UBC Faculty Research and Publications

Repetitive reaching training combined with transcranial Random Noise Stimulation in stroke survivors… Hayward, Kathryn S; Brauer, Sandra G; Ruddy, Kathy L; Lloyd, David; Carson, Richard G May 30, 2017

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12984_2017_Article_253.pdf [ 881.19kB ]
JSON: 52383-1.0366878.json
JSON-LD: 52383-1.0366878-ld.json
RDF/XML (Pretty): 52383-1.0366878-rdf.xml
RDF/JSON: 52383-1.0366878-rdf.json
Turtle: 52383-1.0366878-turtle.txt
N-Triples: 52383-1.0366878-rdf-ntriples.txt
Original Record: 52383-1.0366878-source.json
Full Text

Full Text

RESEARCH Open AccessRepetitive reaching training combined withtranscranial Random Noise Stimulation inConclusion: Reaching training that includes tRNS timed to coincide with generation of voluntary motor commands isfeasible. Clinical improvements were possible even in the most severely affected individuals as evidenced by CSTZCTR)ntHayward et al. Journal of NeuroEngineering and Rehabilitation  (2017) 14:46 DOI 10.1186/s12984-017-0253-yMovement Lab, ETH Zurich, Zurich, SwitzerlandFull list of author information is available at the end of the article* Correspondence: kathy.ruddy@hest.ethz.ch3Department of Health Sciences and Technology, Neural Control ofTrial registration: This study was registered on the Australian and New Zealand Clinical Trials Registry (ANhttp://www.ANZCTR.org.au/ACTRN12614000952640.aspx. Registration date 4 September 2014, first participadate 9 September 2014.Keywords: Stroke, Upper limb, Non-invasive brain stimulation, Magnetic resonance imaging, Functionintegrity.arm paresis is feasible: a pilot, triple-blind,randomised case seriesKathryn S. Hayward1,2, Sandra G. Brauer1, Kathy L. Ruddy3*, David Lloyd4 and Richard G. Carson5,6AbstractBackground: Therapy that combines repetitive training with non-invasive brain stimulation is a potential avenue toenhance upper limb recovery after stroke. This study aimed to investigate the feasibility of transcranial Random NoiseStimulation (tRNS), timed to coincide with the generation of voluntary motor commands, during reaching training.Methods: A triple-blind pilot RCT was completed. Four stroke survivors with chronic (6-months to 5-years) and severearm paresis, not taking any medications that had the potential to alter cortical excitability, and no contraindications totRNS or MRI were recruited. Participants were randomly allocated to 12 sessions of reaching training over 4-weeks withactive or sham tRNS delivered over the lesioned hemisphere motor representation. tRNS was triggered to coincidewith a voluntary movement attempt, ceasing after 5-s. At this point, peripheral nerve stimulation enabled full rangereaching. To determine feasibility, we considered adverse events, training outcomes, clinical outcomes, corticospinaltract (CST) structural integrity, and reflections on training through in-depth interviews from each individual case.Results: Two participants received active and two sham tRNS. There were no adverse events. All training sessions werecompleted, repetitive practice performed and clinically relevant improvements across motor outcomes demonstrated.The amount of improvement varied across individuals and appeared to be independent of group allocation and CSTintegrity.stroke survivors with chronic and severe© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.Hayward et al. Journal of NeuroEngineering and Rehabilitation  (2017) 14:46 Page 2 of 9BackgroundIt is estimated that 30% of stroke survivors have severeupper limb impairment [1], whereby the functional cap-acity of the paretic arm is diminished to the extent thatit cannot be moved against gravity [2]. For these individ-uals, who do not have sufficient movement with whichto work, the provision of effective therapy can be chal-lenging. The associated consequences are poor prospectsfor recovery [3], limited rehabilitation opportunities [4],and ultimately reduced quality of life (QoL) [5]. Yet, iftask-oriented practice can be made possible by somemeans, there exists the potential to promote motor re-covery, and in turn make a significant positive impactupon individual QoL and alleviate burden of care. Inseeking to achieve levels of task-oriented practice be-yond those that are possible through traditional therapyalone, attention has therefore turned to enabling tech-nologies, including “assistive” devices, and adjuvantmethods such as peripheral nerve and brain stimulation.Best evidence syntheses [6, 7] suggest that goal-directed movements can be assisted by minimizing themechanical degrees of freedom to be controlled, in com-bination with the augmentation of voluntary muscle ac-tivity via peripheral nerve stimulation of target muscles,or the use of mechanical actuators. To encourage positivechanges in motor performance, the capacity to increasetask difficulty through small, yet incremental progressionsand provision of meaningful real-time visual and auditoryfeedback have also been highlighted [8, 9]. The authorshave previously sought to implement these principles,using the Sensorimotor Active Rehabilitation Training ofthe Arm (SMART Arm) device to promote functional re-covery in severely impaired stroke survivors [8–10]. It hasbeen shown that 4-weeks (12-h) of community-basedtraining of reaching in people greater than 6-months poststroke improved upper limb function (and increasedreaching distance) [8], enhanced the specificity of musclerecruitment (elevated ratio of biceps to triceps activationduring reaching) [11], and accentuated corticospinal re-activity (decreased motor evoked potential [MEP] onsetlatency) [12]. Of particular interest in the context of thecurrent study is the observation that not all individualsachieved functional gains. In these cases, the intrinsicneurobiological reserve of the injured brain may havebeen insufficient for repetitive training