Upper Limb Motor Improvement after Traumatic Brain Injury: Systematic Review of Interventions

Abstract

Background

Traumatic brain injury (TBI) is a leading cause of adult morbidity and mortality. Individuals with TBI have impairments in both cognitive and motor domains. Motor improvements post-TBI are attributable to adaptive neuroplasticity and motor learning. Majority of the studies focus on remediation of balance and mobility issues. There is limited understanding on the use of interventions for upper limb (UL) motor improvements in this population.

Objective

We examined the evidence regarding the effectiveness of different interventions to augment UL motor improvement after a TBI.

Methods

We systematically examined the evidence published in English from 1990–2020. The modified Downs and Black checklist helped assess study quality (total score: 28). Studies were classified as excellent: 24–28, good: 19–23, fair: 14–18, and poor: ≤13 in quality. Effect sizes helped quantify intervention effectiveness.

Results

Twenty-three studies were retrieved. Study quality was excellent (n = 1), good (n = 5) or fair (n = 17). Interventions used included strategies to decrease muscle tone (n = 6), constraint induced movement therapy (n = 4), virtual reality gaming (n = 5), non-invasive stimulation (n = 3), arm motor ability training (n = 1), stem cell transplant (n = 1), task-oriented training (n = 2), and feedback provision (n = 1). Motor impairment outcomes included Fugl-Meyer Assessment, Modified Ashworth Scale, and kinematic outcomes (error and movement straightness). Activity limitation outcomes included Wolf Motor Function Test and Motor Activity Log (MAL). Effect sizes for majority of the interventions ranged from medium (.5-.79) to large (≥.8). Only ten studies included retention testing.

Conclusion

There is preliminary evidence that using some interventions may enhance UL motor improvement after a TBI. Answers to emergent questions can help select the most appropriate interventions in this population.

Keywords

arm head injury rehabilitation outcomes spasticity virtual reality

Introduction

Traumatic brain injury is a major worldwide cause of morbidity and mortality.¹ In the USA, recent reports indicate 2.87 million TBI-related visits to the emergency room,² with epidemiological data suggesting males being more affected than females.³ Common causes of TBI include falls, motor vehicle accidents, and/or assaults.⁴ The available total annual cost estimates for TBI range from $56–$221 billion.⁵ Individuals sustaining a TBI may face cognitive,⁶ behavioral,⁷ and communication difficulties⁸ lasting from few days post-injury to the rest of their lives.⁹ Additionally, a TBI causes sensorimotor impairments to the upper (UL) and lower limbs (LL).

Motor impairments include abnormal posture, altered muscle tone, paresis, reappearance of primitive reflexes, ataxia, decreased balance, and lack of coordinated movement.¹⁰ Individuals continue to have limited performance of activities of daily living (ADL), especially those relying on coordinated movements and UL muscle strength after a TBI. Persistent UL impairments and limitations in ADL performance impact functional independence in this population.¹¹

Motor improvements post-TBI are attributable in part to motor learning and adaptive neuroplasticity.¹² Provision of rehabilitation, benefits motor recovery by focusing on performing accurate repetitions of desired movement,¹³ is an integral part of motor learning and promotes adaptive neuroplasticity.¹⁴ Recent guidelines stress application of task-specific and intensive repetitive practice of functional reaching and activities including fine motor co-ordination.¹⁵

There is a need to identify the most effective and pertinent interventions with a focus on remediation of impairments and activity limitations in this population.¹⁶ To date, research has focused primarily on cognitive impairments and gait limitations, with less focus on UL issues.^17,18 This is an important topic, given that the UL issues are more diffuse and tend to be long standing in individuals post-TBI.¹⁹ Previous studies have identified deficits in UL functioning including impaired timing, reach accuracy and grasping ability.²⁰ Improving UL motor functioning helps boost the ability of individuals with TBI to perform ADL such as dress, wash clothes, cook, and groom.²¹ Enhanced UL functioning also enables better community reintegration post-TBI. For example, improving the ability to drive helps commute to work and enables the ability to be competitively employed, volunteer and/or attend school.^22,23

Our study objective was to systematically review the available literature focusing on rehabilitation of the UL, in individuals sustaining a TBI. Better identification of useful interventions can help select the best options to be used in the clinic and contribute to evidence-based practice. Our question in the Population, Intervention, Comparison and Outcome (PICO)²⁴ format was, “In individuals sustaining a TBI, does provision of rehabilitation interventions augment UL motor improvement post-intervention compared to pre-intervention?” Preliminary results have previously appeared as an abstract.²⁵

Methods

Systematic Literature Review

We performed a systematic search of the literature using Medline, Google Scholar, ISI Web of Science, Science Direct, and CINAHL. A Health Sciences Library Liaison helped formulate appropriate search strategies. Keywords and MeSH terms used included “traumatic brain injury,” “head injury,” “concussion,” “arm,” “upper limb,” “upper extremity,” “rehabilitation,” “intervention,” “motor recovery,” “impairment,” “activity limitation” and “motor improvement.” We used the terms “AND” and “OR” to combine keywords. Searches involved additional limits to restrict the articles to the English language literature published from January 1990 through August 2020, human species, and adult participants. Inclusion criteria were (i) exposure to or provision of rehabilitation interventions and (ii) assessment of motor impairment and/or limitations in ADL using the UL. Exclusion criteria were (i) studies focusing on effects of provision of only cognitive rehabilitation; (ii) rehabilitation focusing exclusively on LL outcomes; or (iii) review articles, single case studies, and expert opinion articles. We reviewed the reference lists of retrieved studies to identify additional relevant citations. We also checked the excluded reviews to identify any pertinent citations.

Data Abstraction and Analysis

Retrieved articles were grouped according to the intervention used. We developed a data abstraction form to extract data from the selected articles. Data were initially extracted by MKF and AFH. The first author (SKS) then verified that all relevant data were obtained from the selected articles. The extracted data included details about chronicity, type of UL intervention, outcome of intervention, and results of the study.

We quantified the effectiveness of interventions using estimates of effect sizes.²⁶ When pre, post and retentions scores were available, effect sizes were calculated as the mean post-pre/SD_pre values or mean retention - pre/SD_pre values. In case only change scores were reported, we used the ratio of the mean change score to the variability in change scores. Effect sizes ranging from .2–.49, .5–.79, and ≥.8 were interpreted as small, medium, and large, respectively.²⁷ We assessed the quality of the selected articles using the modified version²⁸ of the Downs and Black²⁹ checklist.

The Downs and Black checklist is a reliable and valid assessment.³⁰ It can be used to assess the quality of both randomized and non-randomized study designs. The total scores of this assessment and PEDro scale are highly correlated in studies involving individuals with brain injuries.^31,32 Scores on the modified Downs and Black checklist were rated as “excellent” (score 24–28), “good” (score 19–23), “fair” (score 14–18), or “poor” (score ≤13).³³ The quality of each study was independently evaluated by AFH and MKF, with discrepancies, if any, resolved by SKS.

Results

A total of 140 citations were identified through database searches (Figure 1). After removing duplicates, 120 citations were screened, of which 90 were excluded. Of the 30 full text articles assessed for eligibility, we excluded seven studies, as they were reviews and/or expert opinions. Twenty-three articles were included in the qualitative synthesis. The different interventions used included those to reduce muscle tone (n = 6), constraint induced therapy (n = 4), virtual reality–based gaming (n = 5), non-invasive stimulation (n = 3) [including neuromuscular electrical stimulation (n = 1) and transcranial direct current stimulation (n = 2)], arm motor ability training (n = 1), use of stem cells (n = 1), goal oriented task-specific practice (n = 1), feedback provision (n = 1), and forced use therapy (n = 1). The average (95% CI) age of participants was 36.4 (29.1 to 43.6) years. Brief highlights of the studies are presented below, with details in the accompanying tables. The scoring for the modified D&B checklist for each individual study is available in Supplementary Table 1.

Figure 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram.

Interventions to Reduce Muscle Tone

We found six studies (quality ranging from fair to good) that examined the effects of different interventions on UL muscle tone (Table 1). Studies investigated the effects of different interventions on muscle tone reduction including provision of Botulinum toxin A injections, oral medication, serial casting soft splinting, and acupuncture.

Table 1.

Interventions to reduce muscle tone in the upper limbs.

