline evaluation. Patients read a 3-column article and a 2-page food menu (both subtests from the Behavioral Inattention Test). Over the course of the study, patients alternately read two versions of articles and menus seven times: prior to PAT (baseline), after one week of PAT, after two weeks of PAT, and weekly after PAT for 4 weeks. At baseline, there were more errors on the left column of the article and on the left page of the menu, consistent with egocentric neglect. In addition, within each column or page, the error rate was greater on the left half than the right. Interestingly, there were interactions between egocentric and allocentric bias such that the allocentric bias was greater in the right column of the article and in the right page of the menu. Over the course of the study, there was improvement in both article and menu readings (i.e., error rates reduced over time). 25% of the participants reached the normal range in reading performance after PAT. However, the degree of left-right asymmetry did not change in either egocentric or allocentric reference frames.
Consistent with the literature, our findings suggest that PAT has a therapeutic effect on reading. Inconsistent with the previous study, PAT does not always improve egocentric bias when assessed in a reading task. This pattern of results supports the view that PAT reduces lateralized errors in activities involving motor/manual components, but PAT may not improve left-right asymmetric performance in activities such as reading.
Poster 2
Activity of the Action Observation Network with Lower Extremity Use in Chronic Stroke Participants
1Brain and Creativity Institute, Los Angeles, CA, USA
2Occupational Science and Occupational Therapy, Los Angeles, CA, USA
3Biokinesiology and Physical Therapy, Los Angeles, CA, USA
4University of Southern California, Los Angeles, CA, USA
Neurorehabilitation after stroke, composed of physical and occupational therapy, can be a difficult process for patients, requiring time, energy, and effort. Studies suggest that action observation (AO) therapy may be modestly effective in improving motor ability after stroke. The Action Observation Network (AON), including the posterior inferiofrontal gyrus, rostral inferiopartietal lobule, and posterior superior temporal sulcus, may be implicated in AO therapy. In this study, we aim to investigate if a lower extremity effector (foot) can recruit regions of the AON, measured via functional magnetic resonance imaging (fMRI), during certain conditions of observing an action, imitating an action, and/or executing an action. Actions were composed of plantar flexion, heel rotation, and abduction or adduction similar to pressing on a gas pedal at an angle. These actions were shown as videos or photos to chronic (> 3 months) mild-to-moderately affected left middle cerebral artery stroke (n=16) and non-disabled control (n=13) participants. While watching these videos on a projector, both groups of participants were instructed to observe, imitate, or execute the shown action(s) to the best of their ability. We hypothesize that the non-disabled controls will recruit regions of the AON and that stroke participants will also recruit similar brain regions, yet this activity will be modulated by their degree of motor ability, measured by the Fugl-Meyer Exam. Preliminary results confirm that, the AON is recruited in non-disabled controls during conditions of observation, imitation, and execution of lower extremity movements, however, to differing degrees. Activity in the AON for participants with hemiparesis resembles these patterns from non-disabled controls, yet it is modulated by motor capability. Understanding which conditions and/or effectors engage the AON may help in determining types of conditions or effectors that are best used for AO therapy. This is an ongoing study and further analysis is needed to understand potential, if any, application of this technique to neurorehabilitation.
Poster 3
An iPod-based dual-task test for differentiating freezers and non-freezers in Parkinson’s disease
1University of Calgary, Calgary, Alberta, Canada
2University of Lethbridge, Lethbridge, Alberta, Canada
Poster 4
Cognitive Processing Speed, Mood, and Fatigue Improvements in Persons with Multiple Sclerosis Treated with Dalfampridine
1Mandell Center for Multiple Sclerosis, Mount Sinai Rehabilitation Hospital, Hartford, CT, USA
2Department of Community Health, Brown University, Providence, RI, USA
3Departments of Neurology and Engineering, Brown University, Providence, RI, USA
Poster 5
Reliability and validity of a new forward reach assessment tool in a sample of healthy controls
1Mandell Center for Multiple Sclerosis, Mount Sinai Rehabilitation Hospital, Hartford, CT, USA
2Departments of Neurology and Engineering, Brown University, Providence, RI, USA
Poster 6
A Pilot Randomized Trial Comparing Inter-session Scheduling of Biofeedback Results to Individuals with Chronic Pain: Influence on Psychological and Physical Function
St. Luke’s Rehabilitation Institute, Spokane, WA, USA
Chronic pain is an intractable condition estimated to affect more than a third of the US population. Multidimensional treatment for chronic pain often includes biofeedback therapy to assist in development of pain self-management skills. Acquisition of pain modifying strategies through biofeedback training results from learning to volitionally control a physiological signal, that is, psychomotor learning. Research on variables that influence psychomotor learning has examined the schedule for providing external feedback within a training session that is optimal for long-term retention of skill. The majority of these feedback-scheduling studies have supported a paradoxical relationship that scheduling methods that reduce the frequency at which external feedback is provided to a learner during training (acquisition phase) result in greater levels of skill in retention tests separated from acquisition. Based on this notion, the relative frequency of feedback provided in acquisition training, such as continuous visual feedback offered in biofeedback training, should be reduced as early as possible in acquisition to reduce the likelihood for dependence on external feedback to control the biological process. To assess this premise, it was the purpose of this study to conduct a randomized pilot study comparing two biofeedback schedules for encouraging retention of pain modification ability, and improvement in physical and psychological functioning in people with chronic pain.
Participants engaged in heart rate variability (HRV) biofeedback training for 9 sessions, and then were contacted after 3 months to collect follow-up information. Two groups were compared: The Faded Feedback group received a high relative frequency of HRV biofeedback early in acquisition, with the amount of biofeedback systematically reduced across sessions so that by the final training session, participants were attempting to control HRV without external feedback. In contrast, the Full Feedback group received HRV biofeedback continuously while attempting to control HRV across the acquisition phase. Participants were adult patients participating in individual outpatient cognitive behavioral therapy for pain management. Several measures were used to evaluate the impact of the biofeedback schedules: pain intensity, fear-avoidance beliefs, self-report physical functioning, and post-program use of biofeedback skills.
The Faded Feedback schedule, compared to Full Feedback schedule, resulted in greater use of biofeedback skills post-program, and improved pain intensity and fear-avoidance beliefs at the conclusion of biofeedback training and 3-months following training. No differences were noted between groups for physical functioning.
This pilot study provided evidence that reducing the frequency of external visual feedback given to the learner with chronic pain in biofeedback training was associated with several positive outcomes over an extended period of time following training. Thus, over-presentation of external feedback during biofeedback training can inhibit the ability of people with chronic pain to retain skill in the long-term. We will use these pilot results to plan a larger clinical trial.
Poster 7
Interhemispheric interactions between axial and proximal arm muscle representations of the primary motor cortex
1University of Montreal, Montreal, Quebec, Canada
2Centre de recherche Interdisciplinaire en Réadaptation, Institut de Réadaptation Gingras-Lindsay de Montréal, Montreal, Quebec, Canada
After a unilateral stroke, 30% of individuals have decreased arm function. One mechanism that can support recovery of the arm after stroke is the reorganisation of interhemispheric interactions. Most studies of these interactions, as well as rehabilitation strategies, focus on the hand. However, postural control of proximal arm and trunk muscles are essential to elicit purposeful arm movements. As such, cortical areas involved in the control of these movements are likely to also play a crucial role in recovery of arm after stroke. To date, little is known about the interhemispheric dialogue between cortical representations of proximal arm and axial muscles in M1 of the two hemispheres, and how they change after stroke. Hence, in this study we used two different transcranial magnetic stimulation (TMS) paradigms to compare interhemispheric interactions between axial and proximal arm muscles in healthy volunteers. In the first paradigm, we assessed the ipsilateral silent period (iSP) in right Anterior Deltoid (AD) and Erector spinae at L1 (ES L1) during right isometric shoulder flexion in 9 subjects. In this task, AD was an agonist and ES L1 a postural muscle. As a control, we also assessed the iSP in a distal hand muscle, the right First Dorsal Interosseous (FDI), during isometric abduction of the index. The iSP was elicited by stimulation of the motor cortex ipsilateral to the recorded muscles with current intensities of 130% of active motor threshold (AMT). We found that iSP could be clearly elicited in FDI and AD in 6 subjects, and in ES L1 in only 3 subjects. The iSPs area for FDI, AD and ES L1 were not significantly different from each other (respectively 67%; 71%; 67 % of baseline EMG). Ipsilateral motor evoked potential (iMEP) could also be evoked, mainly in ES L1. In the second part of this study, a paired pulse paradigm was used in 5 subjects performing either isometric index abduction (FDI targeted) or shoulder flexion (AD and ES L1 targeted) at 10% of the maximum voluntary contraction. During contraction, a conditioning stimulus (CS; 1.3AMT) was applied to M1 ipsilateral to the recorded muscles. Following a 10 ms delay, a test stimulus (TS; 1.2-1.3AMT) was applied to M1 of the other hemisphere. The CS decreased the amplitude of the MEP evoked by the TS in FDI (90% of test MEP) and AD (82%). Contrastingly, it increased the MEP in ES L1 (135%). Our preliminary results suggest that whereas distal and proximal arm muscles (FDI, AD) seem to primarily exert an interhemispheric inhibition on their homolog, stabilizing axial muscles may have more facilitatory interactions. Supported by The Sensorimotor rehabilitation research Team (SMRRT-CIHR)
Poster 8
Attitudes of clinicians towards spasticity assessment
1Department of Rehabilitation, Universite Laval, Quebec city, Canada
2School of Physical and Occupational Therapy, McGill University, Montreal, Canada
3Feil and Oberfeld Research Center, Jewish Rehabilitation Hospital, Center for Interdisciplinary Research in Rehabilitation of Greater Montreal, Montreal, Canada
4Department of Physiotherapy, Faculty of Health Sciences, Manipal University, Manipal, India
Considering that the Modified Ashworth Scale (MAS) is the best-known outcome measures to clinically estimate spasticity, participants were asked to evaluate their level of satisfaction with the MAS. Among the participants who answered that they would use the MAS, 90.1% reported that they were satisfied or somewhat satisfied with this spasticity assessment.
Poster 9
Does task engagement level affect how we move?
University of Southern California, Los Angeles, CA, USA
High subjective engagement level is consistently found when motor tasks are performed under virtual or augmented reality conditions. In previous work, attentional demand was greater during a reach task to a virtual target compared to a real target. More importantly, engagement level was generally higher in the more difficult virtual compared to the easier real target condition. This study investigated the impact of engagement level on movement strategy by comparing reach kinematics for those with low- or high-engagement ratings during virtual target (VT) and real target (RT) conditions.
Using a within-subject design, 15 participants (59-88 yrs old) performed reach movements to targets arranged in a circular 8-target pattern; there were 48 reaches per condition and each from standing and stepping. A KinectTM camera positioned in front captured each reach; position data were resampled at 12Hz. Movement onset/offset was defined as the first time at which the tangential velocity of the wrist virtual marker was above/below 1 cm/s. Kinematic variables included movement time (MT), time to peak velocity (TTPV), and time after peak velocity (TAPV). Participants’ self-assessed engagement to each condition was acquired using a 5-level Likert scale for a single item administered after reaching. Participants were categorized into low-, neutral- and high- engagement (E) groups based on scale scores.
Reaches from standing to VT for participants with high-E showed longer MT than those with low-E; the longer MT was attributed to longer TAPV. Interestingly, MT for reaches from stepping to VT were independent of engagement. However, in spite of similar MT, the high-E group demonstrated shorter TTPV and longer TAPV than the low-E group. For easier reaches to RT, there was no evidence that engagement level modulated any kinematic measure.
These findings provide indirect evidence that engagement level may impact movement strategy, but only for relatively difficult tasks. Reaches to RT were relatively easy (from open-ended post-interview); RT reaches exhibited a symmetric velocity profile suggestive of a default strategy regardless of engagement level. In contrast, during more challenging VT reaches, a high-E level was associated with slower movements and a prolonged homing-in phase. We speculate that high-E may harness more attentional resources toward goal achievement. These preliminary group level findings suggest an interaction between engagement level and task difficulty on movement strategy. Further analysis of individual-subject variability within engagement level groups is needed to confirm or refute the group level analysis.
Poster 10
Remote Limb Ischemic Conditioning Enhances Motor Learning
Washington University, St. Louis, MO, USA
Poster 11
Overcoming hurdles towards maximally effective transcranial direct current stimulation
Medical University of South Carolina, Charleston, SC, USA
Transcranial direct current stimulation (tDCS) literature shows mixed opinions on efficacy, which can be attributed to lack of universally agreed dose amplitude, duration and regimens individualized according to skull thickness, subcutaneous fat, skull defects, etc. It is noteworthy that tDCS charge densities that are safely used in animal studies are two orders of magnitude higher than those used in humans1 (~4 C/cm2 in animals2 versus ~0.06 C/cm2 in humans3). While tDCS commonly has minor adverse effects4, low current amplitudes (≤2 mA) for short durations (≤30 min/session) are used to avoid temporary concerns like skin redness and irritation in select patients. Skin being the first contact point that transmits highest electric currents and fields, increased doses can result in severe skin reactions. Here we explain why such limitations have not been observed in animal studies. First, translational studies are performed in smaller animal models like rodents and cats, making absolute distance between electrodes, and therefore electrical path resistance, significantly lesser than that in humans. Second, human skin has higher resistance than other species5 resulting in more conversion of electrical energy into heat and potentially consequent inflammation. Third, tDCS has been applied epicranially in select rodent studies bypassing skin resistance at anode1,6, which cannot be implemented through non-invasive clinical trials. Some of these limitations can be overcome by improving electrical conductivity of electrode-body interface that is stable over the course of tDCS session. It is not uncommon for saline-soaked sponges to offer uneven conductivity as an electrode contact area because of gravity-driven pooling of saline and drying from physical squeezing. Silicon-based lubricant gels, e.g., ones used for ultrasound imaging, offer good contact but are not good electrical conductors. Also, Hydration of the skin is not a major factor determining conductivity of the skin by tDCS (unlike AC current) because capacitance, but not resistance/conductance, of the skin is a function of hydration7. Therefore, using high-salt gel electrodes8 on scrubbed9, moisturized10 skin or other ways to improve conductivity11 can ensure less skin injury with stronger intracranial electrical fields without neural injury12.
Liebetanz, D. et al. Clinical Neurophysiology 2009.
Brunoni, A. R. et al. Rev Neurosci 2011.
Marquez, J. et al. International Journal of Stroke 2013.
Brunoni, A. R. et al. Int J Neuropsychopharmacol 2011.
Davies, D. J. et al. Toxicol In Vitro 2004.
Liebetanz, D. et al. Epilepsia 2006.
Jemec, G. B. & Serup, J. Acta Derm Venereol 1990.
Minhas, P. et al. J Neurosci Methods 2010.
Türker, K. S. Physical Therapy 1993.
Rim, J. H. et al. Clin Exp Dermatol 2005.
Farinha, A. et al. Biomedical Instrumentation & Technology 2006.
McCreery, D. B. et al. IEEE Trans Biomed Eng 1990.
Poster 12
Development Of A Simple App For Stepping Height Measurements During Rehabilitation Training In Parkinson’s Disease
Dept. of Clinical Neurosciences, Hotchkiss Brain Institute. Faculty of Medicine, University of Calgary, Calgary, AB, Canada, Canada
Acknowledgments: CIHR, AIHS, MITACs, Parkinson Alberta Society, the Movement Disorders Clinic-AHS, and the following Brazilian agencies: National Council for Scientific and Technological Development (CNPq) and Coordination for the Improvement of Higher Education Personnel (CAPES).
Poster 13
Repetitive transcranial magnetic stimulation to modulate cortical excitability after stroke: effect of participant characteristics
1Division of Biokinesiology and Physical Therapy, University of Southern California, Los Angeles, CA, USA
2Department of Neurology, University of Southern California, Los Angeles, CA, USA
3Department of Radiology, University of Southern California, Los Angeles, CA, USA
A stroke-induced change in interhemispheric interaction (IHI) may increase the inhibition from the non-lesioned hemisphere toward the lesioned hemisphere, thereby impeding recovery of the paretic extremities. One plausible method to improve functional recovery is to decrease the excitability of the non-lesioned hemisphere and restore IHI balance by applying low-frequency repetitive transcranial magnetic stimulation (rTMS). Although low-frequency rTMS has been demonstrated to be safe and feasible for individuals with stroke, its effectiveness in modulating the excitability of the non-lesioned hemisphere and improving paretic hand function has not been consistently reported. It is essential to understand the factors that may impact the efficacy of rTMS, such as TMS methodology and participant characteristics, in order to optimize its application in stroke rehabilitation. The purpose of this study was to explore the potential participant characteristic factors with our preliminary data. Ten individuals with sub-acute stroke received active and sham rTMS in two consecutive days with a counterbalanced order. In the active condition, 1-Hz rTMS was applied over the representation area of extensor digitorum communis in the non-lesioned primary motor cortex (M1) at 90% resting motor threshold for 25 minutes. The excitability of the non-lesioned M1 was measured before and after rTMS using resting motor evoked potential (MEP) amplitude. Participants were divided into two groups (responders and non-responders), based on their MEP changes following active rTMS. Participant characteristics, such as age, stroke duration, lesion location and initial hand function, as measured by upper extremity Fugl-Meyer assessment (UE-FMA), were compared between the two groups. Among the ten participants, three showed down-regulation of excitability following active rTMS and were categorized as responders. No significant difference in stroke duration, initial hand function and lesion location was found between responders and non-responders (p>0.05). The responders appeared to be younger than the non-responders with a borderline significance (p=0.07). Further analysis showed that age was significantly correlated with the amount of change in excitability (r=0.85, p=0.02), suggesting that the efficacy of inhibitory rTMS was greater in young participants than old participants. Our preliminary data suggests that the ability of rTMS to modulate brain excitability may be a function of age in stroke population. The findings attest to the further need to determine the subject-specific characteristics that would enable prediction of good outcome following rTMS.
Poster 14
Combining Theta Burst Stimulation with Reaching Practice in Individuals with Severe Post-Stroke Arm Impairment: Two Case Studies
1Georgetown University Medical Center, Washington, DC, USA
2Georgetown University, Washington, DC, USA
3Medstar National Rehabilitation Hospital, Washington, DC, USA
Chronic severe arm impairment after stroke is costly and can greatly impact quality of life. There are few validated treatment methods for improving arm function in those with severe impairment and existing approaches often have a low ratio of outcome to effort. For example, repetitive reaching practice often results in improved performance, even in stroke patients with relatively severe motor impairment; however, the amount of improvement is small compared to the amount of time and effort required. It may be possible to enhance the effects of practice by applying non-invasive brain stimulation, such as transcranial magnetic stimulation (TMS), prior to reaching practice to increase the excitability of brain areas that could contribute to proximal arm recovery. Preliminary work in our lab has suggested that dorsal premotor cortex (PMd) of the intact hemisphere may contribute to affected arm reaching to a greater degree than neighboring cortical areas. We aim to determine if upregulation of intact PMd could enhance the effects of reaching practice in persons with severe arm impairment after stroke. Here, we report the results of the first two participants. At the first visit, a double-pulse TMS disruption was applied over intact PMd during a reaching response time task. At the second visit, participants performed a reaching response time pre-test before receiving intermittent theta burst stimulation (iTBS) over intact PMd at 80% of the active motor threshold of primary motor cortex (20 trains of 10 TMS bursts delivered at 5 Hz with 8 seconds between each train). Immediately after stimulation, participants performed an intensive session of repetitive reaching practice followed by
a reaching response time post-test. At the third and final visit, participants performed the reaching response time post-test followed by the TMS disruption test. iTBS and practice were well-tolerated by both participants. Participants 1 (P01) and 2 (P02) both showed improved reaching response time after iTBS + practice (P01 was 35% faster, P02 was 16% faster). Kinematic Improvements were also evident (hand path length (16% decrease for P01, 13% decrease for P02), maximum directional error (8% decrease for P02), initial directional error (12% decrease for P02) and number of movement units (50% decrease for P01, 55% decrease for P02)). Both participants also showed larger disruption effects after iTBS and practice suggesting greater recruitment of intact PMd during performance of reaching movements. In conclusion, these early results suggest that reaching practice preceded by iTBS to intact PMd could improve reaching performance in persons with severe arm impairment after stroke and may increase recruitment of intact PMd when performing affected arm reaching movements.
Poster 15
Using dual-task paradigms to detect motor learning effects in older adults after task-specific training: A feasibility study
1Utah State University, Logan, UT, USA
2University of Utah, Salt Lake City, USA
Poster 16
Enhancing activation of object use information with transcranial direct current stimulation
Moss Rehabilitation Research Institute, Elkins Park, PA, USA
Limb apraxia is a disorder of complex skilled action not attributable to weakness, incoordination, or other elemental sensory or motor impairments. Although it has a significant negative impact on activities of daily living, a number of recent evidence-based reviews indicate that apraxia following stroke is amenable to rehabilitation. In this study we tested whether non-invasive brain stimulation (transcranial direct current stimulation; tDCS) could be used to enhance functioning within the brain regions responsible for object use in neurologically intact participants, with the goal of applying the results to apraxia rehabilitation in the future. We have previously shown that when looking for a target object in an array with 3 non-target objects, neurologically-intact controls looked at use-related non-target objects significantly more than at unrelated non-target objects, even though object use was irrelevant to the task. This finding indicates that action representations associated with object use may be incidentally activated during visual object processing. Activation of these use representations are temporally delayed in individuals with apraxia, with the slowing predicted by lesions to the inferior parietal and posterior temporal lobes. Based on this finding, in the present study we applied anodal tDCS to those regions to determine whether stimulation would speed use activation. Results showed that use activations were speeded with tDCS (relative to sham tDCS), especially when contextual cues for object use were weak. In addition, the time to complete the search task was speeded with tDCS relative to sham. We discuss these results in terms of a neuroanatomic model of object and action processing, and the potential for using tDCS for rehabilitation of apraxia.
Poster 17
Enhancing the mirror illusion with transcranial direct current stimulation
1Moss Rehabilitation Research Institute, Elkins Park, PA, USA
2University of Pennsylvania, Philadelphia, PA, USA
The perceptual-motor system’s estimate of the current arm position is strongly dependent on visual feedback. One compelling example of this phenomena is the mirror illusion (MI). To experience the MI, a participant places her arms on either side of a vertically-oriented mirror, with only the arm on the reflective side of the mirror being visible. When the mirror is placed midway between the two arms, the reflection of the visible arm (viewed in the mirror) appears to be in the same location as the arm behind the mirror. Thus, the MI setup creates a compelling illusion in which visual feedback about movement of the arm behind the mirror comes from movement of the visible arm. This false visual feedback from the MI has been shown, over multiple sessions, to reduce post-stroke movement deficits. During this “mirror therapy”, the impaired limb benefits from the illusory visual feedback from the unimpaired limb. Thus, it is highly probable that over time the MI leads to changes in bilateral interactions between motor networks within the two hemispheres. We hypothesized that the MI would be affected when we modified interhemispheric interactions using transcranial direct current stimulation (tDCS). Specifically, we tested whether bilateral tDCS stimulation to the primary motor cortices (anode-right-cathode-left and anode-left-cathode-right) would modify the MI, as measured using a previously-developed method of objectively quantifying the strength of the MI (Holmes & Spence, 2005). In this method, participants make reaching movements with the unseen arm behind the mirror while viewing the reflection of the other arm. When an offset in the positions of the two limbs relative to the mirror is introduced, the effects of the mirror are manifested in the bias of the reflected arm’s position reaching of the unseen arm. We found that active tDCS in the anode-right-cathode-left montage increased the magnitude of the MI relative to sham tDCS. We take these data as a promising indication that tDCS could improve the effect of mirror therapy in patients with hemiparesis.
Poster 18
Manual Asymmetry in motor skill learning
Moss Rehabilitation Research Institute, Elkins Park, Pennsylvania, USA
Motor skill learning involves a process wherein one learns to synthesize novel movement capabilities in absence of perturbation such that they are able to perform and retain the movement skill with higher accuracy, consistency, efficiency, coordination and flexibility. Performance asymmetry has been reported in motor control and adaptation to kinematic and dynamic perturbations of goal-directed actions. In this study, we investigated asymmetry in acquisition and retention of a complex motor skill that requires speed and accuracy for optimal performance. Further, we examined if skill learning asymmetry is influenced by arm dominance. 6 right-handed (RH) and 6 left-handed (LH) adults practiced 2 distinct tracks during different sessions separated by 2-4 weeks. During separate sessions, participants practiced with their dominant (DOM) or non-dominant (NDM) arm in a pseudo-randomized order. Performance changes during practice were characterized by improvements in accuracy while practicing within prescribed movement time ranges. Learning was quantified by changes in the speed-accuracy tradeoff (SAT) function measured at baseline and a day after practice ended. There were no baseline differences in the SAT function between the DOM and NDM arms. All participants improved their performance with their DOM and NDM arms with practice. With practice, the RH participants demonstrated significantly higher improvements in the SAT function for the dominant compared to the non-dominant arm. We did not find significant differences between the two arms for the LH group. With the limited practice provided for a complex motor skill, right-handed individuals demonstrate dominant arm advantage for learning. Extended practice may be indicated for improving skill learning with the non-dominant arm in the RH individuals. In RH individuals, asymmetry in skill learning may likely be related to the hemispheric specialization effects reported for motor control.
Poster 19
Bimanual coordination for functional tasks in patients post-stroke
Moss Rehabilitation Research Institute, Elkins Park, Pennsylvania, USA
Most, if not all activities of daily living, require some degree of collaboration between two hands. While unimanual functional measures are widely used in stroke rehabilitation research, little is known about how functional bimanual movements are affected in patients post-stroke. The aim of this study was to investigate the kinematics and coordination between two arms as patients with stroke and age-matched controls performed symmetric and asymmetric functional bimanual activities. We hypothesized that patients with stroke will demonstrate significant deficits in asymmetric, but not symmetric functional bimanual tasks compared to age-matched controls. 6 patients with mild- to- moderate impairments (Fugl-Meyer 30-55) after unilateral stroke and 6 control participants participated in the study. Participants performed 20-trial block of each of the three tasks: (1) unimanual goal-directed task (UNI), (2) symmetrical bimanual task (BISYM) which comprised of picking a box using both hands, and (3) asymmetrical bimanual task (ASYM) which comprised of opening a drawer with their non-paretic hand and pressing a button with their paretic hand. Trajectories of each hand were recorded using a 3-D motion recording system with 2 electromagnetic sensors placed on the dorsum of each hand. Kinematic measures included: movement time (MT), maximal velocity (Vmax), directness of trajectory (DT) measured as trajectory/distance ratio. Bimanual coordination was characterized by the time delay between the two hands as they performed the task. Both controls and patients with stroke demonstrated strong spatial and temporal coordination while reaching the box during the BISYM task, but patients demonstrated significant deficits while picking up the box. While coupling the two arms using “in-phase” movements may help reaching; synergistic coordination necessary for picking up the box is impaired in patients post-stroke. For the ASYM task, controls demonstrated greater temporal coordination compared to the patients, with a significantly greater time-lag between the paretic and non-paretic hand (p<0.05). Importantly, time-lag between the two hands in patients with stroke did not correlate with the movement time of the paretic arm during the UNI task. Our preliminary results demonstrate a task-based dissociation in bimanual coordination in patients post-stroke. Patients demonstrate deficits in coordination between two arms during execution of an asymmetric and synergistic aspect of symmetric bimanual task. The deficient bimanual coordination in the ASYM condition was not strongly related to deficits in unimanual control. This likely suggests different control mechanisms for unimanual and bimanual control in patients post-stroke. In addition, this may also indicate (but yet to be empirically established) that unimanual skill practice may not always translate to bimanual skill performance in patients post-stroke.
Poster 20
MEG-based functional connectivity changes with motor imagery training: evidence for motor imagery as an acquired skill
1Dalhousie University, Halifax, NS, Canada
2IWK Health Sciences Centre, Halifax, NS, Canada
Motor imagery (MI) is a form of practice in which an individual mentally rehearses a motor task, facilitating skill acquisition in the absence of physical practice (Jeannerod, 1995). MI has many clinical applications, including brain-computer interface and stroke rehabilitation. Previous research suggests that skill acquisition via MI is facilitated by repetitive activation of brain regions in the sensorimotor network similar to that observed in motor execution (ME; Hetu, 2013). This activation is influenced by differences in one’s ability to perform MI (Wei, 2010), suggesting that MI ability is an acquired skill. In fact, spatial activation patterns during MI become more similar to that of ME with training. These training-related changes in MI activation have yet to be investigated from a network connectivity perspective. In showing how MI training drives network changes, the current study further demonstrates that the ability to employ MI for skill acquisition is a learned skill. Non-disabled participants (N=10; 24.7 +/- 3.8 years) performed both ME and MI of a unilateral seven-sequence button press task over three days. Magnetoencephalography (MEG) was utilized to capture neural activity. Coherence-based functional connectivity analysis was examined between eighty cortical nodes in the beta frequency band (15-30 Hz) and a partial least squares analysis conducted to compare coherence-based functional connectivity between session 1 and 3 (McIntosh, 2004). Preliminary findings indicate that functional connectivity was altered as a function of session, with network composition differing (p < 0.05) from session 1 to 3. Network changes included more lateralized (ie. in the contralateral hemisphere) activity during MI with more similar patterns of activity observed between MI and ME across sessions. Taken together, the current results indicate that the brain network underlying MI changes with training, further indicating that MI itself is a learned skill. Future work will utilize graph theory (Rubinov, 2010) to quantify changes in network characteristics during MI training, and examine the amount of MI training necessary to effectively facilitate skill acquisition in the absence of physical practice.
Poster 21
Cortical patterns and functional recovery in chronic stroke patients by functional electrical stimulation
1Division of Physical Therapy and Rehabilitation, University of Fukui Hospital, Yoshida-gun, Fukui, Japan
2SecondDepartment of Internal Medicine, Faculty of Medical Sciences, University of Fukui, Yoshida-gun, Fukui, Japan
3Departments of Neurosurgery and the Biomedical Imaging Research Center, Fukui Medical University, Yoshida-gun, Fukui, Japan
4Division of Systems Research, Faculty of Engineering, Yokohama National University, Yokohama, Kanagawa, Japan
5Department of Mechanical Engineering and Intelligent System, The University of Electro-Communications, Chofu, Tokyo, Japan
6Department of Orthopaedics and Rehabilitation Medicine, Faculty of Medical Sciences, University of Fukui,, Yoshida-gun, Fukui, Japan
Poster 22
The Use of Body-Machine Interfaces to Examine Developmental Change in Motor Skill Acquisiton
1Michigan State University, East Lansing, MI, USA
2Northwestern University, Evanston, IL, USA
How does learning in children differ from that in adults? This issue has important implications both from a theoretical motor development standpoint as well as in pediatric rehabilitation. Addressing this question has methodological challenges because in typical motor tasks, any differences in learning strategies between children and adults are often confounded by differences in task familiarity and anthropometric changes.
In order to minimize these confounds, we investigated how children and adults acquire a novel motor skill in a virtual task using a body-machine interface (BoMI). A body-machine interface transforms body movement into the control of an external device. In addition to being a novel task, it allows tasks to be body scaled - thereby ensuring that both children and adults could perform the task using their existing movement abilities.
Participants learned to use their shoulder movements to control a computer cursor in a center-out reaching task to 8 different targets. Shoulder movements were measured using 4 inertial measurement units (IMU) attached to the participants’ upper body and mapped to cursor position. Both children and adults practiced for a total of 160 trials toward 4 targets.
Results showed that initially in practice, children had longer movement times (~50%) compared to adults. This difference in movement time was also associated with a change in the movement strategy. The analysis of the task and null space variance showed that children tended to use greater exploration (i.e. greater task space and null space variance) when compared to adults initially in learning. However, by the end of practice, the performance and exploration for the two groups were similar.
These results suggest that children tend to use greater exploration initially in learning. Future studies will investigate how this difference in learning strategy may be exploited to facilitate motor learning in children.
Poster 23
Dissociating the mechanisms of savings and anterograde interference
1Brain and Mind Institute, University of Western Ontario, Ontario, Canada
2Centre for Sensorimotor Neuroscience, School of Human Movement Studies, The University of Queensland, Queensland, Australia
3Institut de Neurosciences Cognitives et Intégratives d’Aquitaine, CNRS UMR 5287, Université Bordeaux Segalen, Bordeaux, France
4School of Psychology, The University of Western Australia, Western Australia, Australia
Motor learning forms the basis of movement rehabilitation. Better understanding of what contributes to the retention of motor learning is crucial to improving outcomes of movement rehabilitation. In motor adaptation learning, retention has been proposed to occur via (1) reinforcement, where repeating an adapted movement associates the adapted movement with rewarding outcomes; and (2) use-dependent plasticity, where repetition biases subsequent movements to be similar to the repeated movement. Retention is evident in savings, where initial learning improves subsequent learning, and anterograde interference, where initial learning impairs subsequent learning. Savings appears to rely on reinforcement mechanisms (Huang et al., 2011), as deficient reinforcement mechanisms in Parkinson’s disease impairs savings despite intact initial learning (Leow et al., 2012). The mechanisms of anterograde interference remain unclear. We propose that use-dependent plasticity contributes to anterograde interference, but not to savings. This proposal was tested by limiting or extending movement repetition while stimulating the motor cortex with anodal transcranial direct current stimulation, which increases use-dependent plasticity when paired with movement repetition. Participants first adapted reaching movements to a counter-clockwise rotation of visual feedback imposed either abruptly (extended repetition) or gradually (limited repetition) in a first block, during which either sham or 2mA anodal transcranial direct current stimulation (tDCS) was applied over motor cortex. Anterograde interference was then assessed in a second block with a clockwise rotation, and savings was assessed in a third block with a counterclockwise rotation. We found that anodal M1 tDCS elicited more anterograde interference than sham stimulation with extended, but not with limited movement repetition. Conversely, anodal M1 tDCS did not affect savings with either limited or extended repetition of the adapted movement. Crucially, the effect of anodal M1 tDCS on anterograde interference did not require large errors evoked by abrupt perturbations. M1 tDCS still increased anterograde interference even with small errors evoked by a gradual perturbation when target manipulation enforced extended movement repetition. These findings demonstrate that use-dependent plasticity contributes to anterograde interference, but not to savings. This and previous work suggests dissociable mechanisms of savings and anterograde interference. Better understanding of the mechanisms underlying savings and anterograde interference allows for sensible manipulation of these mechanisms to optimize retention of rehabilitation gains in clinical practice.
Poster 24
Reduced Intracortical Inhibition is Associated with Bimanual Common and Dual Goal Tasks
1University of Maryland, School of Medicine, Department of Physical Therapy and Rehabilitation Science, Baltimore, MD, USA
2University of Maryland, School of Medicine, Departments of Neurology and Physical Therapy and Rehabilitation Science, Baltimore, MD, USA
3Veterans Administration, Maryland Exercise and Robotics Center of Excellence (MERCE), Baltimore, MD, USA
4Neuroplasticity and Motor Behavior Laboratory, Moss Rehabilitation Research Institute., Philadelphia, PA, USA
Poster 25
Cortico-cortical coupling during bilateral forces
University of Pittsburgh, Systems Neuroscience Institute, Pittsburgh, USA
Interhemispheric interactions between primary motor cortices play an important role in the control of bilateral forces. However, the mechanisms involved in the generation of bilateral isometric contractions of increasing force levels remain poorly understood. Here, we used coherence analysis between electroencephalographic (EEG) signals from the left and right primary motor cortices (EEG-EEG coherence) and the ipsilateral cortical silent period (iSP; a measure of interhemispheric inhibition) to examine the cortico-cortical coupling during bilateral forces. We tested healthy controls subjected during unilateral and bilateral isometric index finger abduction at 10, 40 and 70% of maximal voluntary contraction (MVC). EEG-EEG was calculated in the alpha (8-13 Hz), low (13-21 Hz) and high (21-30 Hz) beta frequency bands. We demonstrate that mean coherence decreased during increasing levels of force generation in all frequency bands. Whereas, the magnitude of the iSP increased with increasing levels of force. We found a significant inverse correlation between the magnitude of the iSP and EEG-EEG coherence in the high frequency band, particularly at low levels of force, suggesting an association between these measurements. Together, our results indicate cortico-cortical coupling during bilateral forces might involve oscillatory activity in the high beta frequency band.
Poster 26
Effects of Short-Term Cycling Exercise on Functional Measures of Aging Related Changes in Upper Extremity Function
1Atlanta VA Center for Visual and Neurocognitive Rehabilitation, Atlanta, GA, USA
2Emory University, Department of Neurology, Atlanta, GA, USA
3Georgia State University, Department of Physical Therapy, Atlanta, GA, USA
Aerobic fitness has long been associated with better upper extremity function (Spirduso, 1975), yet the neural changes largely responsible for these improvements are not well understood. It has been proposed that alteration of inhibitiory systems in the cortex have strong effect on plasticity in aging and regular aerobic exercise may be prophylactic in preventing aging-related changes in cortical inhibition (McGregor et al., 2013). In the current study, we enrolled sedentary older adults (60 years+) in a short-term (12-week) exercise program to explore if improvements in aerobic capacity alter upper extremity motor control. We evaluated changes in neural activity using functional magnetic resonance imaging and transcranial magnetic stimulation. Participants improved estimates of VO2max by 10-15% on average as a result of the aerobic exercise intervention. Preliminary results show behavioral improvements in both affective, cognitive and motor dexterity measures as a result of the exercise intervention. Pre/Post session TMS results indicate an increase in interhemispheric inhibition, though additional data analysis is required. Comparison of functional MRI datasets in conventional pre/post assessment did not show significant differences in interhemispheric communication. A network analysis approach to address systems change is warranted and may yield additional information from MRI datasets.
Poster 27
Home-based Mirror Therapy for Rehabilitation of Hemiparetic Lower Limb Post-Stroke: A Pilot Study
1Moss Rehabilitation Research Institute, Elkins Park, PA, USA
2Stony Brook University, Stony Brook, NY, USA
Mirror therapy (MT) is a relatively new therapeutic intervention for hemiparesis. MT involves performing movements with the unimpaired limb while watching its mirror reflection superimposed over the (unseen) impaired limb, thus creating a visual illusion of enhanced movement capability of the impaired limb. While a growing body of clinical research indicates that upper limb MT benefits stroke patients to a degree comparable to or better than other therapies, only one study has demonstrated efficacy of lower limb MT in a subacute stroke population1, and potential benefits in a chronic population are unknown. The aim of the study was to determine whether a home-based form of MT is an effective treatment for lower limb hemiparesis in chronic stroke survivors.
This study used a home-based MT regimen with a custom made MT device and a DVD providing instructions. The 30 minute training video directed participants to flex and extend the knee, dorsiflex and plantarflex the ankle, and circumduct the foot. Participants performed training at home two times per day, 5 days per week for four weeks. Seven individuals with chronic CVA and lower extremity paresis were enrolled; five completed all sessions. Outcome measures included lower extremity Fugl-Meyer scores (LE-FM); self-selected walking speed (SSWS) and fastest walking speed (FWS); and an accuracy index (AI) for an ankle tracking task, which was performed on both more affected (MA) and less affected (LA) limbs. This final measure used a custom designed apparatus to quantitatively assess voluntary control of ankle movement. All measures were collected during two baseline periods separated by at least 24 hours, during training (weekly assessments), immediately following training (post-treatment), and at 1-month follow-up.
We found that the AI on the more affected side improved significantly from baseline (pre-training) to post-training (p<0.05). AI on the less affected side and scores on the LE-FM trended towards, but did not reach, significance (p = 0.07). There were no significant changes in preferred walking speed (p=0.67) or fastest walking speed (p=0.30) from baseline to post-training; however, trends for improvement were noted at the 1-month follow-up.
Overall, we have identified some positive effects of MT, particularly on volitional control of the more affected ankle. MT may also contribute to improved walking speed over time. This pilot study is a first step towards establishing a home-based therapy for gait rehabilitation for stroke survivors, and may be especially beneficial for those who have limited access to other forms of rehabilitation.
This study was supported by funds from the PA Department of Health
1. Sutbeyaz, S., Yavuzer, G., Sezer, N. & Koseoglu, B. F. Mirror therapy enhances lower-extremity motor recovery and motor functioning after stroke: a randomized controlled trial. Arch Phys Med Rehabil 88, 555-559, doi:10.1016/j.apmr.2007.02.034 (2007).
Poster 28
Diffusion tensor tractography study of sensorimotor pathways in unilateral hand amputees
University of Missouri, Columbia, Missouri, USA
Poster 29
Classification of Normal and Abnormal Gait in Young Children Based on Foot Pressure Data: Towards Remote Monitoring of Medical Interventions
West Virginia University, Morgantown, WV, USA
Poster 30
Severe spondylosis presenting with anterior cervical osteophytes induced respiratory arrest and incomplete quadriplegia secondary to cervical spondylotic myelopathy: A rare case report
1Touro College of Osteopathic Medicine, New York, New York, USA
2Orange Regional Medical Center, Middletown, New York, USA
Patient interview and records revealed six months of worsening lower extremity weakness, difficulty with gait and balance, three months of urinary incontinence, one or two months of hand weakness, and three or four weeks of dysphasia with both liquids and solids, hoarseness, and coughing with swallowing. Prior to the development of his symptoms, he ambulated with a walker.
On physical exam, the patient had bilateral thenar atrophy, right hand asterixis and claw-like stiffness, and an abnormal gait with spasticity. Motor examination revealed 2/5 finger extension bilaterally, 3+ biceps and triceps reflexes bilaterally, positive right ankle clonus, and bilateral Hoffman’s sign. On sensory examination, touch, pinprick, temperature, and proprioception over toes, feet, and fingers were decreased. Fine motor movements were decreased bilaterally.
Neck CT myelogram demonstrated severe spinal stenosis at C5-C6 and large anterior cervical osteophytes from C3 - C6, compressing the esophagus and the hypopharyngeal airway at the level of the epiglottis. In addition, the patient had multilevel lumbar spondylosis with disc bulging and resultant multilevel central canal stenosis. Video swallows further suggested severe pharyngeal dysphagia secondary to large anterior cervical osteophytes.
The patient underwent anterior cervical decompression, which included removal of osteophytes from C3-C6 and discectomy and fusion at C5-C6.
Poster 31
Skill acquisition with the non-dominant hand and associated changes in functional connectivity between sensorimotor hand representations
University of Missouri, Columbia, MO, USA
Unilateral impairment of the dominant right hand can arise from many causes, including diseases or injuries that affect the central or peripheral nervous systems. For some patients, this impairment may be irreversible, which forces them to compensate with the intact non-dominant left hand to perform movements that require high levels of precision. Therefore, there may be broad clinical significance to understanding the behavioral and neural changes involved in learning and retaining precision skills with the left non-dominant hand. We addressed this issue through a novel precision drawing task and resting-state functional connectivity MRI (fcMRI).
We trained 19 healthy right-handed adults (age 27 ± 8, 13 female) to perform a precision drawing task with their left hand. In this task, participants used a pen to draw continuous lines (45-180 mm) within provided boundaries (3, 4, or 5 mm tolerance); primary dependent measures included endpoint speed and smoothness. The training regime involved 10 days of left hand training, with a fixed number of training trials on each day (15-25 minutes). We tested participants’ right hand performance before and after training. Participants returned for follow-up testing sessions 1 week, 1 month, and 6 months after the end of training. In addition, we identified training-correlated changes in functional connectivity with seed regions in bilateral hand sensorimotor cortex, via fcMRI before and after training.
Seventeen participants (89%) showed significantly improved left hand performance across training. The performance difference between hands decreased significantly in the smoothness domain. Despite discontinuation of left hand training, performance improvements remained stable over time: six months after training, 12/14 (86%) participants still showed significantly improved left hand performances.
We found a learning-correlated decrease in functional connectivity between right (trained) sensorimotor hand cortex and bilateral cortical areas including sensorimotor, premotor, and superior parietal cortex. These findings suggest that successful learning of precision skill with the left non-dominant hand may involve increased independence of the non-dominant from the dominant sensorimotor cortex when at rest. We speculate that these changes may be integral to compensation with the left hand, and thus support improved patient ability to perform activities of daily life after impairment of the right hand.
Supported by NIH/NINDS NS053962 to S.H.F.
Poster 32
Predicting variability of distal muscle recruitment curves in stroke using diffusion tensor imaging (DTI)
1Cleveland Clinic Foundation, Cleveland, OH, USA
2Kent State University, Kent, OH, USA
3Sao Paulo University, Sau Paulo, Brazil
Viability of corticospinal tracts (CST) has long been hypothesized to predict functional outcomes of upper limbs following lesions affecting the central nervous system, such as stroke or spinal cord injury. One highly utilized method for understanding CST viability, as a means of predicting and tracking longitudinal patient functional recovery, has been use of the non-invasive neurophysiologic technique transcranial magnetic stimulation (TMS). Specifically, within TMS, electromagnetic induction to the scalp overlying motor cortices (M1) results in an elicited contraction of the contralateral target muscle, termed motor evoked potentials (MEPs). In addition, MEP recruitment curves (RCs) are also determined by assessing MEP amplitudes at incremental stimulus intensities until a MEP plateau is reached. MEP amplitude and the area under the recruitment curve are believed to represent the gain or viability of descending CST to the target muscle and are utilized to track patient recovery. Unfortunately, however, within any patient population, MEPs are highly variable compromising measurement of RC at incremental intensities. Here, in a sample population of stroke patients we investigated whether the extreme variability of RCs can be ascribed to poor white matter integrity of in the CST, as measured by diffusion tensor imaging (DTI). As a metric of overall tract integrity, fractional anisotropy (FA) was compared between the lesioned and non-lesioned hemispheres for tracts projecting between the posterior limb of internal capsule (PLIC) and major cortices including: M1, premotor cortex (PMC) and supplementary motor area (SMA). FA was evaluated at 3 levels: (1) single-slice at the PLIC, (2) weighted average along the length of the tract and (3) segment of tract lying within the stroke/degenerated region. Notably, we found that variability in RCs was strongly correlated with segmental variations of FA for tracts converging from M1 (R=-0.766, p<0.027), PMC (R=-0.647, p<0.083) and SMA (R=-0.673, p<0.067). In addition, we found that RC variability was significantly predicted by segmental variation of FA along tracts emerging from M1 (B=-0.766, p<0.027). Collectively, our results suggest that of the three investigated types of DTI analysis, segmental FA values may provide the strongest clinical sensitivity in predicting variations in MEP generation in distal muscles. In addition, by reconstructing tracts to three major cortices, future work can now begin to predict recovery potential based on the viability of individual CSTs. Finally, by isolating potential sources for variability in of MEP measurement, we hope to further improve the diagnostic ability of TMS.
Poster 33
Increasing Repertoire of Finger Movements to Improve Hand Dexterity in Stroke
Michigan State University, East Lansing MI, USA
Stroke often results in impaired hand function that significantly impacts the ability to perform activities of daily living. Hence there is a need to discover new techniques to improve hand dexterity. Here, we examine the use of a body-machine interface to induce reorganization and increase the repertoire of finger movements in stroke.
8 individuals with chronic stroke (average 6 years post-stroke) were recruited for the study. Participants wore a Cyberglove and the signals corresponding to flexion/extension of the MCP joint of the four digits were mapped linearly on to the position of a cursor on the screen. The task was to move the cursor back and forth between two targets positioned at the left and right edges of the screen.
In order to facilitate exploration of different finger movements, we gradually changed the weightings in the map between the finger movements and the movement of the cursor. For example, the weights of the index, middle, ring, and little fingers (I,M,R,L) were initially set to (1,1,1,1) - which requires a power-grasp pattern. This weights would be gradually changed during the course of the trial to (1,1,-1-1) - which requires individuation between the IM and RL fingers. Each trial consisted of between 50 and 100 targets, and participants practiced the task for 4 sessions spread over two weeks. We measured the repertoire of finger movements generated in these trials using a principal component (PCA) analysis.
Results showed that participants were able to reorganize their finger coordination patterns to adapt to the changing weights and reach the targets. Moreover, in 6 out of 8 subjects, the variance accounted for by the first PC (which indicated the dominant coordination pattern) decreased with practice (on average from 82 to 77%), indicating that participants were able to expand their repertoire of finger movements.
These results suggest that the use of a body-machine interface resulted in increasing the repetoire of finger movements in stroke survivors. This technique provides insight into designing a rehabilitation paradigm targeted at breaking maladaptive synergies.
Poster 34
Efficacy and feasibility of functional upper extremity task-specific training for older adults
1Utah State University, Logan, UT, USA
2University of Utah, Salt Lake City, UT, USA
Poster 35
Learning, retention, and inter-limb transfer of a novel gait training paradigm
1University of Michigan, Ann Arbor, Michigan, USA
2Michigan State University, East Lansing, Michigan, USA
Previous research indicates that motor learning is a key component of recovery after neurological injuries such as stroke or spinal cord injury. We recently developed a novel functional motor learning paradigm for gait rehabilitation; however, little is known on the retention and inter-limb transfer effects of our paradigm. Therefore, the purpose of this study was to evaluate the consolidation and inter-limb transfer effects of our leg motor skill-learning task in a group of neurologically intact adults. A secondary purpose was to evaluate whether transfer effects are side-specific as upper-extremity literature indicates that spatial accuracy tasks are learned better when practiced initially with the non-dominant hand. Twenty-five young adults (13 right-group and 12 left-group) were tested on two consecutive days. Participants performed a target-tracking task that necessitated hip and knee flexion modifications while walking on a treadmill. On Day 1, the right-group performed testing on the right leg. On Day 2, the subjects were tested for right leg retention and left leg transfer effects using the same paradigm. The left group performed the same sequence beginning on the left leg. Changes in tracking error were computed to study the learning, retention, and transfer effects. The results indicated that repeated practice of the leg motor learning task resulted in significant reduction in target-tracking error in both the groups (P<0.05). Subjects in both groups retained their performance similar to those observed at the end of training on Day 1 (Day 1, Block 10 Tracking Error = 12.7 ± 0.5%; Day 2, Block 1 Tracking Error = 13.1 ± 0.9%, P>0.05). Similarly, both groups displayed transfer effects between legs (P<0.05) and the inter-limb transfer appeared to be symmetrical between legs (P>0.05). To address whether the transfer effect was mediated due to consolidation or due to practice of target-tracking prior to testing of transfer effects, a control study was conducted in which twenty young adults (10 right-group, 10 left-group) were tested on a single day on both their legs using the same paradigm. We also evaluated whether there were inherent differences in motor skill learning between legs. The results indicated no differences between legs (P > 0.05) and demonstrated minimal inter-limb transfer in the single-day protocol. Evaluation of no-vision trials indicated that there was a clear transfer effect on the 2-Day protocol, whereas no transfer effect on the 1-Day protocol. These results indicate that repeated practice results in better inter-limb transfer when the learning of a task on the training leg has consolidated. The inter-limb transfer effects displayed in our target-tracking paradigm provide an opportunity to facilitate recovery of the impaired limb via training of the less impaired limb. The results have meaningful implications for gait rehabilitation in individuals with stroke or other neurological disorders.
Poster 36
Use of inertial sensors for determining rate of kicking in infants
University of Southern California, Los Angeles, Ca, USA
Poster 37
Capturing Upper Extremity Function with Acceleration Variability Metrics
Washington University School of Medicine, St. Louis, MO, USA
Poster 38
Amputation-related changes in inter-hemispheric interactions are reversible through transplantation of the human hand
1University of Missouri, Columbia, Missouri, USA
2Brain Imaging Center, Columbia, Missouri, USA
3Christine M. Kleinert Institute, Louisville, Kentucky, USA
Use of the intact hand by unilateral amputees is associated with increased activity in both the contralateral (intact) and ipsilateral (former) sensorimotor hand territories. Ipsilateral responses are believed to reflect reductions in normal levels of inter-hemispheric inhibition maintained by activity-dependent, transcallosal, GABAergic pathways. Former amputees who have undergone hand transplantation provide a unique opportunity to address whether these activity-dependent reorganizational changes can be reversed. If so, then following regeneration of peripheral nerves, movements of the intact (non-transplanted) hand should exhibit the same pattern as healthy adults, primarily activating contra- but not ipsilateral sensorimotor cortex. Here, we used functional MRI to map cortical sensorimotor representations in patient DR, a right-hand dominant male who suffered traumatic amputation of his left hand proximal to the wrist in 1998 (age 23) and underwent hand transplantation 13 years later (age 36). Patient DR completed the same protocol 15 and 26 months post-transplant. At 15 months post-transplant, DR showed a pattern of activation similar to amputees: specifically, he activated both contra- and ipsilateral sensorimotor cortices when moving his intact hand. However, when tested at 26 months post-transplant, patient DR showed a pattern of activation similar to controls: specifically, he activated only contralateral (but not ipsilateral) sensorimotor cortex when moving his intact hand. Between 15 and 26 months post-transplant, DR exhibited substantial improvements in hand functions that reflect regeneration of peripheral sensory and motor nerves. These results are consistent with the hypothesis that amputation-related changes in the normal inhibitory balance between sensorimotor cortices can be reversed when afferent and efferent activity between hand and brain is restored. The fact that this was observed in a fully mature brain has potentially important implications for understanding the role of experience in recovery from injuries to the central and/or peripheral nervous systems.
Poster 39
Restoration of cortical blood flow precedes spontaneous forelimb recovery after cortical infarcts in mice
1The University of Texas, Austin, TX, USA
2Duke University, Durham, NC, USA
There is typically some degree of spontaneous recovery over time after stroke and neural plasticity in the peri-infarct tissue has been implicated in this recovery. The temporal relationship between restoration of blood flow to peri-infarct tissue and spontaneous recovery has not yet been investigated. We studied this in a mouse model of post-stroke upper extremity impairments by repeated within-animal measures of blood flow changes in peri-infarct cortex, as assayed with multi-exposure speckle imaging (MESI), and forelimb functional changes, as assayed with the pasta matrix reaching task (PMRT). Fourteen C57/BL6 mice were first shaped to determine their preferred-for-reaching forelimb on the PMRT. Cranial windows were then installed over the forelimb area of the contralateral motor cortex. After 3 weeks of recovery from cranial window surgery, the mice then received 3 baseline imaging sessions, followed by 14 days of training on the PMRT to establish pre-injury skill. Mice then received photothrombotic cortical lesions of the forelimb representation of the trained motor cortex or sham procedures (n’s = 7). Mice received forelimb probes using the PMRT in tandem with the imaging of cortical blood flow at Days 3, 5, 10, and 20 post-infarct. Infarcts significantly decreased performance on the PMRT and reduced cortical blood flow compared to both baseline levels and the sham group. Additionally, the re-establishment of cortical blood flow proximal to the infarct core preceded the recovery of motor performance. The temporal patterns of results are consistent with the possibility that blood flow recovery enables the adaptive plasticity required for motor recovery.
Poster 40
Forearm muscle activation in children with cerebral palsy and typically developing children during massed practice through adapted video game play
1University of Massachusetts Lowell, Lowell, MA, USA
2Brown University, Providence, RI, USA
3Hasbro Children’s Hospital, Providence, RI, USA
Poster 41
Effect of burst stimulation by high frequency biphasic square-wave pulse on cortical perfusion after stroke: A pilot study
1Second Department of Internal Medicine, Faculty of Medical Sciences, University of Fukui, Fukui, Japan
2Gait Analysis Laboratory, Division of Physical Therapy and Rehabilitation, University of Fukui Hospital, Fukui, Japan
3Division of Systems Research, Faculty of Engineering, Yokohama National University, kanagawa, Japan
4Department of Mechanical Engineering and Intelligent System, The University of Electro-Communications, Tokyo, Japan
5Biomedical Imaging Research Center, University of Fukui, Fukui, Japan
6Department of Orthopedics and Rehabilitation Medicine, Faculty of Medical Sciences, University of Fukui, Fukui, Japan
Poster 42
Paradoxical motor recovery from a first stroke by re-opening a sensitive period with a second stroke
1Johns Hopkins, Maryland, USA
2UCLA, California, USA
3Duke University, North Carolina, USA
Poster 43
Clinical Implementation of an iPod-based Gait Assessment System for In-patient Stroke and Brain Injury Rehabilitation
1School of Rehabilitation Medicine, Capital Medical University, Beijing, China
2Department of Rehabilitation Evaluation, Beijing Charity Hospital, China Rehabilitation Research Center, Beijing, China
3Department of Clinical Neurosciences, Faculty of Medicine, University of Calgary, Alberta, Canada
In-patient programs for stroke and brain injury rehabilitation in a tertiary hospital can consume significant clinical resources. Here we conducted a pilot study on an iPod based gait assessment and training system in the largest rehab center in metropolitan Beijing, China. Ambulosono System (AID Inc., Calgary, Canada) is a mobile technological platform that employs a suite of iPhone Apps to enable motion sensing, automated auditory feedback instructions, and cloud-based data logging and computing. To test its clinical applicability, we recruited consecutively 40 young and adult patients with acute stroke (F=24; M=4; age=37-62yrs) or other brain injury patients (F=7; M=5; age=13-71yr). These patients, who are considered a typical clinical cohort seen at our center, were followed up over a 5 month period. Initial baseline clinical assessment included: Fugl-Meyer Motor Assessment for Lower Extremity, Berg Balance Scale, Functional Ambulation Category, isokinetic knee muscle testing (Biodex System 4 Pro). Ambulosono-based interventions consisted of self-paced over-ground walking and task specific training (e.g. step length triggered auditory and music-feedback, obstacle negotiation, backward walking, and stepping in place, under both mono and dual task conditions). Testing and training data captured by the iPod Touch was transmitted to a cloud server for longitudinal analysis. Our results indicate: 1) gait parameters acquired through Ambulosono system are largely comparable to those reported in the literature. The average walking speed of stroke patients was around 0.51m/s. During training with step length triggered music-feedback, the walking speed increased to 0.89m/s. The average step length and time for music training were 0.55m (CV=0.28) and 0.65s (CV=0.23), respectively. Step length asymmetry between the intact and impaired limb ranged from 5 to 12%. 2) Task specific gait training using Ambulsono system also resulted in significant improvement in patients’ walking speed and distance. 3) Post-Ambulosono training also led to significant improvements in baseline clinical assessment scores such as Berg Balance Scale (pre-Ambulosono training=47.8; post-Ambulosono training=51.1; p = 0.001).
Poster 44
Distinctive Impairment Profiles between Parkinsonian and Hemiplegic Gait Based on 6 Minute Walking and Dual Task Working Memory Tests
1School of Rehabilitation Medicine, Capital Medical University, Beijing, China
2Department of Rehabilitation Evaluation, Beijing Charity Hospital, China Rehabilitation Research Center, Beijing, China
3Department of Clinical Neurosciences, Hotchkiss Brain Institute. Faculty of Medicine, University of Calgary, Calgary, Alberta, Canada
Poster 45
Improved sensation in a replanted or transplanted hand
1University of Missouri, Columbia, MO, USA
2Christine M. Kleinert Institute, Louisville, KY, USA
Former amputees that have had their injured hands replanted (heterotopic hand replants), or received transplantations of a donor hand (allogeneic hand transplants) provide a unique opportunity to evaluate the effects of deafferentation on the central nervous system can be reversed. Following sensory nerve transection and repair, peripheral nerve regeneration is estimated to proceed at a rate of up to 2mm per day. However, patients that undergo with surgical nerve repairs of the arm or hand show persistent difficulties in localization of touch without vision. This may reflect chronic disorganization of finger maps within the primary sensory (S1) cortex, as suggested by studies in nonhuman primates. We tested the hypothesis that central adaptations associated with extended experience can mitigate these functional limitations in right-handed heterotopic replant (N = 4) and allogeneic hand transplant recipients (N = 3).
We adapted the locognosia technique (Nordenboos, 1972) to measure the ability to localize light touch (100mN) on ventral surface of the hand and digits. This method allowed participants to see the position of their hand (eliminating proprioceptive demands), but not the location of the stimulus. On average, healthy adults localized touch with a very high level of precision, and exhibited no differences between sides (Right: Mean ± SD=4.00 ± 3.76, Left: 3.70 ± 3.40 ). Patients showed substantial variability (Affected Hand: 26.03 ± 24.75, Unaffected Hand: 4.77 ± 5.16), and a positive correlation between localization accuracy and time since hand replantation or transplantation. Two complete hand transplant recipients (8 and 10 years post-surgery) one mid-palm replant recipient (3 years post-surgery), and one full hand replant recipient (1.5 years post-surgery) exhibited the ability to localize stimuli on average within 95% confidence intervals of the control group on their affected hands. Time since the transplant/replant was correlated with performance (r=-.63).
Our findings suggest that hand transplant and replant recipients may recover a very high level of accuracy in touch localization. This ability may continue to improve for years following peripheral nerve repair and regeneration, and is perhaps attributable to central adaptations.
Poster 46
Spectroscopic evidence for lower neuronal metabolism and time-dependent increases in inhibition within the former sensorimotor hand territory of chronic unilateral amputees
University of Missouri, Columbia, MO, USA
R
Poster 47
Facilitating supplementary motor area using near-infrared spectroscopy mediated neurofeedback improves postural stability but not hand dexterity
1Neurorehabilitation Research Institute, Morinomiya Hospital, Osaka, Japan
2Department of Neurology, Osaka University Graduate School of Medicine, Osaka, Japan
NF session consisted of 16 repetitions of 5second - task and 8-16 second rest periods. Subjects were asked to control their SMA activation according to the feedback signals without specific strategic instruction. Each subject received the REAL session in which their own SMA activation was fed back and the SHAM session in which the SMA activation of other subjects was fed back, with an interval of more than one week.
Before and after each NF session, postural stability and hand dexterity were assessed using cumulative length of center of pressure (COP) displacement during standing and 9-hole PEG test score (9HPT) with non-dominant hand respectively.
NF effect on the SMA activation was assessed using comparison between first 6 blocks and last 10 blocks in each NF sessions. Behavioral changes were analyzed using repeated measures ANOVA with p<0.05 as significant.
Poster 48
Reliability of Ipsilateral Silent Period to Measure Interhemispheric Inhibition: Preliminary Results in Non-disabled Young Adults
Division of Biokinesiology and Physical Therapy, University of Southern California, Los Angeles, California, USA
The ipsilateral silent period (ISP) is an index of transcallosal inhibition. ISP is a brief reduction in EMG activity following transcranial magnetic stimulation (TMS) applied to the primary motor cortex (M1) during voluntary ipsilateral muscle activation. Measures of ISP, such as duration
or amplitude, are thought to reflect the balance in interhemispheric inhibition (IHI)-the extent to which homologous M1 areas mutually inhibit the contralateral side. An imbalance in IHI may be an important mechanism underlying impaired motor function after stroke, making ISP a useful and often cited measure. In stroke, ISP also allows for greater participant inclusion than other measures of IHI because it is assessed via stimulation of the non-affected M1. Our long-term goal is to measure ISP during volitional activation of the paretic hand in individuals with stroke, yet consistent methods for determining ISP for first dorsal interosseous (FDI) have not been established. Current studies have utilized a maximal contraction of the affected and/or ipsilateral hand, limiting the number of trials that could be completed before fatigue. While other studies have utilized submaximal contractions, it is not yet known whether the amount of force would impact the consistency of the ISP measurement. Therefore, the purpose of the current study is to investigate ISP under different percentages of maximal volitional isometric contraction (MVIC) to determine the optimal percentage of force that produces the most consistent ISP values for FDI among non-disabled young adults.
Four subjects were recruited to press against a force transducer isometrically using their dominant side’s index finger abduction. MVIC was measured initially and used to calculate three force levels: 30%, 50% and 100% of MVIC. The subjects were asked to hold the predetermined force while a TMS pulse was applied over the non-dominant hemisphere. ISP was observed across force levels in all 4 subjects with 30% demonstrating the most robust observation. The ISP durations were 28.7±2.7, 26.4±8.6 and 22.9±11.7ms; while the ISP onset latency was 38.3±1.7, 36.4±2.8 and 38.2±3.7ms under 30%, 50% and 100% of MVIC, respectively. In summary, this is a feasible method to identify ISP for FDI in healthy young adults. ISP duration was longer and less variable (coefficient of variation (CV) = 4.3% vs.7.7% and 9.7%) under 30% MVIC across the 4 subjects. As ISP is a measure of precise interhemispheric interactions, we hypothesize that IHI may be a critical process for adjusting refined muscle contraction rather than for producing maximum force. Future studies will utilize the ISP measurement under submaximal muscle contraction to assess IHI following stroke.
Poster 49
Transcranial direct stimulation: modulating functional connectivity across pain networks
Cleveland Clinic Foundation, Cleveland, Ohio, USA
Poster 50
RhoA Expression in Lamprey Brain Neurons After Spinal Cord Injury
1Temple University School of Medicine, Philadelphia, PA, USA
2Shriners Hospitals for Children, Philadelphia, PA, USA
Disability following spinal cord injury (SCI) is due to failure of axonal regeneration. Axon growth may be inhibited by several types of inhibitory molecules in the CNS, including the chondroitin sulfate proteoglycans (CSPG), myelin-associated glycoprotein (MAG), Nogo and oligodendrocyte-myelin glycoprotein (OMgp). In the lamprey CNS, there are 18 pairs of individually identified reticulospinal neurons with heterogeneous axon regenerative abilities. Of these, the “bad-regenerating” neurons often experience a very delayed form of apoptosis. We have reported that the putative CSPG receptors, PTPσ and LAR, are expressed selectively in the bad-regenerating neurons both in normal lampreys and after SCI, when those same neurons were also undergoing apoptosis, as indicated by labelling with fluorescently labelled inhibitors of caspase activation (FLICA). It is thought that the intracellular signaling pathways activated by those inhibitory molecules, including PTPσ and LAR, converge on RhoA, whose activation has been implicated in both apoptosis and axon growth inhibition. To study the role of RhoA in SCI-induced retrograde neuronal death, we cloned lamprey RhoA, which was highly homologous to mammalian RhoA, and investigated its distribution in lamprey CNS by wholemount in situ hybridization (ISH) before and after SCI. RhoA expression in neurons increased during normal development in lamprey CNS. RhoA was continuously expressed in bad-regenerating neurons and in microglia/macrophages on cord surface after SCI, but was decreased greatly in most other reticulospinal neurons after SCI. Moreover, RhoA mRNA expression was correlated with caspase activation in brain at 2 weeks after SCI. The inverse correlation of RhoA expression with the intrinsic regenerative ability and survival post-axotomy is consistent with a role for RhoA signaling in triggering apoptosis and restricting axon regeneration after SCI.
Poster 51
Genotyping of brain-derived neurotrophic factor predicts response to single-pulse transcranial magnetic stimulation at rest
1Laboratory for Cognition and Neural Stimulation, University of Pennsylvania, Philadelphia, PA, USA
2Department of Neurology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA
3Department of Psychiatry, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA
Polymorphisms in the gene coding for brain-derived neurotrophic factor (BDNF) affect the ability of noninvasive brain stimulation to induce transient changes in corticospinal excitability. Measurements of the motor response to transcranial magnetic stimulation (TMS)-motor evoked potentials (MEP)-have long been used as a means of assessing excitability and plasticity in humans. Evidence suggests that the extent of aftereffects of repetitive TMS, theta burst and paired associative stimulation differ significantly based on BDNF polymorphisms. Specifically, carriers of 66Met allele of the BDNF gene are less susceptible to stimulation-induced modulations compared to those carrying Val66Val allele. While in these studies, susceptibility was defined by marked changes in MEPs after intense repetitive/rapid TMS paradigms or after motor training, it is conceivable that inherent differences in cortical excitability, influenced by BDNF genotype, could also affect evoked responses to simpler and less intense TMS paradigms at rest.
In a retrospective study, we examined a relationship between BDNF genotypes and response to single-pulse, aperiodic TMS in 21 healthy individuals (13 females; age: 25.4±7.98 years). Two single-pulse paradigms were explored: in subgroup 1 (n= 11), MEPs were evoked at 110% of subjects’ resting motor threshold (rMT; ~50µV in 5/10 consecutive trials), while in subgroup 2 (n=10), intensities were adjusted so as to evoke MEPs of 1mV amplitude. In both subgroups, 30-35 MEPs were obtained from the right first dorsal interosseous as the subjects rested their hands. Functional variant Val66Met (rs6265) in the BDNF gene was genotyped using methods described in several prior studies. Genotyping was done by a technician, blinded to the identity of DNA samples, using Taqman assays on an ABI7900HT apparatus.
Fifteen subjects were homozygous for the val allele (Val/Val), 6 were heterozygous (Val/Met), and none were homozygous for the met allele (Met/Met). Controlling for age, sex and rMT, ANCOVA revealed that mean MEP amplitudes differed between BDNF genotypes (F1,16 = 14.9, p=0.0014) such that MEPs were significantly greater in subjects with the Val/Val genotype (1.09±0.3mV) compared to Val/Met group (0.53±0.24mV). In subgroup 1, 7 subjects carrying Val/Val and 4 carrying Val/Met allele were identified. Controlling for rMT, a main effect of genotype was found (F1,8 = 5.38, p=0.049), whereby mean MEPs were greater in subjects carrying Val/Val compared to Val/Met. Although the covariate rMT was not significant, rMT tended to be greater in subjects with Val/Met (66.7 vs. 49.5%). In subgroup 2, we found a similar pattern of results. Together, these findings suggest that although subjects were stimulated at individually-derived and fixed intensities, motor excitability was greatly diminished in Val/Met group.
These preliminary findings suggest that single-pulse TMS can be employed as an index of factors that affect neuroplasticity-an important variable in recovery after a brain injury including stroke-similar to other more complicated TMS paradigms.
Poster 52
Capabilities of the neurorehabilitation are under evaluated. A world wide stigma
Ain - Shams Universty, Cairo, Egypt
Case no.2; 21 years old female with history of TBI affecting higher brain functions. Cognitive assessment & training, PT & life style adjustment was initiated.
Case no.3; 50 years old female with uncontrolled hypertension & refractory headache. Non pharmacological management for muscle spasm & pain(muscle stretching, TENS, Ultrasonic stimulation & Infrared), deep friction massage as a relaxation technique, local nerve block together with patient education were initiated.
Case no.2;was assessed initially by ADDENBROOKE’S COGNITIVE scale, baseline score was 67 after a memory rehab exercises including physical, mental, stress control exercises & Mediterranean diet. Later on the patient was reassessed by the baseline scale & scored 72(5 degrees improvement). The mother confirmed this as the patient was back again to University.
Case no.3; was controlled regarding headache & blood pressure with less side effects & drug-drug interactions. A daily total reduction in the no. of bills received for both illnesses.
Poster 53
Longitudinal imaging of thalamocortical projections after stroke
University of Victoria, Victoria, BC, Canada
The large majority of stroke survivors must cope with chronic disability often affecting the upper limbs. Improved use of the stroke-affected limb is accompanied by neuroplasticity in intact tissues surrounding the stroke and modulating this plasticity should promote further gains in recovery. Sub-cortical brain areas are also affected by cortical stroke, and may contribute to recovery of function. One unknown issue in stroke research involves the role of the thalamus, the brain’s relay center for sensory
information en route to the cortex, in recovery from stroke in the forelimb area of the somatosensory cortex (FLS1). Thus, the aim of the current study was to elucidate the role of thalamocortical projections in recovery of function after stroke. We hypothesized that peri-infarct thalamocortical axons are damaged by stroke, but undergo structural plasticity during recovery. Adult C57BL/6 mice underwent a surgical procedure to inject a green fluorescent protein tagged adeno-associated virus (AAV-GFP) into the ventroposteriolateral nucleus of the thalamus, which sends projections to FLS1. Immediately following virus injection, an imaging window was implanted over forelimb and hindlimb S1. Axon terminals and cerebral vasculature were imaged in vivo using two-photon microscopy before and at various times after stroke to assess acute and long-term changes in vascular and neuronal structure and function. Peri-infarct blood vessels dilate at acute post-stroke periods, but quickly return to baseline diameter in the post-acute period. Preliminary data indicate that, surprisingly, a subset of thalamocortical axons survived within the infarct core, but suffered retraction and branch and bouton loss. Axonal boutons in both the core and peri-infarct cortex underwent high rates of turnover in the first week after stroke, with later stabilization. Overall, peri-infarct thalamocortical axons seem relatively resilient to the effects of ischemia, maintaining most branches and boutons, which may provide a scaffold for functional recovery after stroke.
Poster 54
Transcutaneous spinal cord stimulation to modulate spinal reflex excitability motor output after SCI
1Shepherd Center, Atlanta, GA, USA
2Emory University, Atlanta, GA, USA
3Atlanta VA Medical Center, Atlanta, GA, USA
Spinal cord stimulation (SCS) has been shown to both modulate and generate locomotor like electromyographic activity as well as to alter spasticity in individuals with spinal cord injury (SCI). Recent evidence has shown that transcutaneous (t)SCS can generate reflex responses similar to those elicited by epidural SCS but with reduced cost and invasiveness. It remains unclear however if there are certain stimulation frequencies at which tonic tSCS is most effective at reducing spinal reflex excitability and/or spasticity and whether the addition of stepping related afferent feedback further augments motor output. Therefore the goal of this research was to examine the effects of tSCS frequency on posterior root motor reflexes (PRMRs) and electromyographic muscle activation patterns recorded from healthy non-injured individuals and individuals with incomplete spinal cord injury during stepping in a robotic gait orthosis. We hypothesized that as frequency increases from 10-50Hz, excitability will first rise and then fall as some previous evidence has suggested. To achieve this, tSCS was delivered through stimulating electrodes placed over the T11/T12 intervertebral space (source) and reference electrodes placed on the abdomen (sink) and was applied at sub-motor threshold levels in non-injured and incomplete SCI subjects for up to 30 minutes. The results from the four healthy non-injured subjects, indicate that resting-state excitability is significantly modulated by sub-motor threshold tSCS though the effects of individual frequencies varied. In the individuals with incomplete SCI, tSCS at higher frequencies (50 Hz) demonstrated the ability to depress reflex excitability both during tonic stimulation and up to 5 minutes after the tonic stimulation was removed. Furthermore, during robotic assisted stepping in individuals with SCI, tSCS brought significant reduction ankle clonus during the stance phase. These results and previous evidence support the idea that transcutaneous spinal cord stimulation, like its epidural predecessor at or above 50 Hz tends to decrease responsiveness to afferent inputs and that future studies should focus on training with tSCS.
Poster 55
Impaired Modulation of Corticospinal Drive before Movement Onset after Spinal Cord injury
University of Pittsburgh, Department of Physical Medicine and Rehabilitation, Center for the Neural Basis of Cognition, Systems Neuroscience Institute, Pittsburgh, USA
Corticospinal drive is modulated in a task-dependent manner before movement onset in uninjured controls, which is thought to contribute to features of the upcoming movement. The extent to which corticospinal drive is modulated before movement initiation in individuals with damage to the corticospinal tract due to spinal cord injury (SCI) remains unknown. We tested patients with chronic (≥1 year) incomplete cervical (from C4 to C7) SCI and uninjured age-matched control during a simple index finger reaction time (RT) task. Using single pulse transcranial magnetic stimulation we examined motor evoked potentials (MEPs) in the first dorsal interosseous (FDI) muscle ~120 ms before the onset of ballistic isometric index finger abductions instructed by a visual stimulus presented on a computer screen. We found that index finger RT was shorter in controls (299.1 ± 25 ms) compared with patients (361.1 ± 59 ms). We also found a speed-dependent modulation on MEP size in the FDI muscle in control subjects. FDI MEPs were larger in size when the movement speed increased at similar stimulation periods. In contrast, we did not observe the same modulation in SCI patients. Specifically, in patients FDI MEPs size did not increase progressively with shortening in movement onset. Thus, our results demonstrated an impaired ability to modulate corticospinal output prior to movement initiation after SCI. We propose that understanding these deficits might contribute to facilitate aspects of upcoming movements exerted by a partially paralyzed hand.
Poster 56
Cardiovascular responses to cutaneous nociceptive input after cervical spinal cord injury: role of pain afferent types and their plasticity
1Departments of Neurology and Physiology, Emory University, Atlanta, GA, USA
2Spinal Cord Injury Clinic, Veterans Administration Medical Center, Atlanta, GA, USA
In normal, anesthetized rats, stimulation of segmental (T6 - L1) dorsal cutaneous nerves (DCNs) generates different cardiovascular responses depending on which pain afferent types are activated. Stimulation of A delta fibers alone (0.5 mA) generates increases in heart rate (HR) but relatively limited decreases in blood pressure (BP), with this effect on BP being greater at higher stimulation frequencies (10 Hz vs 1, 2 or 5 Hz). Stimulation of both A delta and C fibers together (5 mA) generates the same increase in HR but now a greater drop in BP, especially at higher stimulation frequencies (2, 5, & 10 Hz). The temporal relationships between BP and HR changes show an initial drop in BP followed by an increase in HR followed by a recovery in BP and then a recovery in HR.
Following C7 crush spinal cord injury (SCI), we have found three cardiovascular responses based on BP, a depressor (normal) response where DCN stimulation at A delta and C fiber strength (5 mA) almost always generates a drop in BP, a pressor (autonomic dysreflexia) response where DCN stimulation almost always generates an increase in BP, and a dysautonomia response where there are mixed depressor and pressor responses to DCN stimulation at different spinal levels. We have now investigated the BP vs. HR relationships within these groups and found that the depressor response after injury is different than the normal response in that the final recovery of increased HR is delayed. This finding is also true in the dysautonomia response. In the pressor response, there are simultaneous increases in both BP and HR and both BP and HR are delayed in their return to baseline. In this last group, the effect is much greater in response to rostral DCN stimulation than to caudal DCN stimulation.
To relate these cardiovascular responses to pain afferent anatomical plasticity in the spinal cord dorsal horn, we transganglionically labeled A and C fibers with CTB and IB4 respectively in the T7 and T13 DCNs. In all injury cardiovascular responses, we found increased A fiber sprouting from both DCNs relative to uninjured animals. C fibers were fewer than, or the same as, in uninjured animals in both the depressor and dysautonomia response groups but showed sprouting in the pressor response group, significantly more so at T7 than at T13. Taken together, the normal animal data and the spectrum of cardiovascular responses and anatomical plasticity after cervical SCI, it would seem that A delta afferent input and plasticity preferentially affect HR responses and C fiber input and plasticity preferentially affect BP responses. This may mean that A delta effects could be predominantly cardiac while C fiber effects could be predominantly on vascular tone.
Poster 57
Effect of daily acute intermittent hypoxia on hand function in persons with incomplete cervical spinal cord injury
1Emory University, Atlanta, GA, USA
2Shepherd Center, Atlanta, GA, USA
Loss of hand function is a debilitating consequence of cervical spinal cord injury (SCI) and often results in significant reduction in functional independence. Most cervical spinal injuries are incomplete and endogenous hand recovery via use of spared pathways is slow and often inadequate to restore normal function. Thus, there is critical need for emerging therapies that enhance recovery of hand function after SCI.
Acute intermittent hypoxia (AIH) induces neural plasticity in rodents by increasing expression of plasticity-promoting proteins and strengthening synapses onto motor neurons to yield profound recovery of breathing and walking function. In rats with cervical SCI, daily exposure to AIH (dAIH) elicits profound recovery of both breathing capacity and forelimb ladder walking (Lovett-Barr et al 2012). We recently showed that AIH also facilitates motor function in persons with chronic SCI. A single AIH exposure increased ankle torque generation, while dAIH increased walking speed and endurance. Similar facilitation has been suggested to occur in the upper limb of spinal-injured humans. However, no studies have addressed this possibility. The purpose of this study was to test the hypothesis that daily AIH (dAIH) enhances hand function in persons with chronic cervical SCI.
To test this hypothesis, we carried out a randomized, double-blinded, crossover study in which 6 male subjects with chronic, cervical SCI (44±11 years old; C5-C7) received 5 consecutive days of dAIH consisting of 15 episodes of 1.5 min hypoxia (FIO2 = 0.09) alternated with 1 min normoxia (FIO2 = 0.21). Hand function was assessed at baseline, after the first and fifth intervention day (D1 and 5), and at follow-up (F) within one week. Primary outcome measures included hand speed (Jebsen-Taylor Hand Function Test), dexterity (Box and Blocks Test), grip strength, and maximum hand opening aperture. Co-activation of antagonistic muscle pairs at the hand and wrist was also measured during maximum hand opening. All results were compared with baseline and daily exposure to normoxia (dSHAM).
Hand speed (D1 p=0.017m F p=0.010), dexterity (D5 p=0.001, F p=0.006), and pinch strength (D1 p=0.010, F p =0.040) all increased more following dAIH as compared to dSHAM. Increases in hand speed were significant from baseline at follow-up (p=0.034), while increases in dexterity and strength were not significant likely due to the small sample size. Subjects also showed a significant increase in maximum hand aperture after dAIH (D5, p=0.031). No adverse events or significant changes in blood pressure or heart rate were found.
Overall, these results showed dAIH was safe and effective in enhancing hand function in persons with incomplete cervical SCI. dAIH may be particularly effective when used as a plasticity-promoting adjuvant with current rehabilitation therapies such as mass practice.
Poster 58
Impairments and Demographics Associated with Lateropulsion after Stroke: A Logistic Regression Analysis
1Burke Rehabilitation Center, White Plains, NY, USA
2Hunter College, City University of New York, New York, NY, USA
3The Hospital for Special Surgery, New York, NY, USA
Poster 59
Robot-assisted hand exercise compared with conventional exercise therapy after ischemic stroke: A pilot study
1Columbia University Medical Center, New York, NY, USA
2Innsbruck Medical University, Innsbruck, Austria
Poster 60
Neurally dissociable information-processing components of reading deficits in subacute stroke
1Rutgers, The State University of New Jersey, Newark, NJ, USA
2Kessler Foundation, West Orange, NJ, USA,
3Rutgers-New Jersey Medical School, Newark, NJ, USA
A prevailing cognitive model of reading proposes that words are processed by interacting orthographic (spelling), phonological (sound) and semantic (meaning) information. To identify the brain regions critical for carrying out this information processing, we have so far performed neuropsychological testing and multi-sequence MRI on 5 patients with left-hemisphere stroke (<= 5 weeks post onset). We expected reading aloud to require all three components, while reading pronounceable letter strings (pseudowords, e.g., blork) can be completed without semantics. If a patient is able to read pseudowords, but not words, and especially words that do not follow typical spelling-sound patterns (e.g., yacht), this is described as surface dyslexia and is thought to arise primarily from semantic deficits. In contrast, if a patient has difficulty reading pseudowords aloud or making phonological judgments about them, this is described as phonological dyslexia and is thought to arise from deficits in mapping orthography to phonology. In our sample, two patients (P1,P3) were unimpaired across all three components, two patients (P2,P5) showed a pattern of phonological dyslexia, and one (P4) had a profound impairment in reading words and pseudowords. While lesion size predicted the overall degree of impairment, lesion location was crucial for predicting the type of impairment. In the two patients with phonological dyslexia, the damaged areas included the left corona radiata (CR)/internal capsule, with extension to thalamus and putamen in P5. P2, with a restricted CR stroke, achieved 83% accuracy in reading aloud words and only 33% accuracy for pseudowords, t(178) = 8.51, p<.0001. P5 also performed better on words (63%), than pseudowords (42%), t(178) = 2.81, p<.01. This is in contrast with similar performance on words and pseudowords in P1 and P4. P2 was relatively unimpaired on a semantic matching task (69%), but impaired on phonological (62%) and orthographic (67%) tasks compared to the two highest scoring patients. Relative to the other patients and a sample of 21 healthy controls, P2 showed exaggerated effects of word frequency, consistency, and their interaction. This is consistent with impaired orthography-phonology mapping and intact semantics. As we continue this study, we will evaluate whether spared deep fronto-parietal white matter connections to the thalamus are associated with intact orthographic-phonological mapping. Future studies will be useful investigating whether targeting reading therapies to impaired information processing components may be appropriate in the first weeks after stroke.
Poster 61
Proportional recovery of the upper limb after stroke depends on corticospinal tract integrity and not therapy
University of Auckland, Auckland, New Zealand
For most patients after stroke, upper limb impairment resolves by approximately 70% of the maximum amount possible i.e., the proportional recovery rule (PRR; Prabhakaran et al., Neurorehabil Neural Repair 2008 22: 64-71). This presentation will demonstrate how the PRR relies on corticospinal system integrity as determined with transcranial magnetic stimulation and diffusion-weighted MRI. Data were obtained from 48 patients previously enrolled in an upper limb rehabilitation interventional trial. Upper limb impairment with the Fugl-Meyer assessment (FM) was measured 2, 6, 12 and 26 weeks after stroke. Patients received a standardised dose of upper limb therapy between weeks 2 and 6 (mean dose=553 min). Linear regression indicated that patients with an FMinitial >10 recovered to a proportion (β) of 0.45, 0.60 and 0.69 of their maximum by week 6, 12 and 26 respectively, while patients with FMinitial<11 did not. Re-coding patients based on the presence or absence of paretic wrist extensor MEPs led to observed β of 0.48, 0.66 and 0.71 for MEP+ patients, with higher positive predictive power and specificity than categorisation based on FMinitial. For MEP- patients linear regression indicated that fractional anisotropy within the posterior limbs of the internal capsules (but not FMinitial) could differentiate between patients whose outcomes conforming to β≈0 versus 0<β≤0.7. Next, a separate group of 45 patients was examined within a rehabilitation setting. All were expected to make a proportional recovery based on the presence of MEPs within the first 5 days after stroke (FMinitial range 3-65). For these patients, upper limb therapy dose was not standardised (mean dose=176 min). By week 12, the observed β was 0.68. Confirmatory findings were that for both samples, upper limb impairment resolved to ≈0.7 of the maximum possible. The novel findings were that proportional recovery occurred only for patients with functional corticospinal integrity. Patients without MEPs did not conform to the PRR but dichotomised based on structural corticospinal integrity. For patients expected to make proportional recovery, a plateau is reached between 6 and 12 weeks post-stroke. Proportional recovery is insensitive to therapy dose and may reflect the spontaneous neurobiological conditions that are unique to the initial days and weeks after stroke. However, there are at least two reasons why the present results should energise those who deliver upper limb therapy after stroke. First, the majority of patients must re-learn functional tasks in the presence of at least some lingering impairment; and second, tools which provide more direct measures of corticospinal integrity allow early identification of patients with “hidden” potential for upper limb recovery.
Poster 62
Interhemispheric imbalance of primary motor cortex excitability during spontaneous recovery after stroke
1University of Auckland, Auckland, New Zealand
2Bionics Institute, Melbourne, Australia
The interhemispheric imbalance model predicts that balancing of corticomotor excitability after stroke is associated with greater improvements in motor function. However most of the evidence has come from studies of patients at the chronic stage. We carried out a retrospective analysis of longitudinal neurophysiological data collected from 46 patients during the first six months of recovery after first-ever ischaemic stroke. Transcranial magnetic stimulation was used to elicit motor evoked potentials and ipsilateral silent periods from both extensor carpi radialis muscles. We hypothesized that ipsilesional corticomotor excitability would increase during the spontaneous recovery process, and that this would be associated with greater improvements in paretic upper limb impairment and function. Corticomotor excitability became more symmetrical during the first twelve weeks of recovery. This was driven by an increase in ipsilesional stimulus-response curve slope and a decrease in ipsilesional rest motor threshold, which was associated with improved upper limb impairment and function. We found no evidence for a decrease in contralesional rest motor threshold or stimulus-response curve slope at any time within the first 6 months after stroke, and no evidence of asymmetric interhemispheric inhibition. Overall there was little support for ‘balancing’ of corticomotor excitability. Instead, neurophysiological recovery was confined to the ipsilesional primary motor cortex, with increasing excitability linked to motor recovery. Interventions applied during the sub-acute stage of stroke may be more beneficial if they facilitate ipsilesional corticomotor excitability directly, rather than attempting to do so indirectly through suppression of contralesional excitability.
Poster 63
Does Increasing Progenitor Cell Survival Improve Stroke Recovery?
University of Ottawa, Ottawa, Canada
There is a significant increase in the number of progenitor cells (PCs) born in the neurogenic niches of the adult brain under ischemic conditions. These cells have been hypothesized to enhance neurogenesis, increase plasticity within the brain and promote stroke recovery. However, it remains to be determined if the PCs per se are causally important for recovery. Additionally, a majority of the PCs die by apoptosis and only a small proportion mature into neurons, suggesting that it would be necessary for cell-based therapies to prevent apoptosis in the PCs following a stroke in order to enhance stroke recovery. To evaluate this hypothesis, we quantified recovery from stroke in an inducible transgenic mouse that ablates the pro-apoptotic protein, Bax, from the PCs and their progeny. Focal cortical sensorimotor strokes were performed using the photothrombosis model in the nestin-CreERT2/ floxed Bax (nBAX) and littermate control mice. Behavioral recovery was measured by the adhesive removal test (tactile response and asymmetries) and horizontal ladder task (forelimb and hindlimb placement, and coordination), which was assessed both prior to stroke and up to two months post stroke. Preliminary data analysis indicates photothrombosis stroke induced significant contralateral deficits in the nBAX and control mice in the adhesive and horizontal ladder tasks. In comparison to control mice, the nBAX mice had an increase in the number of PCs in the peri-infarct region at one month after removal of BAX from the PCs and their progeny, supporting that the nBAX mouse model increases PC survival. Ongoing work is increasing the sample size and adding in cognitive tests to determine if increasing PCs survival and neurogenesis is associated with improvements in learning or memory, which may aid in recovery post stroke. This work will determine if increasing PC survival is a viable option as a cell-based therapy for stroke recovery.
Poster 64
Spatial bias, the superior colliculus, and prism adaptation
1Kessler Foundation, West Orange, NJ, USA
2Rutgers New Jersey Medical School, Newark, NJ, USA
Poster 65
Stroke subtype and motor impairment influence contralesional excitability
1Laboratory for Non-Invasive Brain Stimulation and Human Motor Control, Burke Medical Research Institute, White Plains, New York, USA
2Department of Neurology, Weill Cornell Medical College, New York, New York, USA
3Centre for Neuromuscular and Neurological Disorders, University of Western Australia, Perth, Western Australia, Australia
4Feinstein Institute for Medical Research, Manhasset, New York, USA
5Laboratory of Neuromodulation, Spaulding Rehabilitation Hospital, Harvard Medical School, Boston, Massachusetts, USA
6Mechanical Engineering Department, Massachusetts Institute of Technology, Cambridge, Massachusetts, USA
7Berenson-Allen Center for Non-Invasive Brain Stimulation, Harvard Medical School, Boston, Massachusetts, USA
The non-lesioned motor cortex (M1NL) is thought to be hyperexcitable in patients with subacute or chronic stroke and offers a promising therapeutic target. However, whether M1NL excitability behaves the same for subcortical and cortical strokes is unknown. We looked for correlations between the Fugl-Meyer (FM) score and M1NL resting motor threshold (RMTNL) in 34 stroke survivors. FM correlated with RMTNL for subcortical (r=0.82; p=0.001) but not for cortical strokes (r=0.11; p=0.62). We conclude that targeting M1NL with de-excitatory protocols is indicated more for subcortical than for cortical stroke, and that different therapeutic strategies may be warranted according to lesion location.
Poster 66
Putting transcranial magnetic stimulation, diffusion tensor imaging and functional MRI to the test: A study of interhemispheric imbalance in chronic stroke
1Cleveland Clinic, Cleveland, OH, USA
2Kent State University, Kent, OH, USA
3Kessler Foundation Research Center, West Orange, NJ, USA
It is believed hand deficits persist in stroke due to an imbalance between ipsilesional and contralesional hemispheres. Imbalance has been characterized in a few different ways: output of the corticospinal tracts as well as their integrity using transcranial magnetic stimulation (TMS) and diffusion tensor imaging, mutual transcallosal inhibition studied using TMS, and cortical activation in movement of paretic hand noted with functional MRI (fMRI). However, it is unclear whether different substrates describing imbalance even offer complementary perspectives, and how they subtend clinical function. Across ten stroke patients (63±9 years) with chronic upper-limb paresis, we examined associations between substrates of hemispheric imbalance, and their relation to two widely used outcomes- one measuring impairment and the other perceived disability in voluntary use of paretic hand. We have found that patients with poorer integrity of corticospinal tracts in the ipsilesional hemisphere show greater output of these tracts in the contralesional. However, neither an imbalance in their integrity nor an imbalance of their output relates to transcallosal inhibition. As a converse, imbalance in cortical activation was associated with transcallosal inhibition. Patients with relatively high fMRI activation within ipsilesional than contralesional motor cortices not only possessed stronger ipsilesional corticospinal output, but also showed balance of mutual transcallosal inhibition. Clinically, while patients with poorer integrity of corticospinal tracts in the ipsilesional hemisphere showed greater impairments, those with reduced ipsilesional than contralesional cortical activation had greater perception of disability. In conclusion, although output of contralesional corticospinal tracts and ipsilesional damage relates, mutual callosal influence is only associated with relative cortical activation between hemispheres. Still, viability of corticospinal tracts appear useful in categorizing range of specific impairments, and helping realize potential offered by the contralesional hemisphere in recovery, while fMRI activation serves to mark disability in volitional use of the weak hand.
Poster 67
Positive dose response relationship for upper limb rehabilitation after stroke delivered unsupervised at home using the video game Circus Challenge
Newcastle University, Newcastle upon Tyne, UK
Circus Challenge comprises 10 mini games professionally produced specifically for upper limb rehabilitation controlled by 100 different bimanual movements, which together form the basis for activities of daily living. The time performing therapeutic control moves (not simply playing the game) is automatically recorded. Patients played the games unsupervised in their home over 12 weeks. The Fugl Meyer Upper Extremity Assessment (FMUEA) and the Chedoke McMaster Arm and Hand Activity Inventory (CAHAI) were undertaken at baseline and at 12 weeks and the changes in scores calculated. A MANOVA was performed for Acute and Chronic patient groups - Dependent variables: Change in FMUEA and CAHAI scores; Fixed factors: Sex, Hemisphere of stroke; Covariates: Dose - total time performing therapy moves, Baseline scores, Age, Time from stroke.
Poster 68
An Algorithm Assessing Upper Limb Function After Stroke From Action Video Gameplay For Remote Monitoring Of Home Based Rehabilitation: Validity And Sensitivity To Change
Newcastle University, Newcastle upon Tyne, UK
Poster 69
The impact of metabolic syndrome on cortical microvasculature - form and function
1Canadian Partnership for Stroke Recovery, Ottawa, Canada
2Neuroscience Program, University of Ottawa, Ottawa, Canada
3Cellular and Molecular Medicine, University of Ottawa, Ottawa, Canada
4Sunnybrook Research Institute, Toronto, Canada
5Medical Biophysics, University of Toronto, Toronto, Canada
Metabolic syndrome (MetS) is a leading risk factor for ischemic stroke. Given that nearly 25% of the world population has MetS, the number of stroke patients needing treatment will rise considerably in the coming years. As it stands, the only post-stroke therapy available to patients is rehabilitation. Animal models are extensively used to optimize human rehabilitation by developing therapies aimed at enhancing neuroplastic mechanisms. Nonetheless, the majority of animals used in stroke rehabilitation studies are young and healthy - they lack the co-morbidities of stroke patients. To better approximate the clinical condition, our lab is using a model of MetS, the Cafeteria (CAF) diet. Before initiating rehabilitation studies with this model, we will confirm that this diet exerts effects on the cerebral vasculature at the structural and, even more importantly, at the functional level.
CAF rats were provided with free access to standard rodent pellets and water as well as a varied daily supply of highly processed human junk food items (i.e. chips, cookies, bacon, etc.) and a 12% sucrose solution. After 3 months of treatment, CAF animals exhibit 4/5 criteria for MetS. CAF animals are overweight (CAF=767±16g, SD=676±13g), have increased circulating triglycerides (CAF=183±21mg/dl, SD=95±9mg/dl), reduced HDL cholesterol (CAF=70±4mg/dl, SD=94±3mg/dl) and impaired glucose tolerance (insulin: CAF=7.1±1ng/ml, SD=3.3±0.5ng/ml; blood glucose two hours after i.p. GTT test: CAF=8.5±0.7mmol/L, SD=6.7±0.3mmol/L).
We have succeeded in generating a rat model of MetS that closely approximates the clinical condition. Histological study of the cerebral microvasculature is underway with a focus on neovascularization following stroke as well as on damage to the endothelial wall. Further analysis is being performed to determine whether or not structural damage to the vessels is accompanied by functional deficits, as assessed via in vivo two-photon fluorescence microscopy. Concurrently, cognitive testing is being performed to determine whether CAF animals exhibit signs of emerging vascular cognitive impairment.
Poster 70
Use of Liquid Consistency Modification (LCM) and Augmented Hydration (AH) Orders for dysphagic patients following stroke
Weill Medical College of Cornell University at Burke Rehabilitation Hospital, White Plains, NY, USA
Poster 71
Effects of movement duration on use of the affected limb in individuals post-stroke
University of Southern California, Los Angeles, USA
Patients with stroke often exhibit non-use of their more affected arm: Although capable of generating arm movements, they often chose not to. Here, we hypothesize that the slowness of movements with the more affected side is a significant factor underlying such non-use. Twelve individuals with chronic stroke (mild to moderate impairment; 46.1± 9.0 on Fugl-Meyer assessment) and six age-matched non-disabled participants performed the Bilateral Arm Reaching Test (BART; Han, Kim, et al. 2014) in the forced and free choice conditions with three different movement time constraints: no time constraint, medium (~1000 ms, depending on target location) and fast (~ 500 ms). Arm kinematics, movement time, and task success were recorded across conditions and amount of use in the free choice condition was computed and compared to normative data. Whereas the non-disabled group showed no differences in hand choice across conditions, the stroke group showed decreased affected hand use in the faster conditions. A logistic regression model showed that the difference in affected limb use in the fast condition is predicted by the difference in movement duration between the affected and the less affected limb in the no time constraint condition. Further, individuals with left hemiplegia showed a dramatic decrease in use in the fast condition (68% decrease in paretic limb use compared to no time constraint condition), whereas individuals with right hemiplegia showed a only a 24% decrease in paretic limb use. We discuss our results in lights of the known right/left hand differences in arm control and in the framework of delayed rewards discounting.
Poster 72
Voxel-based lesion symptom mapping of factors related to language deficits after stroke
1Georgetown University Medical Center, Washington, DC, USA
2MedStar National Rehabilitation Hospital, Washington, DC, USA
3First Affiliated Hospital Sun Yat-Sen University, Guangzhou, China
Characterizing language deficits after left hemisphere stroke and understanding how they relate to lesioned tissue is complicated by the fact that multiple cognitive and linguistic skills are involved in performing any language task. To begin to address this complex problem, we combined factor analysis of behavioral data with voxel-based lesion symptom mapping (VLSM) of structural MRI data from 35 people with aphasia. Factor analysis is a statistical technique that provides insight into the relationship between multiple variables by creating a small number of factors onto which all variables load to varying degrees. VLSM relates brain damage to behavior on a voxel-by-voxel basis, so that the contributions of each area to a language task can be assessed.
The factor analysis of 26 different language and cognition tests using Eigenvalue>1 as the cutoff yielded 4 factors: word retrieval, speech comprehension, phonology/repetition, and executive function. Though a wide range of tests were included in the battery, the scores loaded logically onto these 4 factors. For example, the Philadelphia Naming Test loaded heavily on word retrieval, but very little on the other 3 factors. The yes/no questions from the auditory comprehension section of the Western Aphasia Battery loaded heavily on speech comprehension, but very little on the other 3 factors. Complete results from all tests will be shown. In order to determine how these factors related to the lesions of the 35 people with aphasia included here, we combined the results of the factor analysis with a VLSM analysis. Results showed that voxels were more likely to be lesioned in areas expected to be involved in tasks that loaded on the 4 factors described. The executive function factor was associated with middle frontal gyrus and fronto-parietal white matter lesions, the phonology factor with lesions to the ventral premotor and motor cortex, the comprehension factor with superior temporal areas, and the word retrieval factor primarily with inferior frontal cortex.
Poster 73
The Functional Role of Adult Neurogenesis in Promoting Stroke Recovery
1Cellular and Molecular Medicine, University of Ottawa, Ottawa, ON, Canada
2Canadian Partnership for Stroke Recovery, Ottawa, ON, Canada
3Neuroscience, National Institute of Health, Bethesda, MD, USA
Following a stroke patients have an innate capacity to regain function. The cause of this recovery is proposed to be due to many forms of plasticity including adult neurogenesis, the process of generating new neurons in the adult brain. Preclinical studies in rodents have shown a significant increase in the number of adult generated progenitor cells (PCs) post-stroke and ectopic migration of PCs to the site of damage. In order to determine the function of the PCs, a few studies have ablated the PCs prior to generation of large strokes and have reported that PCs support stroke recovery. However, the timing of this effect and whether PCs would aid in recovery from a focal cortical stroke remains unknown. Therefore, this study aims to determine if neurogenesis per se is required for stroke recovery by using an unpublished transgenic GFAP-TK rat model (Heather Cameron, NIH). This model allows for the inducible deletion of GFAP-expressing cells in the adult brain by treatment with the antiviral drug, valganciclovir. PCs were ablated prior to a focal cortical stroke produced through injections of endothelin-1 (ET-1), a vasoconstrictive peptide, into the forelimb motor cortex. Long-term behavioral recovery was measured through three motor tasks: the staircase, cylinder and beam walk, in the GFAP-TK rats and littermate control wild-type (WT) rats. Preliminary results show that both the controls (vgcv treated and non-treated WT rats, and non-treated GFAP-TK rats) and vgcv treated GFAP-TK rats had significant deficits on the contralateral side in the staircase, cylinder and beam walk tests up to 4 weeks post-stroke. These results suggest that PCs are not required for recovery of motor function following a focal cortical stroke and ongoing studies are increasing sample size to confirm these findings. Additionally, we are testing whether neurogenesis per se is required for learning and memory during stroke recovery. Overall, this work will clarify the functional role of PCs during stroke recovery and thus help define their therapeutic potential for stroke patients.
Poster 74
Influence of depression and cognitive deficits on use of feedback for upper limb recovery in chronic stroke
1School of Physical and Occupational Therapy, McGill University, Montreal, Quebec, Canada
2Department of Neurosciences, University of Montreal, Montreal, Quebec, Canada
3Feil-Oberfeld Research Centre, Jewish Rehabilitation Hospital site of Centre for Interdisciplinary Research in Rehabilitation of Greater Montreal (CRIR), Laval, Quebec, Canada
4School of Rehabilitation Sciences, University of Ottawa, Ottawa, Ontario, Canada
Poster 75
Vitamin D Levels and Stroke Severity in the Acute Inpatient Rehabilitation Setting
1Hamilton College, Clinton, NY, USA
2Burke Rehabilitation Hospital, White Plains, NY, USA
Poster 76
Can inertial sensors characterize treatment-induced skill acquisition in chronic stroke?
University of Southern California, Los Angeles, CA, USA
In individuals with stroke, kinematics of upper extremity (UE) movements can provide performance-based information useful to distinguish between the restitution of skilled movements from those performance gains associated with compensatory movement strategies. A challenge of assessing kinematics is the difficulty in quantifying movement quality without expensive laboratory equipment. Wearable inertial sensors that measure limb accelerations and rotations can be used to quantify treatment-induced gains in motor control by characterizing the degree to which movement of the paretic UE approaches a gold standard multi-joint pattern. In this preliminary work, we demonstrate the use of a webcam and two synchronized wireless inertial sensors that are each comprised of a tri-axial accelerometer, gyroscope, magnetometer, and altitude sensor. Our purpose is to determine if inertial sensors can be used to evaluate movement quality through the application of principal component analysis (PCA). Methods: This study was undertaken as part of a larger clinical trial in which individuals with stroke received a physical therapy intervention. Sensors were placed on the proximal and distal aspect of the paretic arm during the performance of a goal-directed, manipulation task--turning a knob using a grasp with repeated forearm pronation/supination movements. We describe the time course of skill acquisition in movement strategies for eight individuals with stroke at three time points. Principal component analysis (PCA) was used to identify the main contributing motions to the execution of the task. Results: PCA results reflected both between- and within-subject differences in multi-joint control. With practice, improvement in skill is characterized by decreased rotation of the proximal arm, and an increase in uni-planar rotation of the distal arm. Discussion: PCA can be used as a means to evaluate changes in movement quality both within and across subjects. We demonstrate that PCA can extract the key planes of motion in a discrete task, and provide a quantitative measure of skilled multi-joint control. This is a novel application of inertial sensors to capture the characteristics of movement, and by utilizing multiple sensors on an extremity, we can provide further insight than is typically afforded by observation alone. This work also forms a platform for developing feedback systems for patients, and may have application in the home environment.
Poster 77
Effects of task-oriented Exoskeleton Robotic Hand Training on Motor Function Recovery in Chronic Stroke: a six month follow-up study
Neurorehabilitation and Robotics Laboratory, Interdisciplinary Division of Biomedical Engineering, the Hong Kong Polytechnic University, Hung Hom, Hong Kong
Poster 78
Improved spasticity correlates with change in contralesional cortical thickness, following upper limb motor therapy
1Case Western Reserve University School of Medicine, Cleveland, OH, USA
2Cleveland VA Medical Center, Cleveland, OH, USA
3College of Medicine, University of Florida, Gainesville, FL, USA
4MR DVA Medical Center, Gainesville, FL, USA
Poster 79
Changes of Motor Recovery in Chronic Stroke Patients
1Universidad Nacional de Colombia, Bogotá, Colombia
2Instituto Roosevelt, Bogotá, Colombia
3Centro de Investigación en Fisiatría y Electrodiagnóstico - CIFEL, Bogotá, Colombia
between 2 consecutive clinical records [Average time between assessments = 8.3 months] in scores of the following scales: Fugl-Meyer
Motor Scale (FM), Postural Assessment Scale For Stroke Patients (PASS), Five Times Sit To Stand Test (FTSST), Modified Rankin Scale (MRS), Barthel Index, Composite Functional Index, Modified Ashworth Scale.
This study strengthens the knowledge about the pattern of functional recovery and prognosis of motor recovery in chronic stroke patients, providing a basis for objective and accurate assessment of therapeutic responses in clinical practice and future research protocols.
Poster 80
Combination of transcranial magnetic stimulation, botulinum toxin A and intensive occupational therapy facilitate functional recovery of stroke patients
1Fukuoka Seisyukai Hospital, Fukuoka, Japan
2Kumamoto Health Science University, Kumamoto, Japan
【Purpose】We examined the effect of combined therapy with low frequency (1Hz) repetitive transcranial magnetic stimulation (rTMS), botulinum toxin A (BTX-A) and intensive occupational therapy (iOT) for functional recovery of paralyzed upper extremities (U/E) after stroke.
【Methods】The study sample includes 18 stroke patients in both recovery (n = 7) and chronic (n = 11) stages who were treated with the combined therapy for 14 days in inpatient hospital settings. The patients at a recovery stage were within 6 months of stroke onset and did not have any improvement of paralyzed U/E before the combined therapy at least for one month. The patients at chronic stage were those whose function of the paralyzed U/E did not change for at least 6 months after the completion of a conventional inpatient rehabilitation program. In order to assess the condition of the corticospinal tract integrity for each patient, we measured fractional anisotropy (FA) at cerebral peduncle using functional MRI, and then, FA asymmetry was calculated for all of 18 patients (Stinear CM et al., 2007). For rTMS treatment, 1,200 stimulations of 1Hz at 90% of rest motor threshold were applied over the motor area on non-lesional hemisphere of brain twice a day. Each patient received the iOT of 2 hours twice a day just after each rTMS application. BTX-A was administered to 10 patients with U/E spasticity on the first day of training. The following measures were evaluated before and after the combined therapy: Fugl-Meyer Assessment (FMA), Action Research Arm Test (ARAT) and Wolf Motor Function Test (WMFT). Wilcoxon’s signed rank test was used to examine the differences.
【Results】FA asymmetry was not associated with the prognosis of U/E function. The scores and time of FMA, ARAT and WMFT (n = 18) before the therapy were 32.2 ± 2.5 (average ± SE), 18.4 ± 3.7 and 658 sec, respectively. These scores and time after two weeks of the therapy were 41.9 ± 2.7, 27.5 ± 4.6 and 488 sec, respectively, showing significantly improvement with the therapy (FMA, p < 0.001; ARAT, p < 0.001; WMF, p < 0.01).
【Conclusion】 Two weeks of the combined therapy for stroke patients exhibited functional improvements of paralyzed U/E.
Poster 81
Heart rate variability is associated with upper extremity recovery after stroke
University of Pittsburgh, Pittsburgh, PA, USA
Among the 795,000 individuals who sustain a stroke annually in the United States, almost 65% are unable to use their affected upper extremity (UE) in daily tasks resulting in costly resource use to support long-term independence. UE rehabilitation is most efficient when restorative interventions, which seek to improve affected UE function, are focused on patients with good potential for UE recovery early in the rehabilitation process. Unfortunately current predictors such as location and volume of stroke and initial UE impairments are poor predictors of UE recovery after stroke. Identifying novel biomarkers of UE recovery will allow clinicians to more accurately predict patients with potential for UE recovery and provide targeted interventions to enhance long-term independence.
Heart rate variability (HRV) or the temporal variations between consecutive heartbeats is a plausible proxy marker for the integrity of cortical pathways controlling UE function. HRV directly measures the intact function of the vagus nerve, which controls the autonomic functions of the heart, and is modulated by the cortical pathways controlling UE function. We examined whether HRV upon admission to acute inpatient rehabilitation is associated with UE recovery three months after stroke. We hypothesized that individuals with higher HRV at acute inpatient rehabilitation admission would exhibit greater UE recovery after 3 months. This is the first step in developing a prediction model using HRV to predict UE recovery after stroke.
A 12-lead Holter monitor was attached for 24-hours to 10 patients with stroke [mean age = 61 years (SD =12), 7 females, 7 ischemic and 3 hemorrhagic stroke] within three days of admission to acute inpatient rehabilitation. Standard deviation between the consecutive heartbeats (SDNN) was used to quantify HRV. Fugl Meyer Upper Extremity Subscale (FMUE) was used to assess UE recovery at three months after stroke. In our sample, HRV upon admission was strongly (R2= .80) and significantly (p =.009) associated with three month FMUE scores. Individuals who had higher HRV showed greater UE recovery three months after stroke.
Our findings suggest that HRV is a plausible biomarker to predict UE recovery after stroke. Using HRV is novel, inexpensive, and easy to administer. Future studies will examine whether HRV can predict UE recovery after controlling for covariates (age, stroke location and volume, time since stroke and amount and type of rehabilitation).
Poster 82
Interhemispheric frontal resting connectivity increases in post-stroke aphasia and is associated with worse performance
1Georgetown University, Washington, DC, USA
2MedStar National Rehabilitation Hospital, Washington, DC, USA
The role of the right hemisphere in aphasia recovery has been debated for over a century. Some argue that the right hemisphere plays a compensatory role, aiding recovery (e.g., Basso et al., 1989), while others posit that right hemisphere activity interferes with recovery (e.g., Barwood et al., 2011). Recently, research has shifted to examining connectivity, rather than activation levels, in order to better understand neural patterns that explain aphasia symptoms and recovery (Bonilha et al., 2014). This approach is critical because some differences in task-related activity in aphasia may relate to differences in the effort required for task performance rather than actual reorganization of language networks. This experiment examined the resting state networks associated with right BA 44 and performance on a range of language tasks. Twenty participants with left hemisphere lesions and aphasia diagnoses, as well as 21 age matched controls, participated in this study. The participants underwent a 7 minute T2* weighted resting state MRI scan, as well as a high-resolution structural scan. Participants also underwent a battery of language and other cognitive tests. The time course in right BA 44 was extracted for each participant. The model used the time course in right BA 44 as the predictor, and included motion parameters as covariates. At the group level, participants with aphasia showed greater connectivity to right BA 44 in the left and right middle temporal gyrus, right inferior temporal lobe and left insula, compared to controls. The peak difference in connectivity was in the left insula. Participants in the aphasia group were then grouped based on whether each individual’s lesion overlapped with this peak (10 in each group). Participants with lesions overlapping with the peak showed significantly impaired performance on a range of language tasks, relative to patients whose lesions did not overlap with the left insula, and controlling for lesion size. Finally, connectivity between the left insula and right BA 44 was measured for each aphasic participant who had a preserved left insula. Connectivity in this group correlated negatively with performance tests of oral word reading tasks and speech apraxia, controlling for lesion size. These results demonstrate that connectivity between left and right frontal lobes increases after a left hemisphere stroke, and that this over-connectivity is related to worse performance on certain speech measures. Additional analyses will be needed to determine if similar connectivity relationships are observed with other areas of the right hemisphere.
Poster 83
Multimodal predictors of rehabilitation related recovery in stroke
Cleveland Clinic, Cleveland, OH, USA
Stroke is the leading cause of upper limb disability. Unfortunately, variability in stroke location and impairment results in different rates of recovery. Therefore, in order to accommodate the wide patient variability, therapists often use functional tests to aide in development of individualized therapy. However, while baseline function is significant in predicting recovery, it is often limited since patients with the same level of impairment can improve via different neural processes. Knowing the unique process of recovery would allow one to individualize therapies for better outcomes. Therefore, here we investigated if initial assessments of corticospinal integrity (diffusion tensor imaging), corticospinal output (transcranial magnetic stimulation), interhemispheric inhibition (ipsilateral silent period), or cortical activation (functional magnetic resonance imaging) could be predictors of recovery following a five week therapy session. Recovery was defined in terms of impairment, dexterity, and perceived disability. Stepwise linear regressions were used to find predictors of recovery. In addition, we also determined if patients’ neural substrates could be predictors of long-term (3 month) follow-up recovery. Notably, we found that predictors of alleviating impairments were both corticospinal integrity (β=0.537, R2=0.975, p=0.001) and corticospinal output (β = -0.604, R2=0.975, p=0.001). Corticospinal integrity and corticospinal output were also found to be predictors of perceived disability, but only after long-term follow-up (β=0.814, R2=0.999, p=0.001; β=0.613, R2=0.999, p=0.003). Finally, we identified interhemispheric inhibition as the only significant predictor of dexterity improvement (β=0.781, R2=0.61, p=0.013). It is important to note that cortical activation did not show significant contribution to recovery. . However, while baseline function is significant in predicting recovery, the only baseline test found to be a predictor of recovery was perceived disability (β=0.726, R2=0.527 p=0.027). Therefore, the results from our study suggest that corticospinal output and integrity may be a better measure to predict reduction of impairments in stroke patients. Likewise, recovery of dexterity may be best predicted by interhemispheric inhibition. Therefore, future work should consider including measures of corticospinal integrity, corticospinal output, and interhemispheric inhibition to further tailor therapy to individual patients’ neural processes.
Poster 84
Aphasia severity relates to right supramarginal gyrus grey matter volume after accounting for lesion-related factors in chronic stroke
1Georgetown University Medical Center, Washington DC, USA
2MedStar National Rehabilitation Hospital, Washington DC, USA
3First Affiliated Hospital, Sun Yat-Sen University, Guangzhou, China
Aphasia is a common consequence of left hemispheric stroke. Evidence has shown that lesion size and location in the left hemisphere contribute to severity of post-stroke aphasia. Enhanced activation in perilesional left hemispheric areas as well as activation in certain areas of right hemisphere have been observed after damage to the left hemispheric language network. However, the potential contribution of residual left or right hemispheric areas to aphasic deficits remains elusive. In the present study, we assessed whether grey matter (GM) in preserved brain areas was associated with aphasia severity in chronic stroke.
Twenty-four patients (age: 58.6±10.2) with chronic post-stroke aphasia and 26 healthy controls (age: 61.0±10.4) participated in the study. All patients were administered a battery of language and cognitive tests and a high-resolution 3D T1-weighted structure image was collected from all patients and controls. Voxel-based Lesion-Symptom Mapping (VLSM) was performed to determine the lesion location that best predicted impairment on different language measures, and the lesion status at the peak voxel in each VLSM analysis was recorded for each patient in order to account for variance related to lesion location in subsequent analyses. For each language measure, voxel-based morphometry (VBM) was then used to examine the correlation between GM volume and the language score, while including age, time since stroke, total lesion volume, and lesion status from the VLSM analyses as nuisance covariates. Regions-of-interest (ROIs) were then applied to compare the difference in GM volume between patients and controls.
The VBM results showed that when controlling for age, total lesion volume, time after stroke, and lesion status, the GM volume within right supramarginal gyrus (SMG) positively correlated with aphasia quotient, spontaneous speech and repetition (p<0.05, AlphaSim correction). Volumes of right SMG negatively related to age in patients (p=0.03). No correlation was found between right SMG volumes and lesion size (p>0.05), and no difference was found between SMG volumes in patients and controls (p>0.05).
These findings suggest that right SMG volume contributes to aphasia severity in chronic stroke. The lack of relationship with lesion volume and the lack of difference in GM volume between patients and controls suggest that SMG volume may not change markedly after the stroke, but rather that right SMG volume prior to left hemisphere stroke determines resilience to aphasia. Longitudinal studies of GM anatomy after left hemisphere stroke will be needed to examine these relationships more thoroughly.
Poster 85
Recovery of Life Role Activities and Underlying Impairment Gains In Response to Comprehensive, Integrated, Milieu Intervention for TBI Survivors
1University of Florida, COM, Dept of Neurology, Gainesville Florida, USA
2Brain Rehabilitation Research Center of Excellence, Gainesville VA Medical Center, Gainesville Florida, USA
3Louis Stokes Cleveland VA Medical Center, Cleveland Ohio, USA
Poster 86
Sensory processing and sensory augmentation for balance control in chronic post-concussive syndrome
Oregon Health & Science University, Portland, USA
Although postural instability is one of the most common and debilitating complaints in post-concussive syndrome (PCS), we currently lack sensitive, objective measures that characterize postural deficits with sufficient specificity to guide rehabilitation. The goals of this study are to 1) explore sensory processing strategies for balance control in PCS subjects and 2) explore if augmentation of sensory information improves balance in people with PCS.
Methods:
Results:
Poster 87
Educate, Train, Treat, Track: Bringing State of the Art Care to our Military with Traumatic Brain Injury
1US Army, OTSG, R2D, Falls Church, VA, USA
2University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
3Sister Kenny Research Center, Minneapolis, MN, USA
4Oak Ridge institute for Science and Education, Bellcamp, MD, USA
A poster describing the U.S. Army Traumatic Brain Injury (TBI) program. Displays progress from the U.S. Army TBI Task Force. Also includes 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. Shows the evolution of, and current, policies and clinical algorithms in the deployed and garrison environments as well as DoD clinical recommendations. Poster includes the Neurocognitive Assessment Tool and role of neurocognitive assessment in return to duty decision making. Presents the DoD TBI coding procedures and challenges in analyzing coded data. Shows Army TBI Research initiatives related to TBI. Finally, displays the Army TBI education and training strategies used to educate a widely-dispersed population of medical staff and providers and specific tools and resources developed to support the TBI mission to include patient education handouts, multi-media educational tools and slide decks, and the TBI Rehabilitation ToolKit.
Poster 88
Creatine Monohydrate as a neuroprotective supplement for mild traumatic brain injury
Centre for Brain Research, The University of Auckland, Auckland, New Zealand
Alterations in brain metabolites after traumatic brain injury (TBI) suggest a period of metabolic vulnerability occurs following the initial trauma, which may persist despite an absence of symptoms and normal appearance of brain structure. However, when subjected to metabolic stressors after injury the brain appears to be incapable of restoring energetic homeostasis and measurable neuropsychological deficits emerge.
Creatine is a naturally occurring compound involved in the buffering, transport and regulation of cellular energy. Dietary creatine supplementation has been associated with improved symptoms in neurological disorders defined by impaired neural energy provision. Creatine is also neuroprotective in vitro against anoxic/hypoxic damage. The utility of creatine supplementation to protect against energetic insult in vivo remains to be investigated in humans.
We present data from healthy individuals who underwent oral creatine supplementation (or placebo control) and neurophysiological and neuropsychological assessment during acute oxygen deprivation - a metabolic stressor that induces energy crisis and cognitive symptoms similar to those which occur following TBI. We also present pilot data from patients with mild TBI.
The healthy volunteers participated in a week-long placebo-controlled creatine monohydrate supplementation protocol within a double-blind, crossover design. Creatine concentration in the sensorimotor cortex was on average 9.2% greater following supplementation (Cr: 7.0 ± 1.6 mmol/L, Pla: 6.4 ± 0.9 mmol/L; p = 0.04) as measured using magnetic resonance spectroscopy. A hypoxic gas breathing protocol severely reduced arterial oxygen saturation (SaO2, Cr: 79 ± 8%, Pla: 79 ± 11%; p < 0.01) and impaired a wide range of neuropsychological processes consistent with common symptoms of mild TBI. A composite neurocognitive index score was reduced by 12 ± 20% (p = 0.02). Impairments were most pronounced for measures of attention (complex attention: -29 ± 55%; p = 0.03). Corticomotor excitability increased in response to hypoxia with creatine supplementation (Δ ΣMEP, Cr: 170 ± 91% vs. Pla: 127 ± 60%; p = 0.04) and restored hypoxia-induced decrements in neuropsychological performance (neurocognitive index -4 ± 12%, complex attention: -8 ± 18%).
Deficits in cognitive function that occur during oxygen deprivation are consistent with dysfunction in frontal and hippocampal neural circuits, similar to injury that occurs with mild TBI. Creatine appears to be neuroprotective toward these deficits, perhaps as a result of a neuromodulatory action when cellular energy provision is compromised.
Dietary creatine monohydrate supplementation augments neural creatine, increases corticomotor excitability, and prevents decline in cognition that occurs during severe oxygen deficit and energy crisis. Seven days of creatine monohydrate supplementation has potential utility as a neuroprotective supplement in those recovering from mild TBI.
Poster 89
Assessing Motor Performance Following mTBI in Military Service Members Using Body Worn Inertial Sensors
University of North Carolina at Chapel Hill, Chapel Hill, NC, USA