alone to drive re-covery of motor function.A variety of non-invasive brain stimulation (NIBS)techniques are now being used with the aim of altering theexcitability of brain networks that have the potential to beengaged during the execution of motor tasks. The mostcommonly applied NIBS techniques are transcranial-directcurrent stimulation (tDCS) and repetitive-transcranialmagnetic stimulation (rTMS) [13]. In general, the applica-tion of these techniques is predicated on the assumptionthat by altering the state of circuits within (contralateral)primary motor cortex (M1) in a manner that produces sus-tained increases in the excitability of corticospinal projec-tions to the impaired limb (or by decreasing the excitabilityof circuits in the M1 ipsilateral to the impaired limb),therapeutic gains will be realised. The fact that these ap-proaches have limited efficacy in severely impaired strokesurvivors notwithstanding [14], there exist other forms oftherapeutic NIBS that are motivated by a different premise.It is well established that in some circumstances, theaddition of random interference or noise, enhances thedetection of weak stimuli, or the information content ofa signal (e.g., trains of action potentials) [15]. In light ofthis phenomenon, it has been proposed that the applica-tion of transcranial random noise stimulation (tRNS)may boost the adaptive potential of cortical tissue [16].The present investigation is motivated by the conjecturethat: if the delivery of random noise stimulation is timedto occur simultaneously with the generation of voluntarymotor commands, it may serve to amplify functional ad-aptations invoked by the intrinsic neural activity.Implemented through a triple-blind pilot randomisedcontrol design, the specific aim of this study was to es-tablish the feasibility of delivering tRNS, timed to coin-cide with the generation of the voluntary motorcommands, in the context of reaching movements per-formed by individuals with chronic and severe upperlimb paresis after stroke. Recognising that the responseto any therapeutic intervention is constrained by thestate of pathways that can convey signals from the brainto the periphery, diffusion-weighted magnetic resonanceimaging (DW-MRI) was performed to characterize thestructural integrity of the descending corticospinal tract(CST) projections for each participant.MethodsDesignA pilot, triple-blind, randomised case series to explorethe feasibility of combining tRNS with reaching trainingwas conducted between September and December 2014in the Department of Physiotherapy at the University ofQueensland, Brisbane Australia. Ethical approval was re-ceived from the University of Queensland Medical Re-search Ethics Committee (2014000263). All participantsprovided written informed consent to participate and havetheir findings published in accordance with the Declarationof Helsinki. This study was registered on the Australian andNew Zealand Clinical Trials Registry (ANZCTR) http://www.ANZCTR.org.au/ACTRN12614000952640.aspx.ParticipantsFour stroke survivors were recruited using two methods.Firstly, we contacted people in our research group data-base that had consented to be contacted for strokeover the contralateral supra-orbital region. The currentwas set to 2 mA for active-tRNS and 0 mA for sham-tRNS, for 5 s duration and 0 s fade in/out. The tRNSintervener informed participants that they may or maynot feel the stimulation irrespective of whether theywere receiving active or sham stimulation. All tRNS pa-rameters were recorded by the independent tRNS inter-vener in a separate log that was stored in a separatelocked filing cabinet to that containing training and as-sessment information.The delivery of tRNS was timed to coincide with vol-untary motor drive. During reaching training, visual (‘go’on computer screen) and auditory (beep) signals wereused to indicate to the participant that they were tocommence their reaching attempt. The delivery of tRNSwas programmed to commence simultaneously with theonset of these visual and auditory signals, and to ceaseafter 5-s – at which point the voluntary reach movementHayward et al. Journal of NeuroEngineering and Rehabilitation  (2017) 14:46 Page 3 of 9research and resided in Brisbane Australia. Secondly, weposted recruitment flyers with the following sources: a)National Stroke Foundation of Australia webpage; b)Queensland Rehabilitation Physiotherapy Network; andc) physiotherapy and speech pathology clinics at theUniversity of Queensland. Eligible participants were firsttime stroke survivors who were between 6-months and5-years post stroke, aged over 18 years, and presentedwith severe upper limb paresis (as indicated by a tricepsmanual muscle test score of 1+, 2- or 2 out of 5, andMotor Assessment Scale item 6 score of <4 out of a pos-sible 6 points). In addition, all participants were requiredto be able to understand single stage commands, andhave not participated in any upper limb related therapywith a therapy service for at least 2-weeks prior to base-line assessment. Exclusion criteria were: 1) any contraindi-cations to tRNS (e.g., history of seizures) or MRI (e.g.,pacemaker); 2) presence of any neurological conditionother than stroke (e.g., Parkinson’s disease), 3) elbow con-tracture greater than 15°, or 4) consumption of medica-tion/s that could alter cortical excitability (e.g., antiepilepticmedications, antidepressants) or have a presumed positiveor negative effect on neural plasticity (e.g., dopamine, dex-amphetamine). The general practitioner of each participantwas contacted to provide a list of current medications thatwas reviewed by a clinical pharmacist to determine knownor presumed effect on neural plasticity.Blinding and randomizationThis was a triple blind study. Study personnel involvedin assessment and training, along with all participantswere blinded to group allocation for the