Study; Sample size (n) and Down’s and Black score	Chronicity and severity of injury	Intervention	Rehabilitation provided/Dose	Outcomes	Timing of assessment	Results
Yablon et al³⁴; n = 21 DBS 17 (fair)	• 9 acute and 12 chronic • Majority had severe injuries (GCS ≤8)	20–40 units of botulinum toxin A injected under EMG guidance	• ROM therapy, casting, and/or modalities provided as required	Modified Ashworth’s scale for wrist flexors and passive ROM using goniometry	Baseline and 2–4 weeks after injections	Acute TBI • Decrease in tone measured by modified Ashworth scale; ES = −2.83 • All participants had a lower MAS score (1–2 points), which is MCID • Improvement in wrist extension ROM; ES = 1.52 Chronic TBI • Decrease in tone measured by Modified Ashworth Scale; ES = −1.63 • 11 out of 12 participants had a lower MAS score (1–2 points), which is MCID • Improvement in wrist extension ROM; ES = 1.74
Pavesi et al³⁵; n = 6 DBS 14 (fair)	• Spasticity present for 4–6 months post-injury • Severe injuries (GCS ≤8)	20-40 units of Botox; injected under EMG guidance	• After injections, casting was provided	Modified Ashworth’s scale for wrist flexors and passive ROM using goniometry	Baseline and 4 weeks after injection	• Decrease in tone measured by Modified Ashworth Scale; ES = −2.38 • All participants had a lower MAS score (1–2 points), which is MCID • Improvement in wrist extension ROM; ES = 2.11
Meythaler et al³⁶; n = 17 (8 TBI) DBS 22 (good)	• Spasticity present for atleast 6 mos. before study participation • Initial injury severity level information missing	Oral tizanidine for a maximum of 36 mg/d or dose tolerated for 8 weeks followed by placebo or vice versa	• No information provided	Muscle tone Combined original Ashworth’s scale score for shoulder abductors, elbow muscles and wrist extensors Activity limitations Functional independence measure (FIM) and Craig Hospital assessment and reporting Technique	Baseline and 4 weeks after start of medication Retention assessment only when active drug administered	• Greater reduction in tone with tizanidine (ES = −.36) compared to placebo (ES = −.23; P < .05) • Effect not retained • No improvements noted in activity limitations
Moseley et al 2006; n = 26 DBS 22 (good)	• Duration since injury ≤6 months • Severe injuries (GCS ≤5)	Serial casting (n=14) or positioning (n = 12)	Serial casting group • Elbows stretched in an extended position for 14 days • Progressive increase of stretch range after first 7 days Positioning group • Passive stretch applied for 1 h/day; 5-7 days/week • Stretch position maintained using sandbags, slings, or splints • Both groups also received additional therapy designed to improve individual motor skill (15 min/day, 5 days/week)	Primary Torque controlled elbow extension Secondary Tone assessed using Modified Tardieu Scale; Function assessed suing the Test Évaluant la Performance des Membres Supérieurs des Personnes Âgées (TEMPA)	Baseline, immediately after cast removal, and at one month retention An additional assessment conducted one day after cast removal for the primary outcome	Primary Outcome • Improved elbow extension range by 22° after serial casting for 2 weeks; ES = 1.31 • One day after cast removal, gain in elbow range decreased to 15°; ES = 1.17 • Gain of 11° elbow extension range maintained 1 day after removal of stretch; ES = .29 • Change was not maintained at retention Secondary Outcomes • Tone reduced immediately after serial casting compared to positioning; ES = −1.19 • Reduction in tone not maintained at retention • Both groups improved on TEMPA; no between group differences seen
Thibaut et al³⁸; n = 17 (7 TBI) DBS 20 (good)	• Duration since injury ≥3 months • Severe injuries – 6 participants minimally conscious and 11 in a vegetative/unresponsive wakeful state	Participants received three different one- hour long interventions in a randomized crossover fashion	• The interventions consisted of a) stretching (30 mins) and splint (30 mins); b) splint (30 mins) and no treatment (30 mins); c) Manual stretching (30 mins) and no treatment (30 mins)	Modified Ashworth Scale for finger flexors and passively measured distance from thumb to fingers	Baseline, immediately after treatment and 60 min after treatment (retention)	Stretching and splinting Tone • Immediate decrease in Modified Ashworth Scale score after intervention; ES = −1.08 • Overall change in MAS score by 1 point, which is MCID. • Reduction maintained at retention; ES = −.54 Hand opening distance
						• Increased hand opening at after intervention; ES = .55• Change not maintained at retentionSplinting onlyTone• Immediate decrease in Modified Ashworth Scale score after intervention; ES = −.68• Overall change in MAS score by 1 point, which is MCID.• Reduction maintained at retention; ES = −.33Hand opening distance• Increased hand opening at POST compared to PRE; ES = .75• Change not maintained at RETStretching only• No significant changes seen at POST or RET.
Matsumoto - Miyazaki et al³⁹; n = 11DBS 20 (good)	• Duration since injury ≥8 months• Severe injuries—7 participants minimally conscious and 4 in a vegetative/unresponsive wakeful state	Acupuncture or sham stimulation provided. one week apart in a crossover randomized manner	Stimulation provided on the face, dorsum of the hand near the second metacarpal and anterior aspect of leg near the tibialis anterior muscleStimulation provided for a total of 10 min	Abductor Pollicis Brevis F/M ratio16 F waves recorded for the Abductor Pollicis Brevis muscle with stimulation provided at the median nerve	Baseline, immediately after removal of the needle (10 min) and at 20 min	AcupunctureDecrease in F/M ratio after immediately after intervention; ES = −.73• Change maintained at retention; ES = −.70Sham stimulation• No change in F/M ratio seen post-stimulation or at retention

Abbreviations: DBS, Downs and Black Checklist Score; TBI, traumatic brain injury; GCS, Glasgow Coma Scale; EMG, electromyography; ROM, range of motion; ABI, acquired brain injury; ES, effect size; MCID, minimal clinically important difference.

Two studies^34,35 investigated the effects of botulinum toxin A injections on wrist flexor muscle tone in 27 individuals post-TBI (18 males, 9 females) with moderate to severe muscle tone. Botulinum toxin A injections were delivered to target muscles under EMG guidance. Changes in muscle tone (quantified using Modified Ashworth’s Scale) and wrist extension range of motion (measured using goniometry) helped assess the effects of the injections. Muscle tone decreased and wrist extension range increased following botulinum toxin A injections (ES >.8).

Meythaler et al³⁶ assessed the effects of oral tizanidine administration on UL muscle tone in 17 individuals (14 males, 3 females) with acquired brain injuries (ABIs; TBI: n = 8, stroke: n = 9). They administered either tizanidine or placebo in a crossover fashion for 6 weeks, tapered the drug for one week and then switched over to other medication after one more week. Oral tizanidine decreased muscle tone (assessed using Original Ashworth’s scale) on the affected side immediately after treatment (ES = −.36) with no retention at 6 weeks (ES = −.1).

Moseley et al³⁷ recruited 26 individuals (23 males, 3 females) post-TBI with elbow flexion contracture, and randomized them into two groups (n = 13/group). One group received serial casting, and the other received static positioning. Serial casting increased elbow range by 22° over static positioning immediately post-intervention (ES = 1.85). One day after cast removal, elbow range gain decreased to 15° in the serial casting group (ES = 1.17), which further decreased to 11° after 2 weeks post-intervention.

Thibaut et al³⁸ randomized 17 participants (10 males, 7 females) with ABIs (TBI: n = 7, stroke: n = 10) to receive either soft splinting, 30 min of manual stretching, or no treatment. Provision of soft splinting resulted in increased hand opening ability (2.39 cm of major-palm distance, ES = .55). Additionally, soft splinting and manual stretching decreased finger flexor muscle tone after 30 min of treatment (ES = −.53 and −.55).

Matsumoto et al³⁹ assessed effects of acupuncture provision on UL muscle tone in 11 unconscious or minimally conscious males. They used a crossover study design providing acupuncture or no treatment, separated by one week. Acupuncture provision reduced the F/M ratio at the end of treatment (ES = −.73) and was retained 10 min later (ES = −.7).

Constraint Induced Movement Therapy (CIMT)

We found four studies (fair quality; Table 2) that assessed the effects of CIMT on UL impairment, activity, and self-reported UL use levels. Page and Levine⁴⁰ in a case series involving three participants (2 males, 1 female) post-TBI, constrained the less-affected side (5 h/day; 10 weeks). All participants improved UL activity performance measured using the Wolf Motor Function Test—Functional Ability Scale (WMFT; ES = 3.0) and Action Research Arm Test (ES = 1.78). Additionally, participants also improved the amount and quality of self-perceived use, assessed using the Motor Activity Log (MAL).

Table 2.

Use of Constraint Induced Movement Therapy (CIMT).

Study; Sample size (n) and Down’s and Black score	Chronicity and severity of injury	Intervention	Rehabilitation provided/Dose	Outcomes	Timing of assessment	Results
Page and Levine⁴⁰; n = 3DBS 15 (fair)	• All participants had chronic injuries• Initial injury severity level information missing	Modified CIMT.Mitt worn on the less-affected side	All individuals wore the mitt 5 days/week, for 5 h/day for 10 weeksThey also received 3 30 mins long sessions/week of both PT and OT for 10 wks	• Motor Activity Log Quality of Movement (MAL-QoM) and Amount of Use (MAL-AoU) scales,• Wolf Motor Function Test—Functional Assessment Scale (WMFT-FAS) and• Action Research Arm Test (ARAT)	• Baseline and after intervention completion	• Improvements in MAL-QoM and AoU scales above MCID levels• Improved WMFT—FAS (ES = 3.0) and ARAT scores (ES = 1.78) after intervention completion
Shaw et al⁴¹; n = 22DBS 18 (fair)	• All participants had chronic injuries• Initial injury severity level information missing	Traditional CIMTThe mitt was worn 90% of waking hours on the less-affected side	All individuals practiced UL activities for 6 h/day, 5 days/week for 2 weeksParticipants were encouraged to perform better and explicit verbal feedback on small improvements was provided	FMA, WMFT-FAS, and MAL-QoM scale	• All assessments completed at baseline and after intervention completion. Only MAL-QoM additionally assessed at retention (1 mo. and 2 yrs)	FMA scores• Improved (ES = 1.4) scores after intervention completion• Greater change seen in those with mild to moderate (ES = 1.4) and severe (ES = 1.6) impairment compared to moderate impairment (ES = 1.0)WMFT scores• Improved WMFT-FAS (ES = .7) after intervention completion• Greater change seen in those with moderate impairment (ES = 1.3) compared to mild to moderate (ES = .6) and severe (ES = .5) impairmentMAL scores• MAL-QoM scores improved at the end of the intervention (ES = 2.1) with change retained at 1 month (ES = 2.1) and 2 years (ES = 1.3). These changes exceeded published MCID levels• Compared to those with severe impairment, greater change seen immediately post treatment in those with and at 1 month retention (mild to moderate impairment; ES = 2.4) and moderate impairment; ES = 3.7)
Morris et al⁴²; n= 29DBS 17 (fair)	• All participants had chronic injuries• Initial injury severity level information missing	Traditional CIMTThe mitt was worn 90% of waking hours on the less-affected side	All individuals were involved in performance of UL activities for 6 h/day, 5 days/week for 2 weeks	• FMA, WMFT-FAS, WMFT time to complete activities and MAL-QoM and AoU scales• Visual attention, task-switching, and global cognition were also assessed	• Baseline and after intervention completion• Cognitive factors were assessed as part of inclusion criteria	FMA scores• The intervention led to improvements in FMA scores (ES = 1.5)WMFT scores• The intervention led to improvements in WMFT-FAS scores (ES = .4) and time to complete activities (ES = .2)• Individuals were 22.6% times faster in completing activities, which is at MCID levelsMAL scores• The intervention led to improvements in and MAL-QoM (ES = 2.1) and AoU (ES = 1.7) scores• These changes exceeded published MCID levels• Increase in self-perceived arm use correlated with better global cognition, visual attention and task-switching
Cho et al⁴³; n = 9 (3 TBI)DBS 14 (fair)	• Injury sustained ≥12 weeks before study participation• Initial injury severity level information missing	Traditional CIMT with splints that prevented contact of thumb and index finger on the less-affected sideThe three participants with TBI for two, three or five weeks	The individuals continued with whatever therapy was previously prescribedExact details of dose in terms of time spent or numbers of repetitions missing	• Perdue Pegboard Test	• Weekly, scores on the Perdue Pegboard test recorded and scoring stopped when no change was seen for 3 consecutive weeks	• Wearing splint resulted in improved performance on the test (ES = 1.31)

Abbreviations: DBS, Downs and Black Checklist Score; TBI, traumatic brain injury; ABI, acquired brain injury; FMA, Fugl-Meyer assessment; WMFT-FAS, Wolf Motor Function Test—Functional Assessment Scale; MAL-QoM, Motor Activity Log Quality of Movement; MAL-AoU, Motor Activity Log Amount of Use; ES, effect size; MCID, minimal clinically important difference.

Two studies examined the effects of CIMT on participants with chronic TBI. In both studies, participants wore the mitt on the less-affected limb for 90% of waking hours. All participants (n = 22; 14 males, 8 females) in the first study by Shaw et al⁴¹ decreased UL motor impairment measured using the Fugl-Meyer Assessment (FMA; ES = 1.4) and improved in performance of ADL (measured using WMFT; ES = .7) immediately after treatment. Participants also reported an increase in self-perceived quality of movement immediately after the intervention (ES = 2.1), which was retained at one month (ES = 2.1) and at two years post-intervention (ES = 1.3).

Participants in the other study by Morris et al⁴² (n = 29; 19 males, 10 females) similarly had better scores on the FMA (ES = 1.5) and WMFT (ES = .4). Participants reported an increase in the amount (ES = 1.7) and quality (ES = 2.1) of self-perceived UL use after the intervention. Participants reporting better use of the more affected UL had better global cognition (assessed using the Mini-Mental Scale) and visual attention and task-switching (measuring using the Trail Making Tests A and B).

Cho et al⁴³ examined the effects of CIMT on fine motor function of the hand in 9 participants (8 males, 1 female) with ABIs (TBI: n = 3, stroke: n = 6). The less-affected side was partially constrained with an opposition restriction splint that blocked use of the thumb and index finger. All participants were evaluated weekly using the Perdue Pegboard test, until no change was seen in three consecutive assessments. Constraining the less-affected side resulted in improved performance on the pegboard test (ES = 1.31).

Virtual Reality Gaming

We found five studies (fair to good quality; Table 3) that assessed the effects of VR on motor performance outcomes and UL functional outcomes.

Table 3.

Use of virtual reality gaming.

Study;Sample Size (n)Down’s and Black score	Chronicity and severity of injury	Intervention	Rehabilitation provided/Dose	Outcomes	Timing of assessment	Results
Ustinova et al⁴⁴; n = 13DBS 18 (fair)	• All participants had chronic injuries• Mild to moderate motor impairment	Single session consisting of 10 trials of games (90 seconds each) to pop balloons	Every trial consisted of 20–25 reaching movements for a total of 200–250 reaches	Trajectory straightness and movement time	• Data for reaching to balloons in front of the participants compared between 1st trial, 10th trial and 30 minutes post-practice (retention)	Trajectory straightness• Participants had straighter movements after 10 trials ES = 1.07• These changes were retained; ES = 1.0Movement time• Participants took less time to complete reaches after 10 trials; ES = 0.5• These changes were retained; ES = .5
Ustinova et al⁴⁵; n = 15DBS 18 (fair)	• All participants had chronic injuries• Severity was either mild (n = 5, PTA <30 min), moderate (n = 8, 30 min <PTA <24 h) or severe (n = 3; PTA >24 hours)	15 sessions of exergaming	Sessions involving re-training whole body co-ordination, including arm co-ordination, posture, and gaitGames included collecting coins, reaching for flowers, and popping bubbles2–3 oneh sessions every week	Trajectory straightness for the upper limb and dynamic stability index calculated from trunk displacement data in the frontal planeKinematic data were obtained using the Kinect Sensor	• Baseline, intervention completion and one-month after end of intervention	• All participants had straighter reaching movements (ES = .92). and better dynamic balance (ES = 1.31) at intervention completion• These changes were not retained
Mumford et al⁴⁶; n = 9DBS 17 (fair)	• All participants had chronic injuries• Severe to very severe injuries (duration of PTA from 26-270 days)	12 one-hour sessions involving reaching movements	Movements involved and goal directed point to point reaching exploring different parts of the arm workspaceEach session was 40 min long and the intervention duration was 4 weeks	Reaching accuracyBox & Block Test	• Twice at baseline and once after intervention completion	Reaching accuracy• Improved reaching accuracy for both left (ES: = .63) and right UL (ES = .54)Box and Block Test• Greater numbers of blocks transferred at intervention completion using both left (ES = .42) and right hands (ES = .61)• Change of 6 blocks in right hand above measurement error and represents true change
Syed and Kamal 2019; n = 34 (9 TBI)DBS 16 (fair)	• Details on chronicity and initial injury severity not provided	Participants allotted to one of two groups, performing exercises in virtual reality (n = 17, 6 TBI) or conventional rehabilitation (n = 17, 3 TBI)	Virtual reality exergaming involved moving within base of support, stepping, sit-to-stand, skipping, jumping, and joggingConventional rehabilitation involved walking, picking up objects from the floor, moving within base of support, jumping, skipping, and joggingAll participants received two 40 min sessions/week for 6 weeks	Disabilities of Arm Shoulder and Hand (DASH) questionnaire and Berg’s Balance scale (BBS)	• Baseline and intervention completion	• For participants with TBI, greater within group changes were noted after exergaming intervention completion for both BBS (ES = 5.73) and DASH (ES = 2.35)• Change in BBS above measurement error and represents true change
Buccalleto et al 2020; n = 21 (17 TBI)DBS 19 (good)	• All participants had chronic injuries• Level of initial injury severity not provided	Participants randomized to an early (n = 11) or a delayed treatment group (n = 10; training 3 weeks after study initiation). The BrightBrainer system was used to perform exergames	All participants started with unimanual games and then progressed to playing bimanual games using handheld controllersGames trained cognitive and motor aspects of movements	FMA, BBT and Jebsen Taylor Hand Function Test.	• Baseline and intervention completion	• No change seen in FMA or BBT scores• Improved scores on the Jebsen Taylor test for both groups after intervention (ES = .52)• Change in BBS above measurement error and represents true change