duration of thestudy. An offsite investigator prepared the concealedrandomization using a computer generated random num-ber sequence (1:1). Groups were: 1) active-tRNS + reachingtraining or 2) sham-tRNS + reaching training. Consecu-tively numbered opaque envelopes containing group allo-cation were collected from the offsite investigator by thetRNS intervener after initial assessment. The blinding codewas shared with the investigator team on completion oftraining and follow-up assessments of all four participants.InterventionParticipants underwent 12 reaching training sessions of45-m duration over 4-weeks (9 h total) in the NeurologicalAgeing and Balance Research Unit at the University ofQueensland. Set up and pack up time was separate fromtraining time. During training, participants were encour-aged to perform as many repetitions as possible withintime allocation. The training set-up and protocol replicatedthat previously established to produce a statistically andclinically meaningful change in upper arm function inchronic stroke survivors [8]. The set-up is visually depictedin Fig. 1. To augment full range reaching and enableindependent practice, reaching training included outcome-triggered electrical stimulation (OT-stim) that was deliv-ered up to the point at which the individual surpassed theirpersonal best reaching distance [9]. The difficulty of train-ing was incrementally progressed to ensure training waschallenging through increased number of repetitions, re-duced rest time, track elevation and addition of load. Allreaching training was recorded in a training log.tRNS was delivered by a battery-driven electricalstimulator (Magstim, UK) through conductive rubberelectrodes, placed in saline soaked sponge sleeves. Forboth active-tRNS and sham-tRNS, the electrodes werepositioned as per the 10/20 international system for C3/C4 EEG electrode placement [17]: the stimulation elec-trode was positioned over the ipsilesional primary motorcortex (M1), and the reference electrode was positionedFig. 1 Representation of the training setup including horizontalreaching track, trunk restraint, visual feedback, transcranial randomnoise stimulation application, and electrical stimulation applicationto lateral head of tricepsattempt was generally still in progress. When required(personal best reaching distance less than passive reachingof personal information and individual experiences, andto facilitate individualized probing. Interviews were ar-ranged by the blinded assessor and conducted by a facili-tator not involved in any aspect of the study, who wasindependent of the research institution and therapy ser-vices the participant had engaged in.Four open-ended stimulus questions (Table 1) under-pinned each interview. In these discussions, the termfunction was used to encapsulate the ICF domains of im-pairment, activity and participation. At the commence-ment of each discussion, the facilitator explained theinterview purpose and then proceeded to ask each of thefour key questions. There was no strict adherence to thestyle and type of questioning beyond these four questions,Hayward et al. Journal of NeuroEngineering and Rehabilitation  (2017) 14:46 Page 4 of 9distance), OT-stim to the impaired lateral head of tricepsbrachii muscle was triggered to augment voluntary move-ment to ensure that full range reaching was achieved.MeasuresParticipants were assessed at baseline (pre-training, 0-weeks), post-training (4-weeks) and follow-up (12-weeks) in the Neurological Ageing and Balance ResearchUnit at the University of Queensland. Personal andstroke-related details were collected and the modifiedRankin Scale performed to describe the severity of dis-ability present at baseline. Adverse events were reportedin the training log (e.g., fall, negative response to tRNSor OT-stim) during training, and the follow-up assess-ment book for post training events (e.g., fall). Trainingoutcomes of sessions and repetitions completed were re-corded in training logs.Clinical outcomesAll clinical measures were obtained at each timepoint.The primary clinical outcome measure was Motor Assess-ment Scale item 6 (Upper Arm Function, MAS6) [18].Secondary outcome measures were MAS item 7 (handactivities) and item 8 (advanced hand activities), alongwith impairment measures of muscle strength of tricepsbrachii and extensor carpi radialis, resistance to passivemovement (Modified Ashworth Scale) and spasticity(Tardieu Scale) of elbow and wrist flexion [19], andshoulder pain on passive external rotation (RitchieArticular Index) [20]. In addition, upper limb partici-pation was evaluated according to the Rating ofEveryday Arm Use in the Community and Home(REACH) scale [21].MRIMRIs were acquired pre and post training at the Centrefor Advanced Imaging at The University of Queenslandon a Siemens Magnetron 3T Trio whole body scannerusing a 32-channel encoding head coil. A high-resolutionT1-weighted anatomical scan (0.45 x 0.45 x 0.9 mm voxelsize) was collected to determine lesion location. A singlehigh-angular resolution diffusion imaging (HARDI)scan was subsequently performed using a single shotecho-planar imaging (EPI) sequence (TR = 9000 ms,TE = 113 ms, FOV = 230 x 230 mm, 55 slices, voxeldimensions = 2.3 x 2.3 x 2.3 mm3). Diffusion weighting wasapplied across 85 independent non-collinear orientations(b = 3000 s/mm2), along with three un-weighted im-ages (b=0 s/mm2). Only pre-training MRIs were evaluated.Post-training interviewIn depth, one-on-one interviews were completed witheach stroke survivor and their main carer (in the case ofaphasic participants) to encourage comfortable sharingwith probes used to explore or challenge emergingthemes, personal experiences and ideas. All discussionswere drawn to a close with the facilitator summarizing themain points raised. The participants were then providedwith the opportunity to add or dispute what