Abbreviations: DBS, Downs and Black Checklist Score; TBI, traumatic brain injury; PTA, post traumatic amnesia; UL, upper limb; FMA, Fugl-Meyer Assessment; BBT, Box and Blocks Test; ES, effect size.

In two studies, Ustinova and colleagues examined the effects of task-practice of reaching movements from a standing position. In the first study,⁴⁴ 13 participants post-TBI (6 males, 7 females) practiced 10 trials of reaching movements. Movements were recorded using a motion analysis system. At the end of 10 trials, participants reached faster to the targets (ES=.54), and further improved (ES = .74) at retention testing (30 min post-intervention). The participants also had straighter reaching movements (ES = −1.07), which were retained (ES = −.97).

The second study⁴⁵ examined the effects of multiple sessions of playing games on balance, reaching and co-ordination. Participants (n = 15; 10 males, 5 females) with chronic TBI played games for 15 sessions (thrice weekly). All participants were assessed at baseline, after practice, and one-month post-practice. Dynamic balance (ES = 1.33) and reaching movement straightness (ES = −1.16) improved after practice and at one month, these changes were retained.

Mumford et al⁴⁶ examined the effects of repetitive practice of unimanual and bimanual UL movements in nine individuals (5 males, 4 females) with severe chronic TBI. Assessments included kinematic measures of reaching as well as the number of blocks transferred on the Box and Blocks Test (BBT). After training, all participants had more accurate movements (ES = .62) and transferred more blocks (ES = .42).

Syed and Kamal⁴⁷ assessed the effects of VR-based gaming on 34 individuals with a variety of neurological disorders (26 males, 8 females) including TBIs (n = 9). Participants received 12 sessions of either VR-based training (n = 17) or conventional training (n = 17). Both groups improved after training with greater changes noted with VR-based (P < .001) compared to conventional training (P < .05) for both balance (assessed using Berg Balance Scale, BBS) and self-reported UL ADL performance [assessed using Disabilities of the Arm, Shoulder and Hand (DASH) questionnaire]. When results were compared only for the participants post-TBI, greater within group changes were noted after VR-based training compared to before for BBS (ES = 5.73) and DASH (ES = −2.35).

In another study, Buccellato et al⁴⁸ examined the effects of VR-based gaming on a group of 21 participants (15 males, 6 females) with ABIs (TBI: n = 13, stroke: n = 4, a combination of stroke + TBI: n = 4). Participants were randomized to an early treatment group (n = 11) or a delayed treatment group (began training 3 weeks after study initiation; n = 10). The effects of this system on UL function, dexterity, and activity performance were assessed using the FMA, BBT, and Jebsen Taylor Hand Function test, respectively. Early or delayed training did not result in improved function or dexterity. However, activity performance was improved (ES = .52).

Non-Invasive Stimulation

We found three studies (fair quality) that assessed the effects of non-invasive stimulation including use of neuromuscular electrical stimulation (NMES; one study; Table 4A) and transcranial direct current stimulation (tDCS; two studies; Table 4B) on UL motor improvement after TBI.

Table 4.

Use of non-invasive stimulation and arm ability training.

Study;Sample size (n),Down’s and Black score	Chronicity and severity of injury	Intervention details	Rehabilitation provided/Dose	Outcomes	Timing of assessment	Results
A. Use of neuromuscular electrical stimulation
Alon et al⁴⁹; n = 20 (7 TBI)DBS = 17 (fair)	• All participants had chronic injuries• Initial injury severity level information missing	Provision of NMES enabling reciprocal finger flexion and extension along with grasp and release	Forearm-hand splint with surface electrodes positioned over the wrist and hand musclesPulses delivered in an interrupted mode; contraction and relaxation intervals: 3–7 sDaily average of 3.5 h stimulation for a total of 4 months	Resting postures, active, and passive ROM at the wrist and elbow joints assessed using goniometry at the beginning and end of the intervention	• Baseline and after intervention completion	Wrist Joint• More extended resting posture (ES = 3.71) at the intervention completion• Increased range of passive (ES = 2.69) as well as active extension (ES = 2.73)Elbow Joint• More extended resting posture (ES = 4.09) at the intervention completion• Increase in active elbow extension ROM (ES = 6.91)
B. Use of Transcranial Direct Current Stimulation
Kang et al⁵⁰; n = 9DBS = 18 (fair)	• Participants had either sub-acute (n = 4) or chronic (n = 5) injuries• Initial injury severity level information missing	Real (2 mA anodal) or sham stimulation	tDCS stimulation over DLPFC applied for 20 minSham stimulation consisted of 1 min ramp up and ramp down	Reaction time on Contrast reaction time task	• Baseline, immediately after intervention completion, 3 h and 24 h after intervention completion• MMSE scores assessed only at baseline	• Tendency (P = .056) to reduce reaction time after real compared to sham stimulation; ES = .89• This change was not maintained 3 hours post stimulation (ES = .1)• However, better retention of change in reaction times 24 hours post stimulation (ES = .96) and this change correlated moderately with initial MMSE scores (r = .67)
Middleton et al⁵¹; n= 5 (2 TBI)DBS = 14 (fair)	• Both participants had chronic injuries• Initial injury severity level information missing	Bi-hemispheric stimulation of 1.5 mA for 20 minTotal of 24 sessions, sessions held thrice weekly	Stimulation followed by intensive task-specific practice of UL gross and fine motor activitiesGross motor activities—reaching for items on shelves, hitting a balloon with a racquet and simulating household choresFine motor activities —flipping playing cards and manipulating small change	Clinical: FMA and BBTKinematic: Movement straightness and speed assessed using the robotic manipulandumAssessments completed immediately after practice and at 6 month retention assessment	• Baseline, immediately after intervention completion, and 6 mos. after intervention completion	In participants with TBI-• Intervention led to better FMA scores (ES = .47), which was retained (ES = .42)• Change in FMA scores exceeded MCID levels• All participants moved faster (ES = .37). At retention, participants continued to move faster (ES = .70)
C. Use of Arm Ability Training
Platz et al⁵²; n = 60 (15 TBI)DBS = 25 (excellent)	• Participants had acute to early chronic injuries (3 weeks—24 months) post-injury• Initial injury severity level information missing	Participants randomized into groups to received Arm Ability Training, Ability Training with knowledge of results feedback or no Arm Ability Training (n = 20 each)	Arm Ability Training included activities involving dexterity manipulation, aiming for targets, and gripping objects of different sizesThe knowledge of results feedback group got average feedbackDetails on what exact intervention the no Arm Ability Training group received are missing	Clinical: Hand function evaluated using TEMPAKinematic: Movement time for aiming movements performed using a stylus	• Baseline, intervention completion and one year after intervention completion	TEMPA• Participants who received Arm Ability Training took less time to complete the TEMPA (ES = −.95) at intervention completion• Changes were retained at one year (ES = −.82)Kinematic outcomes• Participants in the Arm Ability Training group had faster movements at intervention completion (ES = −.73)• No information provided about retention testing• Provision of feedback did not result in additional gains

Abbreviations: DBS, Downs and Black Checklist Score; TBI, traumatic brain injury; ABI, acquired brain injury; ROM, Range of motion; MMSE, Mini-Mental Scale Examination; FMA, Fugl-Meyer Assessment; BBT, Box and Blocks Test; ES, effect size; MCID, minimal clinically important difference.

Alon et al⁴⁹ assessed the effects of provision of NMES enabling reciprocal finger flexion and extension along with grasp and release in 20 individuals (14 males, 6 females) with chronic ABIs (TBI: n = 7, stroke: n = 13). All participants received an average of 3.5 h stimulation daily over the course of the intervention, which lasted for almost four months. All participants had a more extended posture at the elbow (ES = 4.09) and wrist (ES = 3.71) at rest at the end of the intervention. At the wrist, participants improved their range of passive extension (ES = 2.69) as well as active flexion and extension (ES = 2.73). At the elbow, active ROM of elbow movement increased (ES = 6.91).

Kang and colleagues⁵⁰ assessed the effects of 2 mA anodal tDCS to the left dorsolateral prefrontal cortex on reaction time to an attention task. Nine individuals (8 males, 1 female) with chronic TBI participated and were randomized to receive active tDCS for 20 min or sham stimulation after one week in a crossover fashion. Reaction time on a computerized timed task decreased after application of real tDCS vs sham stimulation at the end of the intervention (ES = −.89). However, this change was not maintained at the two retention assessments (3 h and 24 h after the end of stimulation).

Middleton et al⁵¹ examined the effects of bi-hemispheric stimulation followed by robotic training on five participants with ABIs (TBI; n = 1, stroke: n = 3, stroke + TBI: n = 1). All participants performed strengthening and functional activities for a total of 40 min. Each participant received a 1.5 mA intensity concurrent stimulation for the first 15 min. Results for participants post-TBI (2 males) revealed improvements only in FMA scores (ES = −.47), which were retained (ES = −.42). Participants post-TBI also reached the targets faster at the end of the intervention (ES = .37; assessed by the KINARM^© robotic device) and continued to improve 6 months later (ES = .7).

Arm Ability Training (AAT)

We found one study (excellent quality; Table 4C) that assessed the effects of AAT on motor performance outcomes and hand function. In this study by Platz and colleagues,⁵² 60 participants (36 males, 24 females) with ABI (TBI: n = 15, stroke: n = 45) were randomized into three groups: a control group, a group receiving AAT and a group receiving AAT + knowledge of results (KR) feedback (n = 20 each). Activity performance was assessed using the time to complete the TEMPA (Test Evaluant les Membres Superieurs des Personnes Agees). Kinematic assessment of an aiming movement on a stylus between two targets was also conducted. Provision of AAT resulted in faster performance on the TEMPA (ES = −.95), which was retained one year later (ES = −.75). Participants receiving AAT also had faster aiming movements (ES = .67). Providing KR feedback did not enhance task performance.

Stem Cell Transplantation

We found one study (fair quality; Table 5A) assessing the effects of stem cell transplantation on motor impairment. This study⁵³ examined the effects of provision of injection of mesenchymal stem cells derived from the umbilical cord. Forty participants (32 males, 8 females) with moderate to severe TBI were randomized to receive the injections or to a control group (n = 20/group). The cells were injected into the sub-arachnoid space after lumbar puncture performed between the third and fourth or fourth and fifth vertebrae. Motor impairment was assessed using the FMA at baseline and 6 months after the injection. The Functional Independence Measure (FIM) helped quantify assistance in activity performance. The intervention group had significantly better improvement in FMA scores than the control group for both the UL (ES = 1.38) and LL (ES = .88) as well as FIM scores (ES = 1.17).

Table 5.

Use of stem cells and other interventions.

Study;Sample size (n),Down’s and Black score	Chronicity and severity of injury	Intervention details	Rehabilitation provided/Dose	Outcomes	Timing of assessment	Results
A. Stem cell transplantation
Wang et al⁵³ 2013; n = 40DBS = 18 (fair)	• All participants had chronic injuries• All participants had severe injuries (mean GCS score of 7)	Participants were randomized into two groups to receive stem cells or a control groupThe stem cell group received an injection of umbilical cord mesenchymal stem cells2 mL of stem cell suspension (containing 1 × 10⁷ stem cells) injected into subarachnoid space between lumbar vertebrae 3 and 4 or 4 and 5	Details on whether the intervention or control group received any form of rehabilitation are missing	FMA and FIM.	• Baseline and 6 mos. Post injection	• Significant improvement in upper (ES= 1.38) and lower limbs (ES = .88) FMA scores as well as FIM scores (ES = 1.17) for the intervention group• No change seen in control group
B. Other Interventions
Sietsema et al⁵⁴; n = 20DBS = 18 (fair)	• All participants had chronic injuries• All participants had mild to moderate injuries (RLA staging score IV or V)	Computer game involving reaching from a seated position or rote arm reaching	Initially, NDT-based intervention was providedParticipants performed rote reaching exercises or played a computer gameThe game involved repeating a sequence of flashing lights and sounds by pressing certain buttonsThe order of playing the game or rote arm reaching was counterbalanced	Reaching distance between hip and wrist measured using a motion capture system	• Baseline and after 10 trials in each condition	• Greater reaching distance over the 10 trials of playing the game compared to rote reaching (ES = 1.41)
Croce et al⁵⁵; n = 51DBS = 18 (fair)	• All participants had chronic injuries• All participants had moderate injuries (GCS score: 8 - 12)	Practice of an anticipation task that involved pressing a button when they saw the last light	Participants divided into 4 groups depending upon when they received knowledge of results (KR) feedback into a no KR group (n = 12), 100% KR (n = 14), summary KR (after every 5 trials; n = 13) and average KR (average information for all 5 trials; n = 12)Individuals performed 60 acquisition trials (12 blocks of 5 trials each)	Absolute error in terms of timing of responses	• Assessments conducted at intervention completion, 10 min after the last acquisition trial, (immediate retention) and 24 h following the last acquisition trial (late retention)	• Provision of KR better than no KR.100% KR• Effective immediately after practice; ES = .96• Effects reduced at immediate retention testing; ES = .67 and at late retention; ES = .37Summary KR• Beneficial immediately after practice; ES = .97• Changes maintained at immediate and late retention periods; ES = 1.21Average KR• Beneficial immediately after practice; ES = .95• Changes maintained at early retention; ES = 1.02; but declined at late retention; ES = .77
Sterr and Freigvogel⁵⁶; n = 13 (11 TBI)DBS = 18 (fair)	• All participants had chronic injuries• Initial injury severity level information missing	Forced use therapy	All participants initially received OT for 90 min for 4 weeks in phase AThis was followed by forced use therapy involving principles of shaping for another 4 weeks in phase BThe participants practiced 4-10 tasks in each session	The Frenchay Arm Test; MAL-AoU, MAL-QoM as well as WMFT-FAS	• At the end of phases A and B, and 1 month after the end of Phase B	Immediately at the end of Phase B• Significant improvements seen in Frenchay Arm Test scores (ES = .72), MAL-AoU (ES = 2.38), MAL-QoM (ES = 1.98) and WMFT-FAS (ES = 1.76)• Change in MAL-AoU and QoM scores at MCID levelRetention at 4 weeks post practice• Changes were not retained compared to end of Phase A in any of the 4 clinical outcomes

Abbreviations: DBS, Downs and Black Checklist Score; GCS, Glasgow Coma Scale; FMA, Fugl-Meyer Assessment; FIM, Functional Independence Measure; ABI, acquired brain injury; KR, Knowledge of Results; WMFT-FAS, Wolf Motor Function Test—Functional Assessment Scale; OT, Occupational Therapy; MAL-QoM, Motor Activity Log Quality of Movement; MAL-AoU, Motor Activity Log Amount of Use; ES, effect size; ES, MCID, minimal clinically important difference.

Feedback and Other Interventions

We found 3 studies (fair quality; Table 5B) that assessed the effects of different interventions on UL motor impairment and activity levels in individuals post-TBI. Sietsema and colleagues⁵⁴ assessed the effects of playing a game within an occupational context compared to rote exercises on UL movement patterns. Twenty individuals (17 males, 3 females) with mild to moderate TBI participated in the study. Participants practiced 10 trials in both conditions. The total forward reaching distance from the hip to the wrist was measured using motion analysis. Game playing resulted in greater reaching distance (13 cm, ES = .63) than rote arm reaching exercises.

Croce and colleagues⁵⁵ evaluated the effectiveness of provision of KR feedback at different schedules in subjects with severe TBI (n = 51; 42 males, 9 females). All participants practiced 60 trials (5 trials/block, 12 blocks) of an anticipation task. Participants received KR feedback on timing errors after each trial at different schedules—no KR (n = 12), 100% KR (n = 14), summary KR (n = 13) and average KR (n = 12). They were then tested for immediate (after 10 minutes) and delayed (after 1 h) retention. All the three KR groups were more accurate in the last block compared to the first block of trials (ES = .96). At early retention testing, this effect was decreased in the 100% KR group. However, the summary KR (ES = 1.21) and average KR (ES = 1.02) groups continued to improve accuracy. At the late retention testing, the effects were further reduced in the 100% KR group (ES = .37) and average KR group (ES = .77) but was retained only in the summary KR group at the same level (ES = 1.21).