had been saidor contribute any final thoughts.AnalysisTo determine feasibility, we considered adverse events,training outcomes, clinical outcomes, structural integrityof descending motor projections, and reflections on train-ing through in-depth interviews from each individual case.Adverse eventsNumber of adverse events recorded during training (e.g.,complaints of pain, discomfort) and at follow up assess-ment (e.g., falls).Training outcomesNumber of sessions attended and missed were talliedand average repetitions per session (total repetitions/number of sessions) were calculated for each individual.Table 1 Primary question(s) within each category of questingfor in-depth interviews(1) Understanding of upper limb rehabilitation processes:“Prior to commencing this study what was your understanding of upperlimb therapy? How does this differ now that you’ve undertaken thisresearch project of reaching training?”(2) Reaching training:“Thinking about the reaching training, can you tell me how you feltdoing the training?”(3) Problems as well as rewarding situations during your reaching training:“Was there anything that stopped you from wanting to/helped youcontinue with the research project? If so, can you tell me about this?”(4) Advice you might have about upper limb rehabilitation:“Do you think that this type of training helped your arm recover? Why/why not? What would you say to someone about to commence thistype of reaching training? What would you want them to know?”Hayward et al. Journal of NeuroEngineering and Rehabilitation  (2017) 14:46 Page 5 of 9Clinical outcomesChange from pre to post intervention for the primaryand secondary outcome measures were calculated foreach individual. A clinically meaningful change was con-sidered to be a 10%, or a 1-point change on MAS6. Thisis consistent with previous work conducted by our group[9, 22] and others [23].Structural integrity of descending motor projectionsThe high resolution T1 anatomical scans acquired in the‘pre’ sessions were used for transformation of the diffusionweighted imaging (DWI) data in ExploreDTI (Leemanset al 2009) so that both image modalities (displaying whitematter and grey matter together) could be co-registeredand superimposed to facilitate identification of anatomicalstructures in the brain. DWI data were first visuallyinspected for excessive motion artifact or instrumentalnoise using quality assurance tools available in the diffu-sion MRI software package ExploreDTI v.4.8.4 [24]. Forall images, signal intensity was modulated and the b-matrix rotated [25]. Imaging data were then corrected formotion and distortion. As there were only four partici-pants, it was possible to use a manual ‘region of interest’identification procedure, whereby the white matter tractsof the CST in the posterior limb of the internal capsulewere extracted by hand-drawing around the anatomicalregion on the superimposed DWI-T1 native images. Con-strained spherical deconvolution (CSD) was used to modeldiffusion behavior [26] as it is robust in the presence ofcrossing fibre populations [27]. Crossing fibres are esti-mated to occur in greater than 90% of white matter voxelsin the brain [28]. Additionally, CSD does not make as-sumptions regarding uniform diffusion of water within avoxel [26, 29] and is more sensitive in the severely dam-aged brain [27]. CSD-based deterministic whole-brainfibre tractography was initiated at each voxel using the fol-lowing parameters: seed-point resolution of 2 mm3,0.2 mm step size, maximum turning angle of <40°, andfibre length range of 50–500 mm [30]. Tractographyemployed a fibre alignment by continuous tracking algo-rithm approach [31] with Fractional Anisotropy (FA)values extracted from reconstructed streamlines. Frac-tional Anisotropy is a quantitative, unit-less measure ofdiffusion behaviour of water in the brain influenced bymicrostructural properties of white matter and is the mostcommonly reported measure of white matter microstruc-tural properties after stroke [32]. Having extracted CSTFA for both the lesioned and non-lesioned hemispheresfor each participant, a FA ‘asymmetry index’ (AI) was cal-culated according to Eq. 1. This index quantifies the de-gree of degeneration of white matter in the tracts that areresponsible for conveying motor cortical commands tothe muscles of the upper limb. For our purposes, an asym-metry index of 0.01-0.05 was considered mild, 0.06-0.15was considered moderate, and >0.15 was considered se-vere. As the AI increases, greater loss of white matterstructural integrity as a result of the stroke can beinferred.AI ¼ FA non‐lesioned PLIC – FA lesioned PLICð ÞFA non‐lesioned PLIC þ FA lesioned PLICð Þð1ÞReflections on trainingAll audio recordings of the in-depth interviews weretranscribed verbatim and cross-checked by another re-searcher against the audio record to verify accuracy. Anapproach consistent with conventional thematic contentanalysis was used [33]. On completion of the study, thetranscripts of each participant were explored independ-ently through a process of reading and re-reading. Tworesearchers, one of whom was involved in training (KH)and one who was not involved in any data collectionprocedures (SB), independently reviewed all transcripts.On the first reading, transcripts were read in their entir-ety to acquire a whole sense of the data. On the secondreading, line-by-line analysis was used to identifythemes, patterns or concepts. This led to the tentativecollation of predominant themes emerging across partic-ipants. The two reviewers met at this point and dis-cussed their themes, looking for patterns or conceptsthat were both consistent and inconsistent with eachother. Consensus themes were identified. A final readingof the data was used to check the fit of the consensusthemes with the transcripts, pursuing patterns or con-cepts that were both consistent and inconsistent withthe