Sterr and Freivogel⁵⁶ examined the effects of shaping principles on UL activity performance in 13 individuals (9 males, 4 females) with ABIs (TBI: n = 11, stroke: n = 2). All participants were evaluated using the MAL, WMFT, and Frenchay Arm Test. Compared to provision of occupational therapy, task-practice using shaping principles resulted in greater motor improvement on all outcomes (MAL; AoU: ES = 2.23, QoM: ES = 1.98, WMFT; ES = 1.76 and the Frenchay Arm Test; ES = .72).

Discussion

We examined the effectiveness of different interventions to augment UL motor improvement in individuals post-TBI. Majority of the studies reported moderate to large effect sizes for intervention effectiveness. In terms of quality assessment, one study was excellent, five good, and the rest were fair.

Outcomes Used to Assess Motor Improvement

A variety of outcomes were used to assess motor improvements at different levels of the International Classification of Functioning (ICF).⁵⁷ At the motor impairment level, the FMA was commonly used.^{41,42,48,51,53} Goniometry^34,35,49 and torque-controlled passive extension³⁷ helped assess changes in wrist and elbow ranges of motion. In addition, kinematic motor performance outcomes including speed, reaching path straightness, and accuracy helped quantify motor impairment.^44-46 These kinematic measures were obtained using motion capture equipment, robotic manipulandum, or using instrumented tablets. All the above-mentioned measures have well established psychometric properties.^58,59

Muscle tone was most commonly quantified using the Ashworth’s scale or the MAS.^34,35,38 Other measures used included the Modified Tardieu Scale³⁷ or neurophysiological (H-Reflex) measures.³⁹ The MAS has been recommended as a measure of choice in published guidelines.⁵⁸ However, both the MAS and Modified Tardieu Scale have poor inter-rater reliability in individuals post-TBI.⁶⁰ Use of the MAS alone does not distinguish between the tonic and phasic components of spasticity.⁶¹ Changes noted in H-reflex based parameters do not automatically translate to better functional performance after rehabilitation.⁶² The utility of other neurophysiological measures (e.g., based on spatial threshold control of muscle activation)⁶¹ alone or in conjunction with existing clinical measures to assess muscle tone remains to be estimated.

Similar to motor impairment, a variety of assessments were used to measure activity limitations. The WMFT was commonly used^40-42,56 across the different studies. Dexterity was measured by using the BBT,^46,48,51 Purdue Pegboard Test,⁴³ TEMPA,^37,52 and Jebsen Taylor Hand Function test⁴⁸ in different studies. Limitations in ADL performance were also quantified using the FIM,^36,53 the CHART,³⁶ Frenchay Arm Test,⁵⁶ and the ARAT.⁴⁰ In addition, studies also used the DASH⁴⁷ MAL amount and quality scores,^40-42,56 and the Stroke Impact Scale⁵¹ which report participant self-perceived levels of UL use. All the measures have excellent psychometric properties,⁵⁹ and the FIM and ARAT are part of the published guidelines.⁵⁸ Inclusion of the DASH, MAL, and Stroke Impact Scale across studies is encouraging, given the suggestion to use patient reported measures as outcomes in intervention studies.⁶³

Follow-Up Assessments

It has been suggested that motor improvement after TBI is attributable in part to motor learning.¹² Retention of improvements in performance noted at the end of the intervention denote motor learning. However, only 10^{36-39,41,44,51,52,55,56} studies included any form of retention testing. Amongst these studies, the timing of testing varied widely. Retention was tested at the following periods post-intervention: 10 min,⁵⁵ 20 min,³⁹ 30 min,⁴⁴ 1 h,³⁸ 3 h,⁵⁰ 24 h,^37,50,55 4 weeks,^37,41,56 6 weeks,³⁶ 6 months,⁵¹ 1 year,⁵² and 2 years post-intervention.⁴¹

Not all studies found that changes were retained. While hypertonia was reduced in the short-term (≤24 h) using casting³⁷ and acupuncture,³⁹ long-term retention (>24 h) was absent with oral tizanidine.³⁶ Only short-term retention was assessed with VR⁴⁴ and feedback provision.⁵⁵ Use of shaping principles with⁴¹ and without⁵⁶ constraints as well as Arm Motor Ability Training⁵² resulted in long-term retention. Both short⁵⁰ and long-term⁵¹ retentions were seen with the use of tDCS. It remains to be seen if use of VR technology and use of different interventions including acupuncture and botulinum toxin A result in long-term retention in individuals post-TBI.

Presence of Cognitive and Mood Impairments

Dysfunction in different cognitive domains influences generalized motor improvement in individuals post-TBI.⁶ Only two^42,50 of the selected studies, examined the association between UL motor improvement and cognitive impairment. Few other studies provided information on baseline levels of cognitive functioning,^40,48,54,55 but did not examine the effects of baseline cognitive dysfunction with motor improvement. Only one study assessed the levels of baseline depressive symptomatology,⁴⁸ which can predict motor improvement and satisfaction with life after discharge from rehabilitation in this population.⁶⁴ The presence of cognitive impairments⁶⁵ and depressive symptoms⁶⁶ influence motor improvement after a stroke. Future studies will need to focus on the relationship between cognitive dysfunction, mood disorders, and motor recovery in individuals post-TBI to better understand their association with motor improvement.

Level of Injury Severity

Out of the 23 included studies, only few studies reported initial injury severity levels. The Glasgow Coma Scale,^{34,35,37-39,53,55} duration of post-traumatic amnesia,⁴⁵ or Rancho Los Amigos original scale⁵⁴ helped quantify initial injury severity levels. This information is an important prognostic indicator for changes in overall motor improvement and levels of activity performance assessed using the Barthel Index⁶⁷ as well as a composite score of ADL and social participation (assessed using the Glasgow Outcome Scale Extended measure).⁶⁸

The other studies did not specify the injury severity levels, but some provided FMA scores.^{41,42,44,45,47,51} FMA scores ≥50/66 and ≤49/66 represent as mild and moderate to severe levels of post-stroke UL motor impairment.⁶⁹ The FMA scores in the acute post-stroke stage can predict subsequent UL recovery potential.⁷⁰ Whether UL FMA scores can be used to make similar predictions in individuals post-TBI remains to be estimated.

Sex and Gender Considerations

As stated previously, a greater proportion of males (1.2–4.4) sustain TBIs compared to females.^3,71 The greater proportion of males amongst the included participants across the different studies are indicative of the above findings. Only two studies had an almost equal distribution of sexes,⁴⁶ or included more females.⁴⁴ Despite consistent calls for considerations of sex and gender on functional outcomes,^72,73 only one study⁴⁷ assessed the effects of sex on outcomes. Future studies must strive to include more female participants and consider the effects of sex and gender on functional outcomes.

Effect of Chronicity

Studies on interventions including acupuncture, CIMT, VR-based games, NMES, stem cells, game playing, feedback, and forced-use therapy exclusively included participants with chronic injuries. While studies using Botulinum toxin A included acute, sub-acute and chronic participants, separate analyses were conducted by chronicity.³⁴ Use of serial casting³⁷ included only participants in the sub-acute stage. Other studies including participants across all stages did not conduct separate analyses based upon chronicity.^38,50,52 Future studies must strive to include participants across all stages or conduct sub-analyses based on time since injury.

Limitations

Heterogeneity amongst the interventions used prevented performance of an overarching statistical synthesis like a meta-analysis. Among the 23 studies included in this review, only 9 studies were designed as RCTs. Although the wide variability in presenting symptoms and underlying injury severity present serious challenges in designing RCTs involving individuals post-TBI,⁷⁴ encouraging efforts are underway.⁷⁵ Only three studies included in this review had sample size calculations^37,45,52 and one study⁴⁸ provided an estimate for numbers of participants needed for future trials. Nine of the 23 studies included participants with stroke and TBI. Thus, generalization of the findings is limited to a certain extent, except for two studies,^47,51 which conducted separate analyses for individuals post-TBI. Future studies should include only individuals post-TBI or conduct separate sub-analyses for this population. Limits placed on age (adults), language (English only) and non-inclusion of terms like shoulder, elbow, wrist, and hand. may have led to exclusion of some studies.