data. A second meeting of the reviewers occurred toconfirm the themes, or modify them as required to moreappropriately represent the data. At this point, the con-ditions under which each theme arose and its relation-ship to other themes (within and between groups) weredocumented. All findings are anonymised.ResultsFour participants were recruited, with maximum vari-ation in sample achieved (See Table 2), Each participantwas treated as a single case.Adverse eventsThere were no adverse events recorded during thecourse of training, and no adverse events were reportedto have occurred during follow up.Training outcomesAll four participants completed the 12 training sessionsover 4-weeks, with no training sessions missed. Partici-pants completed on average 117 reaching repetitions perTable 2 Participant characteristicsID Age at training Stroke type MSS Dominant arm Paretic arm Aphasia Baseline mobility, FAC,/6 mRS,/5P1 49 Ischaemic 24 Right Left No 6 3P2 53 Ischaemic 32 Right Right Yes 5 3P3 73 Ischaemic 25 Right Left No 1 5P4 70 Ischaemic 37 Right Right Yes 4 3FAC Functional Ambulation Category, MSS months since stroke, mRS modified Rankin ScaleHayward et al. Journal of NeuroEngineering and Rehabilitation  (2017) 14:46 Page 6 of 9session. Participant 1 completed the most repetitions persession on average (n = 147), while participant 3 completedthe fewest repetitions on average per session (n = 85).Clinical outcomesParticipants (P01 and P02) demonstrated an improvementin triceps muscle strength (impairment), and REACHScale (participation) post training, which was maintainedat follow-up. P02 also demonstrated an improvement inMAS6 (activity) post training, which was lost at follow up.P03 demonstrated an improvement in MAS6 (activity),which was maintained at follow up. P04 demonstrated animprovement in triceps muscle strength (impairment)post training that was maintained at follow up. Interest-ingly, both P03 and P04 demonstrated no change in use ofthe arm in everyday tasks (i.e., participation) post training,however they did demonstrate an improvement in thismeasure at follow up. See Table 3.Structural integrity of descending motor projectionsParticipant P01 (0.02) had a mild AI, P02 (0.11) had amoderate AI, and P03 and P04 had a severe AI (SeeTable 3). This suggests that P01 and P02 had residual CSTreserve, and thus perhaps some potential for UL recovery,but were yet to realize this potential. In contrast, P03 andP04 had such extensive structural loss of white matter inthe CST, that no streamlines could be reconstructed inthis region of the ipsilesional hemisphere. Visualizationsof the CST for each individual are displayed in Fig. 2.Reflections on trainingOverall participants and their carers (where involved)spoke positively about the training program with activeor sham tRNS, describing it as feasible and beneficial.Table 3 Training, clinical and descending motor projection outcomID tRNSgroupTotalrepetitionsWrist MRC/15 Triceps MRC/T0 T1 T2 T0 T1P1 Active 1763 2 3 3 10 12P2 Placebo 1128 2 2 2 3 6P3 Active 1015 0 0 1 3 3P4 Placebo 1696 1 1 2 3 6MAS6 Motor Assessment Scale Item 6 Upper Arm Function, MRC Medical Research Cnot possible, REACH Rating Everyday Arm use in the Community and Home, tRNS tr(4-weeks), T2 follow up (12-weeks)No participants described any adverse effects of the stimu-lation, nor was it perceived to interfere with engagementin training. One participant even described it as “trying tosay hey come on, let’s keep moving, P01”. The only negativeof training, described by all, was that training wasending – expressing that they wanted to be able to con-tinue to participate due to the gains experienced. “Nowthat it has stopped … we’d like it to keep going … if youcould extend it, it would be even more valuable, P04carer”. Beyond these general comments, the content ofthe interviews could be ascribed to three themes. Strokesurvivors and their carers described the individualisedtraining helped them to maintain motivation and fos-tered the overwhelming sense that I haven’t lost allmy chances of improving my arm. Each theme will bediscussed in turn.Individualised training was described as a positiveaspect of this combination therapy approach. Partici-pants described it as not just a set program, but ratheran individualised training package that included coach-ing, education and meaningful repetition. The coachingskills of the trainer were described as critical to trainingengagement. Coaching included tailoring training in away that “forces you to move and achieve a goal, P02”. Inthis way it was thought that a better quality of move-ment was achieved. While people felt they could dotraining alone, “what really helped was having someonebeside me, pushing you on and motivating you, P01”. Astrong component of individualization expressed by par-ticipants was education, described as “direction andguidance on what to do and how to do it, P02 carer”.Many commented that previously they did not receivesuch guidance, which made it challenging to maintainconcentration during training. Engagement in meaningfules15 MAS6,/6 REACH,/5 AsymmetryindexT2 T0 T1 T2 T0 T1 T212 1 1 1 1 2 2 0.027 2 3 2 0 1 1 0.112 0 1 1 0 0 1 NP6 1 1 1 0 0 1 NPouncil strength grading including + and – to achieve a possible 15 points, NPanscranial random noise stimulation; T0 baseline (0-weeks), T1 post trainingt isnistreortC rHayward et al. Journal of NeuroEngineering and Rehabilitation  (2017) 14:46 Page 7 of 9repetition performed within a combination-training pro-gram was viewed positively. “This training with the stimu-lation and the movement and actually showing us how itis happening, and then doing it with repetition. It clicked,P02 carer” and “I knew I was doing a defined task … andwas trying to improve each time … it was good, P01”.Individuals described that maintaining motivationwas important throughout training. All described thatduring the first week of training, little gains were experi-enced and recovery was slow. Throughout this periodhope that training might trigger a change in one’s recov-ery pathway was important. “It’s only going to happen ifyou keep working on it, P03” and “she has the determin-ation to concentrate on the arm, P04 carer”. Maintainingmotivation was challenged by not being able to see theachievements that were happening. “It can be frustrating.We like to see everything visually represented as achieve-ment. But some of the achievements that are happeningare quietly occurring in our brain so when the actualproduct will happen might take a few weeks. I think pa-tience is very important, P02 carer”. But with patienceand the ability to just hang in there, training was de-Fig. 2 Corticospinal tract streamline reconstructions: the corticospinal tracslices of T1 weighted anatomical scans with direction encoded fractional aformat (ie. right on the image is the patient’s left side). The reconstructed sby red circles. The posterior limb of the internal capsule (PLIC) within the ceach scan, using anatomical landmarks. No tracts were detected in the PLIscribed to eventually “just sort of click, P01”. Individualsdescribed taking ownership of performance-relatedchanges within a session, small or large, to positivelyreinforce engagement in use of the arm outside oftherapy – an important shift in mindset that contributedto maintained motivation.It was overwhelmingly evident that a major benefit ofthe training was realizing that “I (stroke survivor)haven’t lost all my chances of improving my arm”.Many described recovery of the arm early after stroke as“forgotten and neglected, P01”. Realising that engagementin arm training was possible, many described a renewedhope for recovery. Statements such as “made me feel likethere is hope there for moving my (his) arm, P01 and P03carer” were common. Carers described a sense of relief,that the hope they had sustained in seeing an improve-ment in their loved ones arm function was not hopeless.“We believed there was a way to do it, but just didn'tknow what the way was, so it was like oh yes, it is pos-sible, P02 carer”.DiscussionThis study demonstrated that gains in relation to bothimpairment and function are possible in the chronicphase of recovery in people with severe upper limb im-pairment – even in those with limited neurobiologicalstructural reserve for recovery. While all participants im-proved, the nature of the improvement was specific to eachindividual. Despite demonstrating that tRNS timed to coin-cide with the generation of the voluntary motor commandsduring reaching training was feasible, it was not evident inthis very small sample, that there was a benefit over andabove reaching training alone (sham tRNS). Given all par-ticipants enjoyed the training, perceived it to be beneficialand wanted it to continue, there is impetus to explore thistraining paradigm more extensively.Few studies to date have sought to document the con-dition of the motor system prior to engaging individuals(with severe impairment) in therapeutic training. Withoutindicated for each of the four participants, displayed on coronal (x view)otropy (FA) colour maps superimposed. Images are shown in radiologicalamlines for the corticospinal tract are also superimposed, and indicatedicospinal tract was the region of interest that was delineated manually foregion in the right hemisphere for P03, or the left hemisphere for P04first defining the state of pathways that convey signalsfrom the brain – operationalized in the present instanceas the structural integrity of the corticospinal tract – it ischallenging to determine the extent to which this acts as aconstraint on the gains that can be achieved as a result ofan intervention. In the present study all four individualshad clinically severe upper limb impairment and demon-strated treatment related gains. Yet, they variously exhib-ited mild, moderate, and severe degeneration of the whitematter tracts that pass through the posterior limb of theinternal capsule. It appears therefore that in this study, thegains realised through training were not manifestly con-strained by the integrity of corticospinal projections to theaffected limb. This is consistent with a previous studyusing a similar training intervention [12], whereby someindividuals who failed to exhibit a MEP at the outset,nonetheless accrued benefit from training. In that case,Hayward et al. Journal of NeuroEngineering and Rehabilitation  (2017) 14:46 Page 8 of 9the functional integrity of the corticospinal tract wasinferred from the presence or absence of MEPs invokedby TMS in a principal agonist (triceps brachii). Taken to-gether, these outcomes suggest that for chronic stroke sur-vivors who present with severe impairment of motorfunction, the integrity of corticospinal projections – at leastas contemporary DWI and TMS methodologies assessthem, does not exert a determining influence on the gainsthat can be realised through repetitive reach training.The present study is unique in the field of stroke re-habilitation as it combines assistive therapy, with periph-eral and cortical stimulation techniques. Two key aspectsseparate our work from that conducted previously: 1) cor-tical stimulation was timed to coincide with the prepar-ation and production of active movement, and this wasfollowed up with 2) peripheral stimulation that augmentedvoluntary motor output to enable task completion. It re-mains to be seen however, if this paradigm can enhancerecovery over and above that achieved with a particularintervention in isolation. This is a common challengefor stroke rehabilitation and recovery research, par-ticularly in relation to the application of NIBS [34].Nonetheless, the approach we have described appearsfeasible and was also received favourably by thestroke survivors and carers involved.Strengths and limitationsThis study, implemented through a triple-blind design,adopted strict inclusion criteria. We confined the periodpost stroke to between 6 months and 5 years, and didnot include people taking medications that may have thepotential to influence cortical activity. While this made re-cruitment challenging, it partially mitigated the potentialinfluence of confounding factors. Necessarily however,these may remain influential when a small cohort is in-volved. Additionally, we established through in-depth,one-on-one interviews with each participant, a deep un-derstanding of each stroke survivor’s perceptions of thetraining program.There are however, some limitations to consider.While the attempt was made to couple the delivery ofNIBS to the progression of each voluntary movement, itwas not possible to time this precisely for each reach at-tempt. Specifically, the onset of tRNS was not yoked dir-ectly to the onset of electromyography – for example. Inaddition, the offset of tRNS was determined by a fixed 5-sinterval, rather than triggered by a defined feature of theevolving movement kinematics. In future investigationsthe timing of tRNS delivery could be linked more directlyto the preparation or execution phases of each movementattempt. The sample size employed in the present studywas also rather limited. In order to more effectively ascer-tain the relationships that may exist between the structuralintegrity of motor output pathways and the potential forrecovery of upper limb function in severely impairedstroke survivors, a larger sample – representative of thetarget population, would be preferred.ConclusionThis study highlights that combined interventions that ex-ploit motor learning principles, enable repetitive practice,and seek to enhance cortical drive are feasible and em-braced enthusiastically by stroke survivors. Supporting ac-tive engagement in movement training in even the mostsevere stroke survivors has direct benefits for the strokesurvivor in terms of enhancing motor recovery and main-taining hope. It also has a positive impact upon carers, whooften play a critical role in encouraging and promoting theuse of such interventions within the home and community.AbbreviationsAI: Asymmetry index; CST: Corticospinal tract; FA: Fractional anisotropy;M1: Primary motor cortex; MAS6: Motor activity scale item 6 upper arm function;MRI: Magnetic resonance imaging; NIBS: Non-invasive brain stimulation; OT-stim: Outcome-triggered electrical stimulation; QoL: Quality of life; REACH: Ratingof everyday arm use in the community and home; TMS: Transcranial magneticstimulation; tRNS: Transcranial random noise stimulationAcknowledgementsKatrina Kemp completed all blinded clinical assessments; Elizabeth Hayward,registered clinical pharmacist, completed all medication history checks; EthanMarrinan and Jia Ling Tan were blinded tRNS applicators; Sarah Fitzhenrycompleted all indepth interviews; and Anna Hatton completed therandomisation sequence and stored the envelopes offsite.FundingThis study was funded by a small project grant from the National StrokeFoundation of Australia.Availability of data and materialsFrom the corresponding author on contact. Individual data are contained inTable 2 and 3.Authors’ contributionsKSH, RGC, SGB, KLR designed the study and were awarded funding; KSHdelivered all training, wrote the manuscript, analysed clinical and qualitativedata; KLR analysed all neuroimaging data and edited the manuscript; SGBanalysed qualitative data and edited the manuscript; DL developed thetraining software package and edited the manuscript; RGC designed thetRNS training protocol and edited the manuscript. All authors read andapproved the final manuscript.Competing interestsAuthors KSH, RGC, SGB and DL advise SMART Arm Pty Ltd on researchassociated with the SMART Arm.Ethics approval and consent to participateWas received from the University of Queensland Medical Research EthicsCommittee (2014000263). All participants provided written informed consentin accordance with the Declaration of Helsinki.Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims in publishedmaps and institutional affiliations.Author details1Division of Physiotherapy, School of Health and Rehabilitation Sciences,University of Queensland, Brisbane, Australia. 2Department of PhysicalTherapy, University of British Columbia, Vancouver, Canada. 3Department ofHealth Sciences and Technology, Neural Control of Movement Lab, ETHorientation distribution in diffusion MRI: non-negativity constrained super-resolved spherical deconvolution. Neuroimage. 2007;35:1459–72.27. Auriat AM, Borich MR, Snow NJ, Wadden KP, Boyd LA. Comparing adiffusion tensor and non-tensor approach to white matter fibertractography in chronic stroke. Neuroimage Clin. 2015;7:771–81.28. Jeurissen B, Leemans A, Tournier JD, Jones DK, Sijbers J. Investigating theprevalence of complex fiber configurations in white matter tissue withdiffusion magnetic resonance imaging. Hum Brain Mapp. 2013;34:2747–66.29. Farquharson S, Tournier JD, Calamante F, Fabinyi G, Schneider-Kolsky M,Jackson GD, Connelly A. White matter fiber tractography: why we need tomove beyond DTI. J Neurosurg. 2013;118:1367–77.30. Reijmer YD, Leemans A, Heringa SM, Wielaard I, Jeurissen B, Koek HL,Biessels GJ, Vasc Cognitive Impairment Study G. Improved sensitivity tocerebral white matter abnormalities in Alzheimer's disease with sphericaldeconvolution based tractography. PLoS ONE. 2012;7(8):e44074.31. Mori S, Crain BJ, Chacko VP, van Zijl PC. Three-dimensional tracking ofHayward et al. Journal of NeuroEngineering and Rehabilitation  (2017) 14:46 Page 9 of 9Zurich, Zurich, Switzerland. 4Queensland Brain Institute, University ofQueensland, Brisbane, Australia. 5Trinity College Institute of Neuroscience andSchool of Psychology, Trinity College, Dublin, Ireland. 6School of Psychology,Queens University Belfast, Belfast, UK.Received: 27 March 2017 Accepted: 15 May 2017References1. Houwink A, Nijland R, Guerts AC, Kwakkel G. Functional recovery of theparetic upper limb after stroke: Who regains hand capacity? Arch Phys MedRehabil. 2013;94:839–44.2. Hayward KS, Schmidt J, Lohse KR, Peters S, Bernhardt J, Lannin NA, Boyd LA.Are we armed with the right data? Pooled individual data review ofbiomarkers in people with severe upper limb impairment after stroke.NeuroImage Clin. 2017;13:310–9.3. Beebe JA, Lang CE. Active range of motion predicts upper extremityfunction 3 months after stroke. Stroke. 2009;40:1772–9.4. Hayward KS, Aitken PA, Barker RN, Brauer SG. Admission to andcontinuation of inpatient stroke rehabilitation in Queensland Australia: Asurvey of factors that contribute to the consultants decision. Brain Imp.2014;15:88–98.5. Edwards JD, Koehoorn M, Boyd LA, Levy AR. Is health-related quality of lifeimproving after stroke? a comparison of health utilities indices amongCanadians with stroke between 1996 and 2005. Stroke. 2010;41:996–1000.6. Hayward KS, Barker RN, Brauer SG. Interventions to promote upper limbrecovery in stroke survivors with severe paresis: a systematic review. DisabilRehabil. 2010;32:1973–86.7. Reinkensmeyer DJ, Burdet E, Casadio M, Krakauer JW, Kwakkel G, Lang CE,Swinnen SP, Ward NS, Schweighofer N. Computational neurorehabilitation:modeling plasticity and learning to predict recovery. J Neuroeng Rehabil.2016;13:42.8. Barker RN, Brauer SG, Carson RG. Training of reaching in stroke survivorswith severe and chronic upper limb paresis: a randomised clinical trial.Stroke. 2008;39:1800–7.9. Hayward KS, Barker RN, Lloyd D, Brauer SG, Horsley SA, Carson RG. SMARTArm with outcome-triggered electrical stimulation: A pilot RCT. Top StrokeRehabil. 2013;20:289–98.10. Brauer SG, Hayward KS, Carson RG, Cresswell AC, Barker RN. The efficacy ofSMART Arm training early after stroke for stroke survivors with severe upperlimb disability: A protocol for a randomised controlled trial. BMC Neurol.2013;13:71.11. Barker RN, Brauer SG, Carson RG. Training-induced changes in the pattern oftriceps to biceps activation during reaching tasks after chronic and severestroke. Exp Brain Res. 2009;196:483–96.12. Barker RN, Brauer SG, Barry BK, Gill TJ, Carson RG. Training-inducedmodifications of corticospinal reactivity in severely affected stroke survivors.Exp Brain Res. 2012;221:211.13. Wessel MJ, Zimerman M, Hummel FC. Non-invasive brain stimulation: aninterventional tool for enhancing behavioral training after stroke. FrontiersHuman Neurosci. 2015;9:265.14. Bradnam LV, Stinear CM, Barber PA, Byblow WD. Contralesional hemispherecontrol of the proximal paretic upper limb following stroke. Cereb Cortex.2012;22:2662–71.15. Moss F, Wolf LM, Sannita WG. Stochastic resonance and sensory informationprocessing: A tutorial and review of application. Clin Neurophysiol.2004;115:267–81.16. Miniussi C, Harris JA, Ruzzoli M. Modelling non-invasive brain stimulation incognitive neuroscience. Neurosci Behav Rev. 2013;37:1702–12.17. American Electroencephalographic Society. Guideline thirteen: Guidelines forstandard electrode position nomenclature. J Clin Neurophysiol. 1994;11:111–3.18. Carr JH, Shepherd RB, Nordholm L, Lynne D. Investigation of a new motorassessment scale for stroke patients. Phys Ther. 1985;65:175–80.19. Morris S. Ashworth and Tardieu Scales: Their clinical relevance for measuringspasticity in adult and paediatric neurological populations. Phys Ther Rev.2002;7:53–62.20. Bohannon RW, Lefort A. Hemiplegic shoulder pain measured with theRitchie Articular Index. Int J Rehabil Res. 1986;9:379–81.21. Simpson LA, Eng JJ, Backman CL, Miller WC. Rating of everyday Arm-use inthe community and home (REACH) scale for capturing affected Arm-useafter stroke: development, reliability, and validity. PLoS ONE. 2013;8:e83405.axonal projections in the brain by magnetic resonance imaging. AnnNeurol. 1999;45:265–9.32. Jang SH, Ahn SH, Sakong J, Byun WM, Choi BY, Chang CH, Bai D, Son SM.Comparison of TMS and DTT for predicting motor outcome in intracerebralhemorrhage. J Neurol Sci. 2010;290:107–11.33. Green J, Thorogood N. Qualitative methods for health research. 2nd ed.London: SAGE Publications Ltd; 2009.34. Elsner B, Kugler J, Pohl M, Mehrholz J. Transcranial direct current stimulation (tDCS)for improving activities of daily living, and physical and cognitive functioning, inpeople after stroke. Cochrane Database Syst Rev. 2016;13:CD009645.•  We accept pre-submission inquiries •  Our selector tool helps you to find the most relevant journal•  We provide round the clock customer support •  Convenient online submission•  Thorough peer review•  Inclusion in PubMed and all major indexing services Submit your next manuscript to BioMed Central and we will help you at every step:22. Hayward KS, Kuys SS, Barker RN, Brauer SG. Can stroke survivors with severeupper arm disability achieve a clinically important change in arm functionduring inpatient rehabilitation? A multicentre, prospective, observationalstudy. Neuro Rehabil. 2014;35:773–9.23. Van der Lee JH, Wagenaar RC, Lankhorst GJ, Vogelaar TW, Deville WL,Bouter LM. Forced use of the upper extremity in chronic stroke patient:Results from a single-blind randomised clinical trial. Stroke. 1999;30:2369–75.24. Leemans A, Jeurissen B, Sijbers J, Jones DK. ExploreDTI: a graphical toolboxfor processing, analyzing and visualizing diffusion MR data. Proc 17thAnnual Meeting Int Soc Magn Reson Med. 2009;209:3537.25. Leemans A, Jones DK. The B-matrix must be rotated when correcting forsubject motion in DTI data. Magn Reson Med. 2009;61:1336–49.26. Tournier JD, Calamante F, Connelly A. Robust determination of the fibre•  Maximum visibility for your researchSubmit your manuscript atwww.biomedcentral.com/submit


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items