Conclusion

Preliminary evidence suggests that different rehabilitation interventions may facilitate UL motor improvement in individuals post-TBI. This review has identified several new questions in individuals post-TBI such as when compared to baseline, does provision of: (i) Botulinum toxin A followed by intensive rehabilitation result in maintained reduction of muscle tone at 1 and 3-month retention assessments; (ii) CIMT results in better motor improvement compared to conventional rehabilitation following the intervention; (iii) a combination of interventions such as VR-based gaming and tDCS is more beneficial than provision of one single intervention following the intervention and at retention; and (iv) provision of knowledge of performance feedback is useful and results in similar or better improvements than KR feedback at the end of the intervention and at retention assessments. We hope that these questions will help guide and foster further research to evaluate the efficacy of the most suitable interventions to reduce impairment and improve activity performance post-TBI.

Supplemental Material

sj-xlsx-1-nnr-10.1177_15459683211056662 – Supplemental Material for Upper Limb Motor Improvement after Traumatic Brain Injury: Systematic Review of Interventions

Supplemental Material, sj-xlsx-1-nnr-10.1177_15459683211056662 for Upper Limb Motor Improvement after Traumatic Brain Injury: Systematic Review of Interventions by Sandeep K. Subramanian, Melinda K. Fountain, Ashley F. Hood and Monica Verduzco-Gutierrez in Neurorehabilitation and Neural Repair

Footnotes

Acknowledgments

The authors would like to acknowledge Dr Kate Aultman for her support and encouragement in this project.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: SKS was funded by a pilot grant awarded by the School of Health Profession, UT Health San Antonio.

ORCID iDs

Sandeep K. Subramanian

Monica Verduzco-Gutierrez

Supplement material

Supplemental material for this article is available online.

References

James

Theadom

Ellenbogen

, et al. Global, regional, and national burden of traumatic brain injury and spinal cord injury, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet Neurol. 2019;18:56-87.

Traumatic Brain Injury & Concussion: Basic Information. [Internet]. Atlanta, GA: Center for Disease Control and Prevention. https://www.cdc.gov/traumaticbraininjury/get_the_ facts.html. Accessed July 25, 2021.

Taylor

Bell

Breiding

. Traumatic brain injury–related emergency department visits, hospitalizations, and deaths—United States, 2007 and 2013. MMWR CDC Surveill Summ. 2017;66:1-16.

Dixon

. Pathophysiology of traumatic train injury. Phys Med Rehabil Clin N Am. 2017;28:215-225.

Capizzi

Woo

Verduzco-Gutierrez

. Traumatic brain injury: an overview of epidemiology, pathophysiology, and medical management. Med Clin North Am. 2020; 104:213-238.

Allanson

Pestell

Gignac

Yeo

Weinborn

. Neuropsychological predictors of outcome following traumatic brain injury in adults: a meta-analysis. Neuropsychol Rev. 2017; 27:187-201.

Weber

Spirou

Chiaravalloti

Lengenfelder

. Impact of frontal neurobehavioral symptoms on employment in individuals with TBI. Rehabil Psychol. 2018;63:383-391.

Meulenbroek

Turkstra

. Job stability in skilled work and communication ability after moderate-severe traumatic brain injury. Disabil Rehabil. 2016;38:452-461.

Oberholzer

Muri

. Neurorehabilitation of Traumatic Brain Injury (TBI): a clinical review. Med Sci 2019;7(3):47. doi:10.3390/medsci7030047

10.

Fulk

Nirider

. Physical Rehabilitation. 7th ed. Philadelphia, PA: FA Davis; 2019.

11.

Lamontagne

Bayley

Marshall

, et al. Assessment of users’ needs and expectations toward clinical practice guidelines to support the rehabilitation of adults with moderate to severe Traumatic Brain Injury. J Head Trauma Rehabil. 2018;33:288-295.

12.

Breceda

Dromerick

. Motor rehabilitation in stroke and traumatic brain injury: stimulating and intense. Curr Opin Neurol. 2013;26:595-601.

13.

Glenn

Shih

. Rehabilitation following TBI. In: JW

Tsao

, ed. Traumatic Brain Injury: A Clinician’s Guide to Diagnosis, Management, and Rehabilitation. Springer; 2020:293-327.

14.

Kleim

Jones

. Principles of experience-dependent neural plasticity: implications for rehabilitation after brain damage. J Speech Lang Hear Res. 2008;51:S225–S239.

15.

Bayley

Swaine

Lamontagne

, et al. INESSS-ONF Clinical Practice Guideline for the Rehabilitation of Adults with Moderate to Severe Traumatic Brain Injury. Toronto, ON, Canada: Ontario Neurotrauma Foundation; 2018. https://braininjuryguidelines.org/modtosevere. Accessed July 25, 2021.

16.

Dobkin

. Motor rehabilitation after stroke, traumatic brain, and spinal cord injury: common denominators within recent clinical trials. Curr Opin Neurol. 2009;22:563-569.

17.

Vanderbeken

Kerckhofs

. A systematic review of the effect of physical exercise on cognition in stroke and traumatic brain injury patients. NeuroRehabilitation. 2017;40:33-48.

18.

Hellweg

Johannes

. Physiotherapy after traumatic brain injury: a systematic review of the literature. Brain Inj. 2008;22:365-373.

19.

Jang

. Review of motor recovery in patients with traumatic brain injury. NeuroRehabilitation. 2009;24:349-353.

20.

McCrea

Eng

Hodgson

. Biomechanics of reaching: clinical implications for individuals with acquired brain injury. Disabil Rehabil. 2002;24:534-541.

21.

Andelic

Sigurdardottir

Tenovuo

. Rehabilitation after severe TBI. In: T

Sundstrom

P-O

Grände

Luoto

Rosenlund

Undén

Wester

, eds. Management of Severe Traumatic Brain Injury Evidence, Tricks, and Pitfalls. 2nd ed. Springer; 2020:547-556.

22.

Perumparaichallai

Lewin

Klonoff

. Community reintegration following holistic milieu-oriented neurorehabilitation up to 30 years post-discharge. NeuroRehabilitation. 2020;46:243-253.

23.

Soeker

Van Rensburg

Travill

. Individuals with traumatic brain injuries perceptions and experiences of returning to work in South Africa. Work. 2012;42:589-600.

24.

Eriksen

Frandsen

. The impact of patient, intervention, comparison, outcome (PICO) as a search strategy tool on literature search quality: a systematic review. J Med Libr Assoc. 2018;106:420-431.

25.

Subramanian

Fountain

Hood

. Interventions to augment upper extremity motor improvement in individuals with a Traumatic Brain Injury: a systematic review [abstract]. Neurorehabil Neural Repair. 2018;32:1110-1111.

26.

Husted

Cook

Farewell

Gladman

. Methods for assessing responsiveness: a critical review and recommendations. J Clin Epidemiol. 2000;53:459-468.

27.

Cohen

The effect size index: d. Statistical power analysis for the behavioral sciences. 1988;2:284-288.

28.

Morton

Barton

Rice

Morrissey

. Risk factors and successful interventions for cricket-related low back pain: a systematic review. Br J Sports Med. 2014;48:685-691.

29.

Downs

Black

. The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J Epidemiol Comm Health. 1998;52:377-384.

30.

Hootman

Driban

Sitler

Harris

Cattano

. Reliability and validity of three quality rating instruments for systematic reviews of observational studies. Res Synth Methods. 2011;2(2):110-118. doi:10.1002/jrsm.41

31.

Aubut

Marshall

Bayley

Teasell

. A comparison of the PEDro and Downs and Black quality assessment tools using the acquired brain injury intervention literature. NeuroRehabilitation. 2013;32:95-102. doi:10.3233/NRE-130826

32.

Subramanian

Caramba

Hernandez

Morgan

Cross

Hirschhauser

. Is the Downs and Black scale a better tool to appraise the quality of the studies using virtual rehabilitation for post-stroke upper limb rehabilitation? In: WG

Wright

Fluet

Subramanian

Agmon

, eds. Proceedings of the 2019 International Conference on Virtual Rehabilitation (ICVR); July 22–24, 2019; Tel Aviv, Israel. IEEE Publications. doi: 10.1109/ICVR46560.2019.8994724

33.

O’Connor

Tully

Ryan

Bradley

Baxter

McDonough

. Failure of a numerical quality assessment scale to identify potential risk of bias in a systematic review: a comparison study. BMC Res Notes. 2015;8(1):224. doi: 10.1186/s13104-015-1181-1.

34.

Yablon

Agana

Ivanhoe

Boake

. Botulinum toxin in severe upper extremity spasticity among patients with traumatic brain injury: an open-labeled trial. Neurology. 1996;47:939-944.

35.

Pavesi

Brianti

Medici

Mammi

Mazzucchi

Mancia

. Botulinum toxin type A in the treatment of upper limb spasticity among patients with traumatic brain injury. J Neurol Neurosurg Psychiatr. 1998;64:419-420.

36.

Meythaler

Guin-Renfroe

Johnson

Brunner

. Prospective assessment of tizanidine for spasticity due to acquired brain injury. Arch Phys Med Rehabil. 2001;82:1155-1163.

37.

Moseley

Hassett

Leung

Clare

Herbert

Harvey

. Serial casting versus positioning for the treatment of elbow contractures in adults with traumatic brain injury: a randomized controlled trial. Clin Rehabil. 2008;22:406-417.

38.

Thibaut

Deltombe

Wannez

, et al. Impact of soft splints on upper limb spasticity in chronic patients with disorders of consciousness: a randomized, single-blind, controlled trial. Brain Inj. 2015;29:830-836.

39.

Matsumoto-Miyazaki

Asano

Yonezawa

, et al. Acupuncture increases the excitability of the cortico-spinal system in patients with chronic disorders of consciousness following traumatic brain injury. J Altern Complement Med. 2016;22:887-894.

40.

Page

Levine

. Forced use after TBI: promoting plasticity and function through practice. Brain Inj. 2003;17:675-684.

41.

Shaw

Morris

Uswatte

McKay

Meythaler

Taub

. Constraint-induced movement therapy for recovery of upper-limb function following traumatic brain injury. J Rehabil Res Dev. 2005;42:769-778.

42.

Morris

Shaw

Mark

Uswatte

Barman

Taub

. The influence of neuropsychological characteristics on the use of CI therapy with persons with traumatic brain injury. NeuroRehabilitation. 2006;21:131-137.

43.

Cho

Jang

Lee

Song

Lee

. Effect and appropriate restriction period of constraint-induced movement therapy in hemiparetic patients with brain injury: a brief report. NeuroRehabilitation. 2005;20:71-74.

44.

Ustinova

Leonard

Cassavaugh

Ingersoll

. Development of a 3D immersive videogame to improve arm-postural coordination in patients with TBI. J NeuroEng Rehabil. 2011;8:61. doi:10.1186/1743-0003-8-61

45.

Ustinova

Perkins

Leonard

Hausbeck

. Virtual reality game-based therapy for treatment of postural and co-ordination abnormalities secondary to TBI: a pilot study. Brain Inj. 2014;28:486-495.

46.

Mumford

Duckworth

Thomas

Shum

Williams

Wilson

. Upper-limb virtual rehabilitation for traumatic brain injury: a preliminary within-group evaluation of the elements system. Brain Inj. 2012;26:166-176.

47.

Syed

Kamal

. Video game-based and conventional therapies in patients of neurological deficits: an experimental study. Disabil Rehabil Assist Technol. 2021;16:332-339.

48.

Buccellato

Nordstrom

Murphy

, et al. A randomized feasibility trial of a novel, integrative, and intensive virtual rehabilitation program for service members post-acquired brain injury. Mil Med. 2020;185:e203-e211.

49.

Alon

Dar

Katz-Behiri

Weingarden

Nathan

. Efficacy of a hybrid upper limb neuromuscular electrical stimulation system in lessening selected impairments and dysfunctions consequent to cerebral damage. J Neuro Rehab. 1998;12:73-79.

50.

Kang

Kim

Paik

. Transcranial direct current stimulation of the left prefrontal cortex improves attention in patients with traumatic brain injury: a pilot study. J Rehabil Med. 2012;44:346-350.

51.

Middleton

Fritz

Liuzzo

Newman-Norlund

Herter

. Using clinical and robotic assessment tools to examine the feasibility of pairing tDCS with upper extremity physical therapy in patients with stroke and TBI: a consideration-of-concept pilot study. NeuroRehabilitation. 2014;35:741-754. doi:10.3233/NRE-141178

52.

Platz

Winter

Müller

Pinkowski

Eickhof

Mauritz

K-H

. Arm ability training for stroke and traumatic brain injury patients with mild arm paresis: a single-blind, randomized, controlled trial. Arch Phys Med Rehabil. 2001;82:961-968.

53.

Wang

Cheng

Dai

, et al. Umbilical cord mesenchymal stem cell transplantation significantly improves neurological function in patients with sequelae of traumatic brain injury. Brain Res. 2013;1532:76-84.

54.

Sietsema

Nelson

Mulder

Mervau-Scheidel

White

. The use of a game to promote arm reach in persons with traumatic brain injury. Am J Occup Ther. 1993;47:19-24.

55.

Croce

Horvat

Roswal

. Augmented feedback for enhanced skill acquisition in individuals with traumatic brain injury. Percept Mot Skills. 1996;82:507-514.

56.

Sterr

Freivogel

. Motor-improvement following intensive training in low-functioning chronic hemiparesis. Neurology. 2003;61:842-844.

57.

Stucki

Cieza

Ewert

Kostanjsek

Chatterji

Ustun

. Application of the International Classification of Functioning, Disability and Health (ICF) in clinical practice. Disabil Rehabil. 2002;24:281-282.

58.

McCulloch

De Joya

Hays

, et al. Outcome measures for persons with moderate to severe traumatic brain injury: recommendations from the American Physical Therapy Association Academy of Neurologic Physical Therapy TBI EDGE Task Force. J Neurol Phys Ther. 2016;40:269-280.

59.

Rehabilitation Measures Database [Internet]. www.sralab.org/rehabilitation-measures. Accessed July 25, 2021.

60.

Mehrholz

Wagner

Meissner

, et al. Reliability of the Modified Tardieu Scale and the Modified Ashworth Scale in adult patients with severe brain injury: a comparison study. Clin Rehabil. 2005;19:751-759.

61.

Subramanian

Feldman

Levin

. Spasticity may obscure motor learning ability after stroke. J Neurophysiol. 2018;119:5-20.

62.

Elovic

Simone

Zafonte

. Outcome assessment for spasticity management in the patient with traumatic brain injury: the state of the art. J Head Trauma Rehabil. 2004;19:155-177.

63.

Winstein

. Thoughts about the negative results of clinical trials in rehabilitation medicine. Kinesiol Rev. 2018;7:58-63.

64.

Proctor

Best

. Social and psychological influences on satisfaction with life after brain injury. Disabil Health J. 2019;12:387-393.

65.

Subramanian

Chilingaryan

Sveistrup

Levin

. Influence of training environment and cognitive deficits on use of feedback for motor learning in chronic stroke. In: J

Deutsch

Wright

Weiss

, eds. Proceedings of the 2015 International Conference on Virtual Rehabilitation (ICVR); June 9–12, 2015; Valencia, Spain. IEEE Publications. doi: 10.1109/ICVR.2015.7358582

66.

Subramanian

Chilingaryan

Sveistrup

Levin

. Depressive symptoms influence use of feedback for motor learning and recovery in chronic stroke. Restor Neurol Neurosci. 2015;33:727-740.

67.

Kim

Pyun

S-B

. Prediction of functional outcome and discharge destination in patients with traumatic brain injury after post-acute rehabilitation. Int J Rehabil Res. 2019;42:256-262.

68.

Kodliwadmath

Koppad

Desai

Badiger

. Correlation of Glasgow outcome score to Glasgow coma score assessed at admission. Int Surg J. 2016;3:1959-1963.

69.

Subramanian

Yamanaka

Chilingaryan

Levin

. Validity of movement pattern kinematics as measures of arm motor impairment poststroke. Stroke. 2010;41:2303-2308.

70.

Stinear

Byblow

Ackerley

Smith

Borges

Barber

. Proportional motor recovery after stroke: implications for trial design. Stroke. 2017;48:795-798.

71.

Critchley

Memon

. Epidemiology of head injury. In: PC

Whitfield

Thomas

Summers

Whyte

Hutchinson

, ed. Traumatic Brain Injury: A Multidisciplinary Approach. Cambridge: Cambridge University Press; 2020:1-10.

72.

Mollayeva

Colantonio

. Traumatic brain injury: sex, gender and intersecting vulnerabilities. Nat Rev Neurol. 2018;14:711-722.

73.

Hanafy

Amodio

Haag

, et al. Is it prime time for sex and gender considerations in traumatic brain injury? Perspectives of rehabilitation care professionals[published ahead of print June 23, 2020]. Disabil Rehabil. doi:10.1080/09638288.2020.1774670.

74.

Turner-Stokes

Pick

Nair

Disler

Wade

. Multi-disciplinary rehabilitation for acquired brain injury in adults of working age. Cochrane Database Syst Rev. 2015(12):CD004170. doi:10.1002/14651858.CD004170.pub3

75.

Morris

Taub

Mark

, et al. Protocol for a randomized controlled trial of CI therapy for rehabilitation of upper extremity motor deficit: the bringing rehabilitation to American veterans everywhere project. J Head Trauma Rehabil. 2019;34:268-279.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB