10. Upper Extremity Interventions

Document Sample
10. Upper Extremity Interventions Powered By Docstoc
					                    10. Upper Extremity Interventions
 Norine Foley MSc, Robert Teasell MD, Jeffrey Jutai PhD CPsych, Sanjit Bhogal MSc, Elizabeth Kruger

Key Points
                                                                                    The Evidence-Based
Initial degree of motor impairment is the best predictor of motor                   Review of Stroke
recovery following a stroke. Functional recovery goals are                          Rehabilitation (EBRSR)
appropriate for those patients who are expected to achieve a greater                reviews current
                                                                                    practices in stroke
amount of motor recovery in the arm and hand. Compensatory
treatment goals should be pursued if there is an expected outcome of
poor motor recovery.                                                                Contacts:
                                                                                    Dr. Robert Teasell
Attempts to regain function in the affected upper extremity should be               801 Commissioners
limited to those individuals already showing signs of some recovery.                Road East
                                                                                    London, Ontario,
Neurodevelopment techniques are not superior to other therapeutic                   Canada
approaches in treatment of the hemiparetic upper extremity.                         N6C 5J1
It is uncertain whether enhanced therapy results in improved short-                 519.685.4000
term upper extremity functioning.                                                   Web:

It is uncertain whether repetitive task specific training techniques                Email:
improve upper extremity function.                                                   Robert.teasell@sjhc.lo

It is uncertain whether sensorimotor training results in improved
upper extremity function.

It is uncertain whether mental practice results in improved motor and
ADL functioning after stroke.

Hand splinting does not improve motor function or reduce
contractures in the upper extremity.

Constraint-induced movement therapy is a beneficial treatment
approach for those stroke patients with some active wrist and hand

Sensorimotor training with robotic devices improves functional and
motor outcomes of the shoulder and elbow, however, it does not
improve functional and motor outcomes of the wrist and hand.

There is preliminary evidence that virtual reality therapy may improve
motor outcomes post stroke.

 10. Upper Extremity Interventions                                                     pg. 1 of 150
It remains uncertain if hand splinting decreases spasticity.

Botulinum Toxin decreases spasticity and increases range of motion;
however, these improvements do not necessarily result in better
upper extremity function.

Botulinum Toxin in combination with electrical stimulation improves
tone in the upper extremity.

More research is needed to determine the effectiveness of Nerve
Blocks for spasticity.

Physical Therapy may not be effective for reducing spasticity in the
upper extremity.

EMG/Biofeedback therapy is not superior to other forms of treatment
in the treatment of the hemiparetic upper extremity.

Intermittent pneumatic compression is not an effective treatment for
hand edema.

It is uncertain whether transcutaneous electrical nerve stimulation
improves outcomes post-stroke

Functional Electrical Stimulation therapy improves hemiparetic upper
extremity function.

Antidepressant drugs may improve short-term motor performance.

Last updated August 2010

10. Upper Extremity Interventions                                      pg. 2 of 150
Table of Contents
Key Points .......................................................................................... 1
Table of Contents ................................................................................. 3
10. Upper Extremity Interventions Post-Stroke ............................................ 5
  10.1 Consensus Panel Treatment and Recommendations .....................................6
  10.2     Upper Extremity Interventions..............................................................8
    10.2.1 Neurodevelopmental Techniques............................................................ 8
    10.2.2 Bilateral Arm Training ....................................................................... 13
    10.2.3 Additional/Enhanced Upper Extremity Therapy ......................................... 18
    10.2.4 Strength Training............................................................................. 22
    10.2.4 Repetitive/Task- Specific Training Techniques .......................................... 24
    10.2.5 Truck Restraint ............................................................................... 28
    10.2.6 Sensorimotor Training and Somatosensory Stimulation............................... 29
    10.2.7 Mental Practice ............................................................................... 37
    10.2.8 Hand Splinting ................................................................................ 41
    10.2.9 Constraint-Induced Movement Therapy.................................................. 44
    10.2.10 Mirror Therapy .............................................................................. 63
    10.2.11 Feedback .................................................................................... 64
  10.3 Robotic Devices For Movement Therapy................................................... 67
    10.3.1 MIT-Manus .................................................................................... 68
    10.3.2 Mirror-Image Motion Enabler Robots (MIME) ............................................ 71
    10.3.3 Assisted Rehabilitation and Measurement (ARM) Guide............................... 72
    10.3.4 Bi-Manu-Track ................................................................................ 73
    10.3.5 Neuro-Rehabilitation-Robot (NeReBot) ................................................... 74
    10.3.6 Continuous Passive Motion (CPM) ........................................................ 75
    10.3.8 GENTLE/s..................................................................................... 76
    10.3.9 Other Devices ................................................................................ 77
  10.4 Virtual Reality Technology..................................................................... 79
  10.5 Treatment for Spasticity or Contracture in the Upper Extremity ..................... 84
    10.5.1 Splinting ...................................................................................... 86
    10.5.2 Stretching Programs to Prevent Contracture............................................. 87
    10.5.2 Botulinum Toxin Injections ................................................................. 88
    10.5.3 Electrical Stimulation Combined with Botulinum Toxin Injection ...................... 98
    10.5.4 Nerve Block and Spasticity ................................................................ 99
    10.5.5 Physical Therapy in the Treatment of Spasticity.......................................100
    10.5.6 Shock Wave Treatment ...................................................................101
    10.5.7 Centrally Acting Muscle Relaxants (tolperisone) .......................................102
  10.6 EMG/Biofeedback .............................................................................. 103
  10.7 Transcutaneous Electrical Nerve Stimulation (TENS) ................................. 107
  10.8 Functional Electrical Stimulation (FES) ................................................... 109
  10.9 Medications Used in Motor Recovery...................................................... 122
    10.9.1 Stimulants ....................................................................................122
    10.9.2 Levodopa.....................................................................................123
    10.9.3 Antidepressants .............................................................................124
  10.10 Treatment of Hand Edema .................................................................. 125
  10.11 Summary ........................................................................................ 128

10. Upper Extremity Interventions                                                               pg. 3 of 150
 References.............................................................................................. 131

10. Upper Extremity Interventions                                                              pg. 4 of 150
                                             circulation infarcts, right hemispheric
10. Upper Extremity                          strokes, homonymous hemianopia,
                                             visual gaze deficits, visual inattention
Interventions Post-Stroke                    and paresis were associated with poor
Impaired upper extremity function is a       arm function. When Dominkus et al.
common and often devastating                 (1990) assessed motor recovery in the
problem for stroke survivors. In the         upper extremity with the Motricity
population-based Copenhagen Stroke           Index (Demeurisse et al. 1980), a
Study (Nakayama et al. 1994), 32% of         patient with initial paresis was 4.58
stroke patients had severe arm paresis       times more likely to show motor
at admission and 37% had mild                recovery compared to a patient with
paresis. In 64 out of 491 (13%)              initial paralysis. This finding has led to
stroke survivors, the arm remained           recommendations regarding which
entirely non-functional despite              patients should receive more
comprehensive rehabilitation efforts.        aggressive therapy (ie. therapy aimed
Regaining lost function in the upper         at strengthening and increasing range
extremities may be more difficult to         of motion), or to less aggressive
achieve than return of normal function       therapy (ie. therapy aimed at
(ambulation) in the lower extremities        minimizing pain and contractures).
(Hiraoka 2001). Similarly, Barreca et        Barreca et al. (2001) recommended
al. (2001) noted that, “Rehabilitation       that for patients with a poor prognosis
of the hemiplegic upper limb remains         for recovery, defined as a Chedoke
difficult to achieve, with only 5% of        McMaster score of less than stage 4,
stroke survivors who have complete           treatment should focus on minimizing
paralysis regaining functional use of        contractures and pain in the involved
their impaired arm and hand                  upper extremity. However, there is
(Dombovy 1993, Gowland 1982,                 evidence from a number of studies
Kwakkel et al. 2000). Limited                that treatment gains, albeit sometimes
rehabilitation resources, time               small, are observable in patients with
constraints, and a lack of early motor       severe initial impairment (Partridge et
recovery in the arm and hand tend to         al. 2000, Lincoln et al. 1999,
focus therapy on improving balance,          Sunderland et al. 1992, Kwakkel et al.
gait and general mobility.”                  1999, Feys 1998). There is also
                                             evidence that motor rehabilitation of
There is much discussion regarding           chronic stroke patients remains
which patients benefit the most from         successful several months or years
therapy. Nakayama et al. (1994)              after the acute stroke (Hummelshein &
reported that in a sample of stroke          Eickhof 1999, Kraft et al. 1992, Junkel
patients with severe arm paresis, with       et al. 1999). In terms of patients with
little or no active movement on              less severe initial impairment (defined
admission, that 14% of patients              by a Chedoke McMaster score of stage
experienced complete motor recovery,         4 or greater), Barreca et al. (2001)
while 30% achieved partial recovery          have recommended that an aggressive
(Hendricks et al. 2002). Similarly,          restorative program geared towards
Kwakkel et al. (2003) reported that          regaining function in the affected
11.6% of patients had achieved               upper extremity should be adopted.
complete functional recovery at 6
months, while 38% had some
dexterity. Patients with anterior

10. Upper Extremity Interventions                                           pg. 5 of 150
Previous Reviews                                       physical therapy (d= -0.01); there was
Several previous reviews have focused                  a medium effect of conventional
on upper extremity therapies for                       physical therapy compared to no
stroke survivors. Two non-systematic                   therapy (d=0.51) and a large effect of
reviews concluded that exercise                        EMG biofeedback treatment compared
therapy was beneficial and highlighted                 to conventional physical therapy
the importance of extensive practice                   (d=0.85).
(Duncan and Lai 1997, Richards and
Pohl 1999), while a recent systematic                  10.1 Consensus Panel Treatment
review of 13 randomized controlled
trials concluded that insufficient
                                                       and Recommendations
evidence was available to support the                  Barreca et al. (2001) provided
effectiveness of exercise therapies on                 consensus treatment recommendations
arm function (van der Lee et al. 2001).                for management of the post stroke
However, these authors suggested that                  arm and hand, based on a synthesis of
more intensive therapies might be                      best evidence. After reviewing the
beneficial.                                            evidence the panel came to a
                                                       consensus agreement that a
Two reviews were able to pool their                    hemiplegic upper extremity must be at
results quantitatively (Barreca et al.                 least at a Chedoke-McMaster Stroke
2001, Hiraoka 2001). Barreca et al.                    Impairment Inventory (CMSII) stage 4
(2001) reported that the pooled effect                 before full rehabilitation efforts
sizes associated with upper extremity                  designed to restore function in the
treatments were: Z=4.87 for                            arm, are attempted. The panel
sensorimotor training (including 4                     concluded that attempts to rehabilitate
RCTs); Z=3.43 for EMG-electrical                       the upper extremity of a person with a
stimulation (including 3RCTs); and                     score less than a level 4 will not meet
Z=4.44 for electrical stimulation                      with success. A more palliative
(including 2 RCTs). Hiraoka (2001)                     compensatory approach is
included 14 RCTs evaluating upper                      recommended in such a case.
extremity therapies and found an
overall effect size (d) of 0.33,                       The stages of motor recovery assessed
suggestive of a small to medium                        using the Chedoke McMaster Stroke
impact of therapy. Subgroup analyses                   Impairment Inventory, which is very
suggested that there was no treatment                  similar to the Brunnstrom Recovery
effect of neurodevelopmental                           Stages, are described below.
treatment compared with conventional

Stages of Motor Recovery of the Chedoke McMaster Stroke Impairment Inventory
(Gowland et al. 1993)
 Stage                                             Characteristics
   1     Flaccid paralysis is present. Phasic stretch reflexes are absent or hypoactive. Active
         movement cannot be elicited reflexively with a facilitory stimulus or volitionally.
  2      Spasticity is present and is felt as a resistance to passive movement. No voluntary
         movement is present but a facilitatory stimulus will elicit the limb synergies reflexively.
         These limb synergies consist of stereotypical flexor and extensor movements.
  3      Spasticity is marked. The synergistic movements can be elicited voluntarily but are not
  4      Spasticity decreases. Synergy patterns can be reversed if movement takes place in the

10. Upper Extremity Interventions                                                           pg. 6 of 150
        weaker synergy first. Movement combining antagonistic synergies can be performed when
        the prime movers are the strong components of the synergy.
  5     Spasticity wanes, but is evident with rapid movement and at the extremes of range.
        Synergy patterns can be revised even if the movement takes place in the strongest synergy
        first. Movements that utilize the weak components of both synergies acting as prime movers
        can be performed.
  6     Coordination and patterns of movement can be near normal. Spasticity as demonstrated as
        resistance to passive movement is no longer present. Abnormal patterns of movement with
        faulty timing emerge when rapid or complex actions are requested.
  7     Normal. A “normal” variety of rapid, age appropriate complex movement patterns are
        possible with normal timing, coordination, strength and endurance. There is no evidence of
        functional impairment compared to the normal side. There is a “normal” sensory-perceptual
        motor system.

 2001 Consensus Panel Recommendations for Patients with Severe Impairment

 “For the client with severe motor, sensory and functional deficits in the involved limb
 after stroke, the effectiveness literature indicates that additional treatment for the upper
 limb will not result in any significant neurological change. The evidence to date suggests
 that interventions may not lead to meaningful functional use of the affected limb at this
 stage of motor recovery.”

 1. Maintain a comfortable, pain-free, mobile arm and hand
    • emphasize proper positioning, support while at rest and careful handling of the
       upper limb during functional activities.
    • engage in classes overseen by professional rehabilitation clinicians in an
       institutional or community setting that teach the client and caregiver to perform
       self-range of motion exercises.
    • avoid use of overhead pullies that appear to contribute to shoulder tissue injury
    • use some means of external support for the upper limb in stages 1 or 2 during
       transfers and mobility
    • place upper limb in a variety of positions that include placing arm and hand within
       the client’s visual field.
    • Use some means of external support to protect the upper limb during wheelchair

 2. To maximize functional independence, stroke survivors with persistent motor
 and sensory deficits and their caregivers should be taught compensatory
 techniques and environmental adaptations that enable performance of
 important tasks and activities with the less affected arm and hand.

2001 Consensus Panel Recommendations for Patients with Moderate Impairment

“For clients with moderate impairments who demonstrate high motivation and potential for
functional motor gains

1. Engage in repetitive and intense use of novel tasks that challenge the stroke
survivor to acquire necessary motor skills to use the involved upper limb during
functional tasks and activities.

2. Engage in motor-learning training including the use of imagery.”

10. Upper Extremity Interventions                                                     pg. 7 of 150
                                                     are presented in sections 10.2.1 to
 Conclusions Regarding Management of                 10.2.9.
 the Post Stroke Arm and Hand
                                                     10.2.1 Neurodevelopmental
 There is consensus (Level 3) opinion                Techniques
 that in severely impaired upper
 extremities (less than stage 4) the focus           A variety of treatment approaches are
 of treatment should be on palliation and
                                                     in use currently. Arguably, the Bobath
                                                     approach (a neurodevelopmental
  For those upper extremities with signs             technique) is the most commonly
 of some recovery (stage 4 or better)                used, although other methods, such as
 there is consensus (Level 3) opinion                motor re-learning, orthopedic or mixed
 that attempts to restore function                   technique are also used.
 through therapy should be made.
                                                     The concepts of NDT emphasize that
  Attempts to regain function in the                 abnormal muscle patterns or muscle
  affected upper extremity should be                 tone have to be inhibited, and that
  limited to those individuals already               normal patterns should be used in
  showing signs of some recovery.                    order to facilitate functional and
                                                     voluntary movements. There are a
                                                     number of approaches that fall under
10.2 Upper Extremity                                 the heading of neurodevelopmental
Interventions                                        techniques. These include the Bobath,
A variety of treatment interventions to              Brunnstrom and Proprioceptive
improve motor recovery in the upper                  Neuromuscular Facilitation approaches.
extremity have been evaluated. They                  Therapy approaches aimed at the
                                                     rehabilitation of the lower extremity
                                                     are also discussed in Module 9.

Table 10.1 Neurodevelopmental Training (NDT) Approaches
    Approach                                          Description
Bobath             Aims to reduce spasticity and synergies by using inhibitory postures and
                   movements in order to facilitate normal autonomic responses that are involved in
                   voluntary movement (Bobath 1990).
Brunnstrom’s       Emphasis on synergistic patterns of movement that develop during recovery from
Movement Therapy hemiplegia. Encourages the development of flexor and extensor synergies during
                   early recovery, assuming that synergistic activation of the muscle will result in
                   voluntary movement (Brunnstrom 1970).
Proprioceptive     Emphasis on using the patient's stronger movement patterns for strengthening
Neuromuscular      the weaker motions. PNF techniques use manual stimulation and verbal
Facilitation (PNF) instructions to induce desired movement patterns and enhance motor function
                   (Meyers 1995)
                                                     to other types of interventions. Van
In their review of NDT vs. other                     Peppen et al. (2004) recently
treatment approaches, Barreca et al.                 conducted a systematic review of
(2003) included five RCTs (Basmajian                 specific neurological treatment
et al. 1987, Dickstein et al. 1986,                  approaches and also concluded that
Gelber et al. 1995, Logigian et al.                  compared to a Bobath approach, no
1983, van der Lee et al. 1999) and                   one particular program was favoured
concluded that NDT was not superior                  over another with respect to

10. Upper Extremity Interventions                                                      pg. 8 of 150
improvement in functional outcomes                     evaluated the effect of
(ADL), muscle strength or tone,                        neurodevelopmental techniques (NDT)
dexterity, although motor relearning                   (Logigian et al. 1983, Lord and Hall
programs were associated with shorter                  1986, Dickstein et al. 1986, Basmajian
lengths of hospital stays.                             et al. 1987, Wagenaar et al. 1990,
                                                       Gelber et al. 1995, van der Lee et al.
Paci (2003) conducted a review of 15                   1999, Langhammer and Stanghelle
trials including six RCTs (Langhammer                  2000, 2003, VanVliet et al. 2005, Platz
and Stanghelle 2000, van der Lee et                    et al. 2005, Hafsteinsdóttir et al.
al. 1999, Gelber et al. 1995, Partridge                2005). Eight of the eleven studies
et al. 1990, Basmajian et al. 1987,                    were RCTs. Another systematic review
Mulder et al. 1986), six non-                          (Luke et al. 2004) which included the
randomized controlled trials and three                 results from 8 trials (5 RCTs) came to
case series to determine if                            similar conclusions.
neurodevelopmental treatment is an
effective approach. They concluded                     Trials evaluating Neurodevelopmental
that there is no evidence that supports                techniques are summarized in Tables
neurodevelopmental treatment as                        10.2 and 10.3.
being the superior type of treatment.
We included eleven studies that

Table 10.2 Studies of Neurodevelopmental Techniques
Author/                         Methods                                        Results
PEDro score
Logigian et al.  42 stroke patients were randomly          No differences between the groups on any of
1983             assigned to receive either facilitated    the functional assessments (Barthel Index,
USA              therapy or traditional techniques for     manual muscle test). Unclear when the
4 (RCT)          remediation of motor loss in the          assessments were performed.
                 affected upper extremity. Treatment
                 duration was unclear.
Lord and Hall    39 patients 8-38 months post stroke       There was no change in the overall reported
1986             from 2 different rehabilitation centers   self-care status between the groups. Of the 4
USA              were identified retrospectively. One      upper extremity functional skill levels
No Score         of the centers used neuromuscular re-     (feeding, brushing hair, brushing teeth and
                 education therapy (NRT) (n=20) and        upper extremity dressing), NRT patients
                 the other used a traditional functional   showed slightly greater independence in
                 retraining program (TFR) (n=19). A        feeding.
                 telephone questionnaire regarding
                 current functional state was
                 administered to patients/families.
Dickstein et al. 131 stroke patients were randomized       No statistically significant differences between
1986             to receive one of three treatments 1)     the groups were reported on any of the
Israel           conventional therapy (n=57), 2)           outcome measures (Barthel Index, muscle
5 (RCT)          Proprioceptive neuromuscular              tone or active range of motion).
                 facilitation techniques (n=36) or 3)
                 Bobath techniques (n=38), for 30-45
                 min/day x 5 days/week x 6 weeks.
Basmajian et     29 hemiparetic stroke patients were       There were no differences between the
al. 1987         randomized to receive either              groups on any of the outcome measures
Canada           integrated behavioural and physical       (Upper Extremity Function Test, finger

10. Upper Extremity Interventions                                                            pg. 9 of 150
6 (RCT)        therapy (n=13) or physical therapy        oscillation tests).
               based on neuro-facilitated techniques
               (n=16), for45 min x 3 days/week x 5
Wagenaar et    7 patients alternated between 2           There were no differences in Action Research
al. 1990       therapy approaches 5-9 days post          Arm test scores between the 2 groups.
Netherlands    stroke: 1) Brunnstrom approach and
No Score       2) Neuro-developmental treatment
               (NDT). Therapies were provided for
               30 min/session for 21 weeks. Starting
               order was randomized.
Gelber et al.  20 patients with pure motor               There were no significant differences between
1995           hemiparesis following a stroke within     the groups at any of the testing intervals.
USA            the previous month were randomized        Length of hospital stay was similar for both
5 (RCT)        to neurodevelopmental technique           groups.
               (NDT) (Bobath) or traditional
               functional retraining (TRF) treatment
               approaches for the period of inpatient
               rehabilitation. FIM, Box & Block test
               and Nine Hole Peg Test were
               evaluated at admission, discharge, 6
               and 12 months.
van der Lee et In an observer blind trial, 66 patients   Mean improvement on Action Research Arm
al. 1999       were randomized to receive either         test in patients with sensory disorder was
Netherlands    forced use therapy with                   significantly greater in that receiving force
7 (RCT)        immobilization of the unaffected arm      use rather than bimanual training. During
               combined with intensive treatment or      treatment, force use patients also showed
               to receive intensive bimanual training    greater clinical significant improvement on
               based on Neuro-Development                Motor Activity Log than bimanual training
               Treatment.                                patients.
Langhammer     61 first-ever stroke patients with        Patients in the MRP group had shorter
and Stanghelle hemiparesis were block randomized         hospital stays compared to the Bobath group.
2000, 2003     into 2 groups and stratified according    Both groups improved in motor function
8 (RCT)        to gender and hemiplegic site. Group      (Motor Assessment Scale and Sodring Motor
               1 had physiotherapy according to the      Evaluation Scale), but the MRP group had
               Motor Relearning Programme (MRP)          significantly better gains than the Bobath
               and group 2 received physiotherapy        group. No differences were seen between
               according to the Bobath approach.         groups in the life quality test, use of assistive
                                                         devices or accommodation after discharge
                                                         from hospital. Follow-up 1 year and 4 years
                                                         post stroke did not reveal any major influence
                                                         of the two approaches on long-term function.
Van Vliet et al.   120 patients admitted to a stroke     There were no significant differences between
2005               rehabilitation ward were randomized   the two groups. Scores on the subsections of
UK                 to two rehabilitation approaches      both RAM and MAS associated with upper
7 (RCT)            Bobath based (BB) or movement         extremity function were similar.
                   science base (MSB). Rivermead Motor
                   Assessment (RMA) and Motor
                   Assessment Scale (MAS) scores were
                   assessed at 1,3 and 6 months.
Platz et al.       62 patients with severe arm paresis There were no differences in the mean Fugl-
2005               having suffered from a stroke       Meyer, Action Research Arm or Ashworth
Germany            between 3 weeks and 6 months        scores between the groups at the end of the

10. Upper Extremity Interventions                                                          pg. 10 of 150
8 (RCT)           previously were randomized to 3            treatment period.
                  different upper extremity regimens: i)
                  no augmented therapy (n=20), ii)
                  augmented therapy (Bobath) (n=21)
                  or iii) augmented therapy
                  (impairment –oriented training-BASIS
                  training) (n=21). The treatments
                  were provided for 4 weeks. Additional
                  therapy was provided for 45 min x 5
Hafsteinsdóttir   A controlled, multi-site cluster trial.    There were no differences in the proportion of
et al. 2005       225 patients in 6 hospitals received       patients experiencing a poor outcome
Netherlands       rehabilitation on units using the NDT      between groups. The adjusted odds ratio
No Score          (Bobath) approach and 101 patients         associated with the NDT approach was 1.7
                  on 6 wards received rehabilitation on      (95% CI: 0.8 to 3.5). There were no
                  units using a conventional (non-NDT)       differences in median QoL scores between the
                  approach. The primary outcome was          groups at 12 months.
                  a poor outcome (Barthel Index scores
                  < 12 or death) at one-year. Quality of
                  life (QoL) was also assessed.
Hafsteinsdóttir   Additional analyses from 2005 study.       There were no differences between the
et al. 2007       Health-related QoL (HRQoL) was             groups on any of the outcomes assessed at
Netherlands       measured using the SF-36;                  either 6 or 12 months. The percentage of
No Score          depression was measured with the           patients with depression at discharge, 6
                  Center of Epidemiological Studies          months and 1 year were: Bobath: 55, 40 &
                  Depression Scale and shoulder pain         30%; Conventional rehab: 46, 45 and 43%.
                  was measured with the Visual               The percentages of patients with shoulder
                  Analogue Scale at discharge, 6 and         pain were: Bobath: 18, 22 and 20%,
                  12 months.                                 Conventional rehab: 22, 28 and 19%.
Platz et al.      148 anterior circulation ischemic          At the end of follow-up, there were no
2009              stroke patients were randomly              significant differences in FM scores among
Germany           assigned to receive 45 minutes of          study groups (either groups: a vs. b/c or b
8 (RCT)           additional arm therapy daily over 3 to     vs. c). There was a significant interaction
                  4 weeks as either (a) passive therapy      effect favouring the use of IOT therapy
                  with inflatable splints or active arm      among subjects with mild paresis.
                  motor therapy as either (b)
                  individualized best conventional
                  therapy (CONV) or (c) standardized
                  impairment-oriented therapy (IOT),
                  as Arm BASIS training for severe
                  paresis or Arm Ability training for mild
                  paresis. The main outcome measures,
                  assessed at baseline, post treatment
                  and 4 weeks were the Fugl-Meyer
                  (FM) arm motor score (severely
                  paretic arms) and the TEMPA time
                  scores (mildly affected arms).

Discussion                                                   differed. Langhammer & Stanghelle
                                                             (2000) reported improvements in
The results from two recent, high                            upper extremity function and a shorter
quality RCTs assessing similar                               length of hospital stay associated with
treatment approaches and outcomes                            the motor relearning, while Van Vliet

10. Upper Extremity Interventions                                                            pg. 11 of 150
et al. (2005) did not report any                      treatment approach (BASIS arm
significant difference between                        training or Bobath) or following
treatment approaches. Van Vliet et al.                passive, conventional or impairment-
(2005) speculate that earlier, more                   oriented training (2010).
intensive training provided in the L &
S study as well as and higher (albeit                 Hafsteinsdóttir et al. (2007) reported
non-statistically significant) baseline               that the Bobath approach was not
difference may have contributed to                    superior to that of non-NDT approach.
the differences. The content of the                   There were no differences between
treatment programs within the two                     the groups on any of the outcome
studies may also have differed. Platz                 measures assessed including FIM,
et al. (2005) failed to demonstrate an                quality of life, health- related quality
effect of augmented arm therapy (in                   of life, shoulder pain or depression at
addition to regular rehabilitation) upon              up to 12 months following stroke.
motor recovery, regardless of the

Table 10.3 Summary of RCTs Evaluating Neurodevelopmental Techniques
         Author            n                  Intervention                    Main Outcome(s)
      PEDro Score                                                                  Result
Platz et al. 2005          62         No augmented therapy vs.          Fugl-Meyer arm motor score
8 (RCT)                                augmented therapy time                       (-)
                                       (Bobath) vs. augmented
                                         therapy time (BASIS)
Platz et al. 2009         148       Passive therapy (with splints)             Fugl-Meyer (-)
8 (RCT)                              vs. conventional therapy vs.                TEMPA (-)
                                    impairment-oriented training
                                      (BASIS training for severe
                                    paresis or Arm Ability training
                                            for mild paresis)
Langhammer and             61       Motor Relearning Programme             Hospital stays (+ MRP)
Stanghelle 2000,                           (MRP) vs. Bobath              Motor Assessment Scale (+
2003                                                                                 MRP)
8 (RCT)                                                                     (- at 1 and 4 yrs F/U)
                                                                       Sodring Motor Evaluation Scale
                                                                                    (+ MRP)
                                                                            (- at 1 and 4 yrs F/U)
                                                                              Life Quality Test (-)
Van Vliet et al. 2005     120        Motor Relearning Programme        Rivermead Motor Assessment (-
UK                                        (MRP) vs. Bobath                             )
7 (RCT)                                                                  Motor Assessment Scale (-)
van der Lee et al.         66        NDT vs. Forced-use therapy           Action Research Arm test
1999                                                                            (+) Forced-use
7 (RCT)
Basmajian et al. 1987      29         Physical Therapy based on        Upper Extremity Function Test
6 (RCT)                             neuro-facilitated techniques vs.                (-)
                                                  EMG                    Finger Oscillation test (-)
Gelber et al. 1995         20           Bobath vs. Traditional                    FIM (-)
5 (RCT)                                       techniques                    Box & Block test (-)
                                                                           Nine Hole Peg test (-)
                                                                                  LOS (-)

10. Upper Extremity Interventions                                                      pg. 12 of 150
Dickstein et al. 1986     131       PNF vs. Bobath vs. Traditional       Barthel Index (-)
5 (RCT)                                      techniques                   Muscle tone (-)
                                                                    Active range of motion (-)
Logigian et al. 1983        42            Facilitated therapy vs.        Barthel Index (-)
4 (RCT)                                   traditional techniques      Manual muscle test (-)
- Indicates non-statistically significant differences between treatment groups
+ Indicates statistically significant differences between treatment groups

 Conclusions Regarding                              hemisphere through neural networks
 Neurodevelopmental Techniques                      linked via the corpus callosum (Morris
                                                    et al. 2008, Summers et al. 2007).
 There is strong (Level 1a) evidence that
 neurodevelopmental techniques are not              In a systematic review which included
 superior to other therapeutic                      the results from 11 trials, Stewart et
 approaches.                                        al. (2006) reported that bilateral
                                                    movements alone or in combination
 There is moderate (Level 1b) evidence
                                                    with auxiliary sensory feedback are
 from one “good” RCT that indicates
 that when compared to the Bobath                   effective stroke rehabilitation
 treatment approach, Motor Relearning               protocols during the sub-acute and
 Programme may be associated with                   chronic phases of recovery. The
 improvements in short-term motor                   overall effect size was relatively large,
 functioning, but not long-term motor               at 0.732. A second, more conservative
 functioning and is associated with                 analysis, excluding several studies,
 shorter lengths of hospital stay.                  still produced a moderate effect size of
                                                    0.582. Another narrative review,
  Neurodevelopmental techniques are                 (Latimer et al. 2010), which included
  not superior or inferior compared with            the results from 9 studies (3 RCTs) in
  other therapeutic approaches in                   the chronic stage of stroke, also
  treatment of the hemiparetic upper                reported a benefit of bilateral training
  extremity.                                        in recovery associated with motor
10.2.2 Bilateral Arm Training
                                                    A recent Cochrane review on the
The use of bilateral training                       subject, which included the results
techniques with the upper limb                      from 18 RCTs, including 549 subjects,
following stroke has been advocated                 reported that there was no significant
recently as new theories of neural                  improvement in ADL function
plasticity have developed. Bilateral                (standardized mean difference of
arm training is a technique whereby                 0.25, 95% CI: -0.14 to 0.63),
patients practice the same activities               functional movement of the arm
with both upper limbs simultaneously.               (SMD-0.07, 95% CI -0.42 to 0.28) or
Theoretically, the use of the intact                hand, (SMD -0.04, 95% CU -0.50 to
limb helps to promote functional                    0.42) compared with usual care
recovery of the impaired limb through               following stroke (Coupar et al. 2010).
facilitative coupling effects between
the upper limbs. Practicing bilateral
movements may allow the activation
of the intact hemisphere to facilitate
the activation of the damaged

10. Upper Extremity Interventions                                                  pg. 13 of 150
Table 10.4 RCTs Evaluating Bilateral Arm Training
   Author/                      Methods                                     Results
 PEDro score
Whitall et al.   16 patients received 5-minute        Significant improvement on Fugl-Meyer Upper
2000             sessions, 3 times per week of bilateral
                                                      Extremity Motor Performance Test of
USA              arm training with rhythmic auditory  impairment, Wolf Motor Function Test,
No Score         cueing, performed with use of        University of Maryland Arm Questionnaire for
                 custom-designed arm training         Stroke was observed. Isometric strength in
                 machine and scores compared to their elbow flexion and wrist flexion for the paretic
                 pre-treatment baseline scores.       arm and in elbow flexion and wrist extension
                                                      for the non-paretic arm was noted. Active
                                                      range of motion improved for paretic-side
                                                      shoulder extension, wrist flexion and thumb
                                                      opposition and passive range of motion
                                                      significantly improved for paretic wrist flexion.
Cauraugh and 25 chronic stroke patients with mild     Patients in the bilateral training group moved
Kim 2002      to moderate paresis were randomized more blocks on the Box and Block test
USA           to receive: 1) coupled protocol of      compared to the other two groups.
5 (RCT)       EMG-triggered stimulation and
              bilateral movement (n=10); (2) EMG-
              triggered stimulation and unilateral
              movement (n=10); or (3) control
              (n=5). All participants completed 6
              hours of rehabilitation during a 2-
              week period according to group
              assignments. Motor capabilities of the
              wrist and fingers were evaluated.
Luft et al.   21 chronic stroke patients with         On fMRI there were significant changes in
2004          residual upper extremity weakness       activation in portions of the cerebrum and
USA           were randomized to receive bilateral cerebellum, for patients in the BATRAC group
7 (RCT)       arm training with rhythmic auditory     compared to control, although 3 BATRAC
              cueing (BATRAC) or to the control       patients showed no fMRI changes. There were
              condition of therapeutic exercises.     no differences in functional outcome between
              (Same intervention as Whitall et al.    groups (Fugl-Meyer, shoulder strength, elbow
              2002). Treatment sessions lasting 1     strength, Wolf weight, Wolf time of ADL).
              hour were provided 3x/week for 6
              weeks. FMRI, motor function and
              electromyography assessments were
Desrosiers et 41 inpatients, 10-60 days post stroke Although both groups improved from baseline
al. 2005      were randomized to receive a training to end of treatment, there were no significant
Canada        program focusing on symmetrical         differences in outcomes between the groups.
7 (RCT)       bilateral tasks consisting of 15-20, 45
              minute sessions, based on motor
              learning based principles, in addition
              to routine rehabilitation or to routine
              rehabilitation, based on a
              neurodevelopmental approach.
              Outcome measures included: motor
              function (Fugl-Meyer upper
              extremity), grip strength, gross

10. Upper Extremity Interventions                                                        pg. 14 of 150
                 manual dexterity (Box & Block test),
                 fine manual dexterity (Purdue
                 Pegboard test), motor coordination
                 (Finger-to-nose test), ADL (FIM)
Summers et       12 chronic stroke patients were          Individuals receiving bilateral training showed
al. 2007         randomly assigned to one of two          a reduction in movement time of the impaired
Australia        training protocols involving six daily   limb and increased upper limb functional
5 (RCT)          practice sessions. Each session          ability compared to individuals receiving
                 consisted of 50 trials of a dowel        unilateral training.
                 placement task performed either with
                 both impaired and unimpaired arm
                 moving synchronously (bilateral
                 training group) or with only the
                 impaired arm moving (unilateral
                 training). Outcomes assessed before
                 and after treatment included the
                 Modified Motor Assessment Scale.
                 Kinematic measurements of upper
                 limb movements were made in four
                 unilateral test trials performed prior
                 to and following each practice
Morris et al.    106 acute stroke patients (2-4 weeks     While subjects in both groups improved over
2008             post stroke) were randomized to          time, there were no significant differences in
UK               receive bilateral arm training (n=56)    the change scores in short-term improvement
7 (RCT)          or unilateral arm training (n=50). The   (0-6 wk) on any measure. At follow-up, (0-18
                 supervised training was provided for     wk), the only significant between-group
                 20 min 5x/week x 6 weeks. The main       difference was a change in the 9HPT and ARAT
                 outcome measure was the Action           pinch section, which was lower, indicating less
                 Research Arm Test (ARAT), which was      recovery for the bilateral training group.
                 assessed before/after treatment and      Baseline severity significantly influenced
                 at follow-up (18 weeks). Additional      improvement in all upper-limb outcomes,
                 outcomes assessed included the           irrespective of the treatment group.
                 Rivermead Motor Assessment upper-
                 limb scale, Nine-Hole Peg Test
                 (9HPT), the Modified Barthel Index,
                 Hospital Anxiety and Depression
                 Scale, and Nottingham Health Profile.
Stinear et al.   32 patients with upper limb weakness     Immediately after the intervention, motor
2008             at least 6 months after stroke were      function of the affected upper limb improved in
New Zealand      randomized to a 1-month intervention     both groups (p < 0.005). One month after the
6 (RCT)          of self-directed motor practice with     intervention, the APBT group had better upper
                 their affected upper limb (control       limb motor function than control patients.
                 group)(n=16) or to Active-Passive
                 Bilateral Therapy (APBT)(n=16),
                 using a device that mechanically
                 couples the two hands, for 10-15 min
                 prior to the same motor practice
                 (manipulating wooden blocks). Upper
                 extremity portion of the Fugl-Meyer
                 (FM) score was assessed at baseline,
                 post intervention and 1 month after
                 the intervention.

10. Upper Extremity Interventions                                                          pg. 15 of 150
Cauraugh et     30 chronic stroke subjects were            From the pretest to the posttest, both the
al. 2009        randomly assigned to one of three          coupled bilateral no load and load groups
USA             behavioral treatment groups and            moved more blocks and demonstrated more
 5 (RCT)        completed 6 hours of rehabilitation in     regularity in the sustained contraction task
                4 days: (1) coupled bilateral training     compared with the control group. Reaction
                with a load on the unimpaired hand,        times were faster across test sessions for the
                (2) coupled bilateral training with no     coupled bilateral load group.
                load on the unimpaired hand, and (3)
                control (no stimulation assistance or
                load). Both bilateral groups received
                EMG-triggered electrical stimulation.
                The Box & Block test and reaction
                times were assessed before and after
Stoykov et al. 24 subjects with moderate                   Subjects in both groups had significant
2009            impairment with stroke onset > 6           improvements on the MSS and measures of
USA             months participated in an 8-week           strength. There were no differences between
5 (RCT)         training program. Subjects were            groups on the total MSS score or either of its 2
                randomized to a bilateral group (n =       subscales (shoulder/elbow, wrist/hand). The
                12) in which they practiced bilateral      bilateral group had significantly greater
                symmetrical activities, or a unilateral    improvement on the Upper Arm Function scale
                group (n = 12) in which subjects           (a subscale of the MAS-Upper Limb Items).
                performed the same activity with the       There were no significant differences between
                affected arm only. The activities          groups on the 2 other subscales (advanced
                consisted of reaching-based tasks          hand activities, hand movements and upper
                that were both rhythmic and discrete.      arm function).
                The Motor Assessment Scale (MAS),
                Motor Status Scale (MSS), and
                muscle strength were used as
                outcome measures. Assessments
                were administered at baseline and
Lin et al. 2009 60 patients > 6 months post stroke         The CIT and BAT groups showed better
a)              with a Brunnstrom stage II or greater      performance in the overall and the distal part
Taiwan          in the proximal and distal part of the     score of the FMA than the control group. The
7 (RCT)         arm were randomized to one of 3            BAT group exhibited greater gains in the
                groups. Subjects received constraint-      proximal part score of the FMA than the
                induced therapy (CIT), bilateral arm       distributed CIT and control groups. Enhanced
                training (BAT), or a control               performance was found for the distributed CIT
                intervention of less- specific but         group in the MAL, the subtest of locomotion in
                active therapy. Each group received        the FIM, and certain domains of the SIS (eg,
                intensive training for 2 hours/day, 5      ADL/IADL).
                days/week, for 3 weeks. Outcomes
                assessed included Fugl-Meyer
                Assessment (FMA), FIM, Motor
                Activity Log (MAL), and Stroke Impact
                Scale (SIS). The proximal and distal
                scores of FMA were used to examine
                separate upper limb (UL) elements of
Lin et al. 2009 33 stroke patients, 6 to 67 months         The BAT group showed a significantly greater
b)              after onset of a first stroke and with a   improvement in the mean FMA than the
Taiwan          Brunnstrom stage II or greater in the      control group (total score at end of treatment:
6 (RCT)         proximal and distal part of the arm,       57.6 vs. 55.0, p=0.041) but not in the FIM

10. Upper Extremity Interventions                                                           pg. 16 of 150
                   were randomized to either a bilateral (118 vs. 117, p=0.18) or MAL (amount of use
                   arm training (BAT) program            subscale 1.34 vs. 1.61, p=0.12 quality of
                   concentrating on both upper           movement 1.56 vs. 1.86, p=0.17).
                   extremities moving simultaneously in
                   functional tasks by symmetric
                   patterns or a control treatment for 2
                   hours on weekdays for 3 weeks.
                   Outcome measures included the Fugl-
                   Meyer Assessment (FMA) of motor-
                   impairment severity and the
                   Functional Independence Measure
                   (FIM) and the Motor Activity Log
                   (MAL) evaluating functional ability.
The results from the RCTs presented                        stroke, at least 6 months post onset,
above are summarized in Table 10.5                         although one was specifically designed
The majority of trials were conducted                      to assess patients acutely following
on patients in the chronic stage of                        stroke (Morris et al. 2008).

Table 10.5 Summary of Studies Evaluating Bilateral Arm Training
        Author/             n               Intervention                       Main Outcome(s)
     PEDro Score                                                                    Result
Morris et al. 2008         106    Bilateral vs. unilateral training               ARAT (-)
7 (RCT)
Desrosiers et al.          41     Symmetrical bilateral tasks vs.               Fugl Meyer (-)
2005                                 conventional therapy                      Grip strength (-)
7 (RCT)
Luft et al. 2004           21         Bilateral arm training +                  Fugl-Meyer (-)
7 (RCT)                            rhythmic auditory cueing vs.
                                       therapeutic exercises.
Lin et al. 2009b           33     Bilateral vs. unilateral training            Fugl-Meyer (+)
6 (RCT)                                                                            FIM (-)
                                                                          Motor Assessment Log (-)
Stinear et al. 2008        32    Active-Passive Bilateral Therapy             (Fugl-meyer (+)
6 (RCT)                          vs. self-directed motor practice             Grip strength (-)
Stoykov et al. 2009        21     Bilateral vs. unilateral training     Motor Assessment Scale (-)
5 (RCT)                                                                    Motor Status Scale (-)
Summer s et al.            12     Unilateral vs. bilateral training Modified Motor Assessment Scale (+)
5 (RCT)
Cauraugh and Kim           25    Electrical stimulation + bilateral   Box & Block test (+ bilateral group)
2002                                   training vs. Electrical
5 (RCT)                          stimulation + unilateral training
                                             vs. control

 Conclusions Regarding Bilateral Arm                        There is conflicting (Level 4) evidence
 Training                                                   that bilateral arm training is superior to
                                                            unilateral training.

10. Upper Extremity Interventions                                                           pg. 17 of 150
                                                         additional or enhanced upper
10.2.3 Additional/Enhanced Upper                         extremity therapy, usually compared
Extremity Therapy                                        to conventional therapy. The results
In this section we included studies                      are found in Table 10.6 and 10.7.
that examined the effects of providing

Table 10.6 RCTs Evaluating Additional/Enhanced Upper Extremity Therapies
    Author/                   Methods                                        Results
 PEDro score
Trombly et al. 20 patients randomly assigned to          No significant differences were noted between
1986           receive 1 of 4 treatment conditions:      any of the groups on any of the outcome
USA            (1) resisted grasp therapy; (2)           measures.
4 (RCT)        resisted extension therapy; (3)
               ballistic extension therapy; or (4)
               therapy that did not involve affected
               hand for 20 sessions or until patient
               was discharged.
Sunderland et 132 stroke patients were randomized        Repeated measures analysis of Extended
al. 1992       to receive enhanced therapy (ET)          Motricity Index scores showed that patients in
UK             (n=67) or conventional therapy (CT)       the ET group had improved arm function
6 (RCT)        (n=65). ET consisted of Bobath            within the first month. Median Motor club
               exercises, EMG biofeedback computer       assessment and nine-hole peg test scores
               games and goal setting, for 10 weeks.     were higher for patients with mild strokes in
               Patients were divided into mild and       the ET group at six months.
               severe sub-groups.
Sunderland et One-year follow-up of 97 patients          No significant differences between enhanced
al. 1994       from 1992 study.                          therapy and conventional therapy sub-groups
UK                                                       on any of the measures at follow-up.
6 (RCT)
Butefisch et   27 hemiparetic stroke patients, 3-19      Grip strength, peak force of isometric hand
al. 1995       weeks post stroke were assigned to        extensions and peak acceleration of isotonic
Germany        an enhanced non-specific therapy          hand extensions, significantly improved during
3 (RCT)        (n=12) or to enhanced specific            training, for both groups. No between
               therapy + TENS (n=15). Both groups        treatment group statistics were reported.
               received conventional OT/PT. 2
               phase, multiple baseline study.
Dickstein et   Randomized controlled trial of 15         There were similar improvements in Barthel
al. 1997       patients who received 8 repeated          Index and Fugl-Meyer scores between the two
Israel         movement exercises for 19-21 days         groups.
3 (RCT)        as compared to controls who
               performed conventional
Kwakkel et al. 101 patients were randomized 14           At week 26, significant differences in median
1999           days following stroke to receive one      Action Research arm (ARA) scores between the
Netherlands    of 3 therapies: 1) arm training, 2)       three groups were observed. Median Barthel
8 (RCT)        leg training or 3) basic rehabilitation   Index and ARA scores of patients in both arm
               only. Leg and arm treatments were         and leg training groups were significantly
               applied for 30 min 5 days/week x 20       higher when compared to the control group.
               weeks. All patients received basic
Lincoln et al. A single blind trial of 282 patients      No significant differences between the groups

10. Upper Extremity Interventions                                                         pg. 18 of 150
1999           randomized to receive either routine     on any of the outcome measures (Rivermead
UK             physiotherapy, or additional             Motor Assessment Arm Scale, Action Research
7 (RCT)        physiotherapy (10 hrs over 5 weeks)      Arm test or Barthel Index) were observed post
               from a qualified therapist or a          intervention, at 3 or 6 month follow-up.
               physiotherapy assistant.
Platz et al.   A single blind trial of 60 patients      All patients who received ATT demonstrated a
2001           randomized to one of three groups:       significantly greater mean improvement in
Germany        Group 1 received Arm Ability Training    time needed to perform all TEMPA tasks.
7 (RCT)        (ATT) with knowledge of their results.
               Group 2 received ATT without
               knowledge of their results and Group
               3 did not receive ATT.
Bourbonnais    25 chronic stroke patients with          With the exception of the handgrip force,
et al. 2002    hemiplegia were randomized to an         strength measurements of the treated limb
Canada         upper-limb (UL) or lower-limb motor      increased after completion of the treatment.
5 (RCT)        re-education program. 13 patients        The outcome measurements of the upper limb
               with UL weakness received 3 weeks of     of the subjects included in the upper paretic
               force-feedback program 3 times a         limb were not significantly different after
               week. Patients in the LE group           treatment from those measured in the lower
               served as the control. Upper limb        paretic limb.
               performance was evaluated at 8
               weeks using the TEMPA, finger to
               nose test and the Fugl-Meyer.
Rodgers et al. 123 patients with stroke causing         There was no significant difference between
2003           upper limb impairment within the         groups on any outcome measure (Action
United         previous 10 days were randomized to      Research Arm Test, Motricity Index, Frenchay
Kingdom        either an experimental group or into a   Arm Test, upper limb pain, Barthel ADL,
7 (RCT)        control group. The experimental          Nottingham E-ADL) at 3 and 6 months after
               group received stroke unit care plus     stroke. There was no significant difference in
               enhanced upper limb therapy from         service costs between groups.
               both a physiotherapist and an
               occupational therapist commencing
               within 10 days of stroke and available
               up to 30 minutes/day, five days/week
               for 6 weeks. The control group
               received stroke unit care.
Duncan et al. A multi-centre, single-blind clinical     While gains were made in balance, gait and
2003           trial of 92 subjects randomized to       endurance, no significant gains were made in
USA            receive either a structured therapist    upper extremity function.
8 (RCT)        supervised home program of 36 90-
               minute sessions over 12- 14 weeks,
               or to the usual care group with
               services assigned by their physician
               and home visits every 2 weeks for
               health education, vital signs and a
               test of oxygen saturation.
Pang et al.    63 chronic stroke patients (≥50          There was significant improvement made for
2006           years) were randomly assigned into       the upper-extremity exercise group compared
Canada         an upper-extremity exercise program      to the lower-extremity group for the WMFT
7 (RCT)        or a lower-extremity program for 3 1-    and the Fugl-Meyer Assessment. Patients who
               hour sessions/week for 19 weeks.         benefited most from the exercise program
               Each therapy session had 9-12            where those with moderate arm deficits.
               participants and consisted of

10. Upper Extremity Interventions                                                        pg. 19 of 150
                physiotherapy, an exercise instructor
                and an occupational therapist. Main
                Outcome measures included: The
                Wolf Motor Function Test (WMFT),
                Motor Activity Log (MAL), the hand-
                held dynamometry (grip strength)
                and the Fugl-Meyer Assessment
Harris et al.   103 patients admitted for inpatient       Subjects in the GRASP group showed greater
2009            rehabilitation participated in a 4-week   improvement in upper limb function (CAHAI)
Canada          program of upper extremity therapy.       compared to the control group (mean change
8 (RCT)         Patients were randomized to either a      score: 14.1 vs. 7.9, p<0.001). The GRASP
                graded repetitive upper limb              group maintained this significant gain at 5
                supplementary program (GRASP              months poststroke. Significant differences
                group, n=53) or the control group         were also found in favor of the GRASP protocol
                (education protocol, n=50). The           for grip strength and paretic upper limb use.
                primary outcome measure was the
                Chedoke Arm and Hand Activity
                Inventory (CAHAI). Assessment was
                conducted before and after treatment
                and at 5 months post stroke.
                Secondary measures were used to
                evaluate grip strength and paretic
                upper limb use outside of therapy
Ross et al.     39 subjects adults with hand              The mean and (standard deviation) Action
2009            impairment following chronic stroke       Research Arm Test values for experimental
Australia       (90% stroke) or traumatic brain           participants improved from the beginning to
8 (RCT)         injury (10%) were randomized to an        the end of study from 10 points (15) to 21
                experimental group (n = 20) and           points (23) and the equivalent values for the
                received an additional one-hour           Summed Manual Muscle Test improved from
                session of task-specific motor training   35% (33) to 49% (35). There were similar
                for the hand 5x/week over a six-week      improvements in control participants. There
                period. The control group (n = 19)        were no significant between-group differences
                received standard care which              for either outcome.
                consisted of 10 minutes of hand
                therapy three times a week. Both
                groups continued to receive therapy
                directed at the shoulder and elbow.
                The primary outcomes were the
                Action Research Arm and Summed
                Manual Muscle Tests measured at the
                beginning and end of the six-week
Donaldson et    30 subjects with upper limb weakness      Attrition rate was 6.7% at the end of
al. 2009        and within 3 months of anterior           treatment and 40% at follow-up. Median
UK              circulation infarction were randomized    (interquartile range) increases in ARAT scores
6 (RCT)         to receive conventional physical          were 11.5 (21.0) for CPT; 8.0 (13.3) for CPT +
                therapy (CPT), CPT + CPT, and CPT +       CPT; and 19.5 (22.0) for CPT + FST. The
                functional strength training (FST).       results were not statistically significant,
                The intervention lasted for 6 weeks       although subjects in the CPT + FST group
                (24 hours in total). Primary outcome      achieved the clinically important improvement
                measure was the Action Research           of 5.7 points.
                Arm Test (ARAT), assessed

10. Upper Extremity Interventions                                                         pg. 20 of 150
                                  before/after treatment and at 12
                                                                                          conclusions difficult to draw.
Discussion                                                                                Additionally, most of the
                                                                                          interventions were non-specific in
A variety of treatments were                                                              nature. Rodgers et al. (2003)
delivered and outcomes assessed,                                                          reported no benefit of therapy
under the rubric of enhanced                                                              associated with any of the
therapy, making general                                                                   outcomes assessed (Figure 10.1).
                           Figure 10.1 M edian O utcome Measures at 6-mo nth s: Enhanced Upper Limb
                                                    Rehabilitation vs. Control
                                      Barthel Index                         p=.276
                                           OHS 3-5
     Outcome Me asure s

                                           OHS 0-2                             p=.681
                                                                                                 Intervention        Control
                                    Upper limp Pain                              p=.307
                                Nottingham E-ADL                   p=.168
                                Frenchay Arm Test            p=.679
                          Upper Limb Motricity Index
                                             ARAT                                                           p=.786

                                                       0       10      20            30     40     50           60    70        80      90

                                                                                6 month Outcome Values

Table 10.7 Summary of Studies Evaluating Enhanced/Additional Therapies
       Author/                                  n                     Intervention                           Main Outcome(s)
     PEDro Score                                                                                                  Result
Kwakkel et al. 1999                            101         Arm training vs. leg training vs.                Barthel Index (+)
8 (RCT)                                                               basic rehab                     Action Research Arm Test (+)
Ross et al. 2009                                39          Additional task-specific motor            Action Research Arm Test (-)
8 (RCT)                                                       training vs. standard care            Summed Manual Muscle Tests (-)
Harris et al. 2009                             103          Upper extremity task-specific            Chedoke Arm and Hand Activity
8 (RCT)                                                         therapy vs. education                         Inventory (+)
Duncan et al. 2003                              92          Supervised home program vs.                    Fugl-Meyer Score (-)
USA                                                                   usual care                             Grip Strength (-)
8 (RCT)                                                                                                    Functional Reach (-)
                                                                                                       Wolf Motor Function Test (-)
Lincoln et al 1999                             282           Routine physiotherapy vs.             Rivermead Motor Assessment (arm)
7 (RCT)                                                       additional physiotherapy                              (-)
                                                                                                      Action Research Arm Test (-)
                                                                                                             Barthel Index (-)
Pang et al. 2006                                63          Arm training vs. leg training                 Fugl-Meyer Score (+)
 7 (RCT)                                                                                              Wolf Motor Function Test (+)
Platz et al. 2001                               74         Arm ability training vs. routine         Tests of upper extremity function
7 (RCT)                                                                therapy                                      (+)
Rodgers et al. 2003                            123          Stroke unit care + enhanced               Action Research Arm Test (-)
7 (RCT)                                                         upper limb rehab vs.                        Motricity Index (-)

10. Upper Extremity Interventions                                                                                               pg. 21 of 150
                                 conventional stroke unit care           Frenchay Arm Test (-)
                                                                          Upper limb pain (-)
                                                                            Barthel ADL (-)
                                                                         Nottingham E-ADL (-)
Donaldson et al.         30     Conventional physical therapy        Action Research Arm Test (-)
2009                            (CPT) vs. CPT 2 vs. functional
6 (RCT)                            strength training + CPT
Sunderland et al.       132         Enhanced therapy vs.           Extended Motricity Index
1992, 1994                          conventional therapy                 (+ first month)
6 (RCT)                                                             Motor Club Assessment
                                                                  (+ mild strokes/6 months)
                                                                         9-hole Peg Test
                                                                  (+ mild strokes/6 months)
                                                                 (- at 1 year for all outcomes)
Dickstein et al. 1997    27      Repeated movement therapy              Barthel Index (-)
3 (RCT)                            vs. conventional therapy          Fugl-Meyer scores (-)
                                                                       Frenchay tests (-)
- Indicates non-statistically significant differences between treatment groups
+ Indicates statistically significant differences between treatment groups

 Conclusions Regarding
 Enhanced/Additional Therapies                            It is uncertain whether enhanced
                                                          therapy results in improved short-
 There is conflicting (Level 4) evidence                  term upper extremity functioning.
 that enhanced therapies improve short-
 term upper extremity function. There is
 evidence that results may not be long
                                                       performance following stroke. They
10.2.4 Strength Training                               identified 14 studies in total. Six
                                                       studies (306 subjects) evaluated the
A small group of studies were                          effect on grip strength. There was a
identified that evaluated treatments                   significant effect associated with
directed at increasing strength in the                 training (standardized mean
upper extremity as opposed to                          difference=0.95, 95% CI 0.05 to 1.85,
function. A much larger pool of studies                p=0.04). Two trials assessed other
has been published on strength                         measures of strength with conflicting
training in the lower extremity.                       results.

Harris & Eng (2010) conducted a                        We identified 5 studies that evaluated
systematic review and meta-analysis                    strength training and that assessed
of strength training on upper-limb                     measures of strength.
strength, function and ADL

Table 10.8 RCTs Evaluating Strength Training in the Upper Extremity
   Author/                     Methods                                    Results
 PEDro score
Trombly et al. 20 patients randomly assigned to       Significantly more subjects assigned to ballistic
1986           receive 1 of 4 treatment conditions to or resisted extension conditions improved in

10. Upper Extremity Interventions                                                        pg. 22 of 150
USA             help improve improve finger             their ability to rapidly reverse movement over
4 (RCT)         extension function: (1) resisted        the course of treatment as opposed to those
                grasp therapy; (2) resisted extension   assigned to resisted grasp or control
                therapy; (3) ballistic extension        conditions; however, there were no significant
                therapy; or (4) therapy that did not    differences were noted between any of the
                involve affected hand for 20 sessions   groups on any of the outcome measures.
                or until patient was discharged.
Butefisch et    27 hemiparetic stroke patients, 3-19    Grip strength, peak force of isometric hand
al. 1995        weeks post stroke were assigned to      extensions and peak acceleration of isotonic
Germany         an enhanced non-specific therapy        hand extensions, significantly improved during
3 (RCT)         (n=12) or to enhanced specific          training, for both groups. No between
                therapy + TENS (n=15). Both groups      treatment group statistics were reported.
                received conventional OT/PT. 2
                phase, multiple baseline study.
Badics et al.   56 chronic and acute stroke             The extensor strength of the legs
2002            patients (having sustained a            increased by 31.0% from baseline. Upper
Austria         stroke from 3 weeks to 10 years         limb strength increased by 32%. Both
No Score        previously) underwent a 4 week          increases were statistically significant for
                strength training program which         both variables. The extent of strength
                targeted both upper and lower           gain was positively correlated with the
                limbs.                                  intensity and the number of exercising
                                                        units. Muscle tone, which was abnormally
                                                        high at baseline, did not further increase
                                                        in any subjects.
Bourbonnais     25 chronic stroke patients with         With the exception of the handgrip force,
et al. 2002     hemiplegia were randomized to an        strength measurements of the treated limb
Canada          upper-limb (UL) or lower-limb motor increased after completion of the treatment.
5 (RCT)         re-education program. 13 patients       The outcome measurements of the upper limb
                with UL weakness received 3 weeks of of the subjects included in the upper paretic
                force-feedback program 3 times a        limb were not significantly different after
                week. Patients in the LE group          treatment from those measured in the lower
                served as the control. Upper limb       paretic limb.
                performance was evaluated at 8
                weeks using the TEMPA, finger to
                nose test and the Fugl-Meyer.
Carr & Jones    40 subjects with chronic stroke were The ATO group increased a mean of 3.2 ft-lbs
2003            randomly assigned to two groups, an during peak shoulder extension compared with
USA             aerobic training only (ATO) group and an increase of 3.8 ft-lbs in the A&ST group.
2 (RCT)         an aerobic and strength training        Arm flexion increased by 0.07 ft-lbs in the ATO
                (A&ST) group. Both groups were          group and by 3.7 ft-lbs in the A&ST group.
                required to exercise aerobically for 20
                minutes for 3 days a week at a
                moderate intensity for 16 weeks. The
                A&ST group also completed a series
                of eight strength-training activities.
                Outcomes were assessed before and
                after treatment.

10. Upper Extremity Interventions                                                        pg. 23 of 150
 Conclusions Regarding Strength                  There is strong (Level 1a) evidence that
 Training of the Upper Extremity                 strength training increases grip
                                                 strength following stroke.

                                               upper limbs. According to Page
10.2.4 Repetitive/Task- Specific               (2003), other, task-specific, low-
Training Techniques                            intensity regimens designed to
                                               improve use and function of the
Schmidt and Wrisberg (1999) note               affected limb have also reported
that it is well established that task-         significant improvements (Smith et al.
specific practice is required for motor        1999, Whitall et al. 2000, Winstein
learning to occur. According to                and Rose 2001).
Classen et al. (1998) focal transcranial
magnetic stimulation and functional            Barreca et al. (2003) reviewed 2
magnetic resonance imaging have                studies (Butefisch et al. 1995,
shown that task-specific training, in          Dickstein et al. 1997) of repetitive
comparison to traditional stroke               training, including repeated practice of
rehabilitation, yields long-lasting            elbow, wrist and finger flexion and
cortical reorganization specific to the        extension, and concluded that there
corresponding areas being used. More           was a positive treatment effect.
specifically, Karni et al. (1995), using
functional magnetic resonance                  A recent Cochrane review authored by
imaging, and Classen et al. (1998),            Thomas et al. (2007) evaluated the
using transcranial magnetic                    effect of task-specific training, on both
stimulation, both reported a slowly            upper and lower-extremity function.
evolving, long-term, experience-               Trials were included if one of the
dependent reorganization of the adult          intervention arms included “an active
primary motor cortex following daily           motor sequence [that] was performed
practice of task-specific motor                repetitively within a single training
activities. Also of interest is that task-     session, and where the practice was
specific sessions ie. thumb and hand           aimed towards a clear functional
movements, for as short as 15                  goal.” Eight and five RCTs respectively
minutes are also effective in inducing         were identified that assessed arm and
lasting cortical representational              hand function and their results pooled.
changes (Classen et al. 1998,                  Task-specific training was not
Butefisch et al. 1995). According to           associated with improvement in either
Page (2003) intensity alone does not           hand or arm function. The
account for the differences between            standardized mean differences were
traditional stroke and task-specific           small (0.17 and 0.16) and not
rehabilitation. For example, Galea et          statistically significant.
al. (2001) reported that stroke
patients who underwent a 3-week long           Our review included several additional
program that consisted of 45-minute            studies. The results are summarized in
task-specific, upper limb training             Tables 10.9 and 10.10.
showed improvements in measures of
motor function, dexterity, and
increased use of the more affected

10. Upper Extremity Interventions                                           pg. 24 of 150
Table 10.9 Studies Evaluating Repetitive, Task- Specific Training Techniques
    Author/                           Methods                                     Results
  PEDro Score
Woldag et al.   21 in-patients admitted an average of 7 weeks       Although Rivermead Motor
2000            following stroke (range 2-46 weeks), received       Assessment scores and grip
Germany         standard rehabilitation therapies supplemented      strength improved over the study
No score        by repetitive training of (1) grasping and          period, there was no additional
                transport movements and (2) sawing                  benefit associated with the
                movements of the affected arm over 10 minutes       additional training
                each, twice daily, five days per week following a
                baseline period.
Cauraugh & Kim 34 patients were randomly assigned to 1 of 3         At follow-up, the number of blocks
2003            treatment groups: blocked practice (same            moved (Box and Block Test) and
USA             movement is performed repetitively on               reaction time improved
6 (RCT)         successive trials) + active neuromuscular           significantly for both the blocked
                stimulation, random practice (different             and random practice groups in
                movements on successive trials) + active            comparison to the control group.
                stimulation, or no active stimulation assistance    No differences were found
                control group. Subjects completed training for      between the block and random
                90 minutes/day, 2 days/week for 2 weeks. A          practice groups.
                session consisted of 3 sets of 30 successful
                active neuromuscular trials with 3 movements
                executed 10 times/set.
Blennerhassett 30 stroke patients were randomized to either an      Only the Upper Limb Group made
& Dite          Upper Limb or Mobility Group. All subjects          a significant improvement on the
2004            received their usual rehabilitation and an          JTHFT and MAS upper arm items.
Australia       additional session of task-related practice using   The JTHFT dexterity scores in the
9 (RCT)         a circuit class form for 4 weeks. Outcome           Upper Limb Group were
                measures were assessed pre-and post-                significantly faster than the
                treatment and at six months and included three      Mobility Group.
                items of the Jebsen Taylor Hand Function Test
                (JTHFT), two arm items of the Motor
                Assessment Scale (MAS), and three mobility
                measures, the Timed Up and Go Test (TUGT),
                Step Test, and Six Minute Walk Test (6MWT).
Thielman et al. 12 patients matched using the Motor                 For data analysis, patients were
2004            Assessment Scale (MAS) were randomized to           divided into high and low
USA             receive 12 sessions (4 weeks) of home-based         functioning subgroups based on
4 (RCT)         unrestrained trunk training, while sitting          the results from pretests.
                unrestrained in a chair, using one of two           Kinematic analysis of arm
                treatments. i) Task-related training (TRT)          trajectories revealed that hand
                involved asking patients to grasp objects, which    paths of low-level subjects
                differed in size, shape and weight. ii)             straightened significantly after
                Progressive resistive exercises (PRE) involved      TRT, but not PRE. After training,
                whole-arm pulling against resistive therapeutic     high-level PRE subjects used less
                tubing in planes and distances similar to that in   trunk motion, while reaching for
                TRT. Pre- and post-test kinematic analysis of       an ipsilateral target, while high-
                arm movements MAS and Rivermead Motor               level TRT group patients showed
                Assessment scores were collected.                   no change in truck movement
                                                                    after training.
Winstein et al.   64 patients with recent stroke admitted for       Compared with SC participants,

10. Upper Extremity Interventions                                                        pg. 25 of 150
2004              inpatient rehabilitation were randomized into 1   those in the FT and ST groups had
Canada            of 3 intervention groups: Standard care (SC),     significantly greater increases in
6 (RCT)           functional task practice (FT), and strength       FM scores (P=.04) and isometric
                  training (ST). The FT and ST groups received 20   torque (P=.02) posttreatment.
                  additional hours of upper-extremity therapy       Treatment benefit was primarily in
                  beyond standard care distributed over a 4- to 6-  the less severe participants.
                  week period. The main outcome measures            Similar results were found for the
                  assessed before and after treatment included      FTHEU and isometric torque.
                  Fugl-Meyer (FM) Assessment, isometric torque,     During the long term, at 9
                  and Functional Test of the Hemiparetic Upper      months, the less severe FT group
                  Extremity (FTHUE).                                continued to make gains in
                                                                    isometric muscle torque,
                                                                    significantly exceeding those of
                                                                    the ST group (P<.05).
Higgins et al.   47 chronic stroke patients were randomized to      There were no statistically
2006             receive a 6-week program (3 sessions/week x        significant changes in any of the
Canada           90 min) of arm training (treatment condition) or outcome measures between
8 (RCT)          to leg training (control condition). Arm           groups.
                 interventions were tailored to each subject’s’
                 perceived need for improvement in various
                 tasks of ADL. Evaluations performed at baseline
                 and study end included the Box & Block test,
                 the nine-hole peg test, TEMPA, grip strength,
                 STREAM, Barthel Index, OARS-IADL, SF-36,
                 Geriatric Depression Scale
McDonnell et al. 20 subacute hemiparetic stroke patients were       Patients in both groups improved
2007             randomized to receive a course of task-specific on mean ARAT and FM scores
Australia        training with or without afferent stimulation. All although the differences were not
7 (RCT)          patients received 9 sessions of physiotherapy      statistically significant. Of the 20
                 training over 3 weeks. Prior to each training      patients, only 14 could perform
                 session, electrical stimulation of the motor point the grip-lift task, which is an
                 of 2 hand muscles was given in the stimulation objective measure of dexterity.
                 group, whereas the control group received sham Patients in the stimulation group
                 stimulation. Changes in dexterity were assessed exhibited significantly greater
                 using a grip-lift task, and standard measures of improvements in this task than
                 upper-limb function including the Action           the control group.
                 Research Arm test (ARAT) and the Fugl-Meyer
                 (FM) Assessment were made before and
                 following the intervention.
Boyd et al. 2010 18 subjects with chronic stroke were               The task-specific group made
Canada           randomized into either a task-specific group or a greater gains in RT and MT
5 (RCT)          general arm use group. Five sessions were          compared with the control group
                 completed within 2-weeks. All patients             on both the random and repeated
                 completed a serial targeting task during a fMRI sequences of testing.
                 scan on day 1 and were re-tested (retention) on
                 session 5. Three intervention sessions were
                 performed on days 2-4 in which patients in the
                 task-specific group performed serial targeting
                 practice, while patients in the general arm use
                 group underwent training sessions of increased
                 but non-task specific use of the hemiparetic
                 arm. Both groups performed a repeated
                 sequence of responses that may be learned, and
                 random sequences of movement, which cannot

10. Upper Extremity Interventions                                                         pg. 26 of 150
                   be learned. Mean reaction time (RT) and
                   movement time (MT) were calculated. RT was
                   the time from target highlight to the beginning
                   of the subject’s response. MT was movement
                   onset to target hit. A change score for was
                   calculated for RT and MT.

Many of the treatments reviewed were                   while others did not; typically there
non-specific in nature, not well                       were improvement on impairment level
described and were evaluated on                        outcomes, which did not transfer to
patients at different stages of                        functional improvements (disability
neurological recovery. Sample sizes                    level). The conclusions that we draw
were generally small. Furthermore, the                 pertain only to the basket of
interventions varied across studies                    interventions that were assessed, and
severely limiting comparability. Often,                cannot be generalized to any specific
multiple outcomes were assessed,                       treatment within the broader group.
some of which demonstrated a benefit,

Table 10.10 RCTs of Repetitive Task- Specific Techniques for the Upper Extremity
      Author           n                Intervention                           Main Outcome(s)
    PEDro Score                                                                     Result
Blennerhassett &       30     Upper or lower extremity task-         Jebsen Taylor Hand Function Test (+)
Dite 2004                    related practice -1 hour a day x            Motor Assessment Scale (+)
9 (RCT)                              5 days x 4 weeks
Higgins et al.         47     Upper or lower extremity task-                 Box & Block test (-)
2006                           related practice: 90 min x 3
8 (RCT)                          sessions/week x 6 weeks
McDonnell et al.       20      Task-specific training with or                  Fugl-Meyer (-)
2007                            without afferent stimulation                      ARAT (-)
7 (RCT)                                                                         Dexterity (+)
Cauraugh & Kim         34       Blocked practice + active                  Box and Block Test (+)
al. 2003                    stimulation vs. random practice                  Reaction time (+)
6 (RCT)                   + active stimulation vs. no active
                            stimulation assistance (control)
Winstein et al.     64 Standard care (SC) vs. functional             Fugl-Meyer (+ FT/ST vs. SC)
2004                         task practice (FT) vs. strength   Functional Test of the Hemiparetic Upper
6 (RCT)                               training (ST)                   Extremity (+ FT/ST vs. SC)
Boyd et al. 2010    18     Task-specific training vs. general   Changes in reaction and movement time
5 (RCT)                               arm training                                (+)
Thielman et al.     12          Task-related training vs.     Kinematic analysis of arm movements (+/-)
2004                         Progressive resistive exercises                    MAS (-)
4 (RCT)                                                            Rivermead Motor Assessment (-)
- Indicates non-statistically significant differences between treatment groups
+ Indicates statistically significant differences between treatment groups

 Conclusions Regarding Repetitive Task                   techniques improve measures of upper
 Specific Techniques                                     extremity function.

 There is conflicting (Level 4) evidence                  Repetitive task-specific training may
 that repetitive task specific training                   improve upper extremity function.

10. Upper Extremity Interventions                                                          pg. 27 of 150
10.2.5 Truck Restraint                                movements may encourage recovery
                                                      of “normal” reaching patterns in the
Reaching movements made with the                      hemiparetic arm when reaching for
affected arm in hemiparetic patients                  objects placed within arm’s length
are often accompanied by                              (Michaelson et al. 2004). Several trials
compensatory trunk or shoulder girdle                 have evaluated the effectiveness of
movements, which extend the reach of                  trunck restraint combined with task-
the arm (Michaelson et al. 2001).                     specific training to improve the
Restriction of compensatory trunk                     movement quality of reaching tasks.

Table 10.11 Studies Evaluating Trunk Restraint
     Author/                         Methods                                   Results
  PEDro Score
Michaelsen &     28 chronic, hemiparetic stroke patients were     The trunk restraint group used
Levin            randomized to a trunk restraint group with       more elbow extension, less
2004             practiced reach-to-grasp movement tasks or to    anterior trunk displacement, and
Canada           un restrained group, with verbal instruction not had better interjoint coordination
5 (RCT)          to move trunk. Training consisted of 60 trials.  than the control group after
                 Kinematics of reaching and grasping an object    training, and range of motion was
                 placed within arm's length were recorded         maintained 24 hours later in only
                 before, immediately after and 24 hours after     the trunk restraint group.
                 training.                                        Significant between group
                                                                  differences for changes in trunk
                                                                  displacement (mm) and elbow
Michaelsen el al. 30 chronic stroke patients with hemiparesis     The TR training group experienced
2006              were randomized into a Trunk-restraint (TR) or larger decreases in impairment
Canada            non-restraint group. Patients received 3 1-hour (Fugl-Meyer Arm Section;
7 (RCT)           sessions per week with object-related reach-to- p<0.035) and greater gains in
                  grasp training supervised by a therapist for 5  function (Upper Extremity
                  weeks. Outcome measures included: motor         Performance Test; p<0.05)
                  function (Upper Extremity Performance Test),    compared with non-restraint at
                  effectiveness for improving arm impairment      follow-up. Both groups showed
                  (Fugl-Meyer Arm Section) and Isometric force    significant improvements for
                  and manual dexterity (Box and Blocks Test).     elbow strength (p<0.002), Box
                                                                  and Blocks Test (p<0.01), peak
                                                                  velocity (p<0.002), trajectory
                                                                  smoothness (p<0.001) and
                                                                  straightness (p<0.01).
Woodbury et al. 11 chronic stroke patients with baseline Fugl-    Mean FMA scores improved from
2009              Meyer Assessment (FMA) scores 26 to 54 were 38 to 49 in the trunk restraint
USA               randomized to 2 constraint-therapy intervention group and from 42 to 46 in the
5 (RCT)           groups. All participants wore a mitt on the     control group. WMFT scores
                  unaffected hand for 90% of waking hours over    improved from 9.41 to 5.16 in the
                  14 days and participated in 10 days/6 hours/day trunk restraint group and from
                  of supervised progressive task practice. During 9.35 to 6.16 in the control group.
                  supervised sessions, one group trained with a   Posttraining, the trunk-restraint
                  trunk restraint (preventing anterior trunk      group demonstrated significantly
                  motion) and one group did not. Outcome          straighter reach trajectories and
                  measures included FMA, Wolf Motor Function      less trunk displacement. The
                  Test (WMFT) Motor Assessment Log (MAL)          trunk-restraint group achieved

10. Upper Extremity Interventions                                                     pg. 28 of 150
                 Amount of Use (AOU) and Quality of Movement      significantly greater gains in
                 (QOM), kinematics of unrestrained targeted       shoulder flexion and elbow
                 reaching and tests of functional arm ability     extension
                 assessed before and after treatment.

The majority of the trials examing                    included trunck restraint combined
interventions to improve reaching                     with constraint-induced movement
tasks evaluated trunck restraint                      therapy.
+task-specific therapy, although one

Table 10.12 RCTs Examining Truck Restraint to Improve Reaching
      Author         n               Intervention                         Main Outcome(s)
   PEDro Score                                                                 Result
Michaelsen el al.    30     1 hr sessions 3x /wk x 5 wks        Upper Extremity Performance Test (+)
2006                        with object-related reach-to-           Fugl-Meyer Arm Section (+)
Canada                    grasp training for Trunk-restraint           Box and Blocks Test (-)
7 (RCT)                           vs. non-restraint
Woodbury et al.      11   CIMT + trunk restraint vs. CIMT                  Fugl-Meyer (-)
2009                                                                 Wolf Motor Function test (-)
5 (RCT)                                                          Kinematic analyses of reaching (+)
Michaelsen & Levin   28     Trunk restraint group vs. no               Trunk displacement (+)
2004                                 restraint                   Performance outcome measures (-)
5 (RCT)

 Conclusions Regarding Interventions                   There is conflicting (Level 4) evidence
 designed to Improve Reaching                          that specialized programs improve

                                                      improvement at the end of the
10.2.6 Sensorimotor Training and                      treatment phase compared to the
Somatosensory Stimulation                             control group. This improvement was
                                                      still seen at follow-up 12 months later.
Somatosensory deficit is common
following stroke. Connell et al. (2008)               A review of sensory-motor training by
reported that among 70 patients with                  Steultjens et al. (2003) included three
first-ever stroke, 7-53% had impaired                 RCTs (Feys et al. 1998, Jongbloed et
tactile sensations, 31-89% impaired                   al. 1989, Kwakkel et al. 1999), one
stereognosis, and 34-64% impaired                     case control trial (Turton and Fraser
proprioception. Sensorimotor                          1990), and one noncontrolled trial
stimulation can be applied by a                       (Whitall et al. 2000). The authors
number of methods ie. an inflatable                   concluded that sensorimotor training
splint or a robotic device that moves                 was not effective for improving ADLs,
the paretic limb. Barreca et al. (2003)               extended ADLs, social participation, or
included four studies in their review of              arm and hand function. In a more
sensorimotor training for the upper                   recent review, including the results of
extremity (Feys et al. 1998, Jongbloed                14 RCTs (Schabrun & Hiller 2009), the
et al. 1989, Volpe et al. 1999, 2000).                authors distinguished between passive
The authors concluded that stroke                     forms of sensory retraining through
survivors who obtained sensorimotor                   electrical stimulation (TENS) and
stimulation showed more                               active forms, primarily through

10. Upper Extremity Interventions                                                       pg. 29 of 150
specific exercises. The included trials                  In the following section we also
assessed the outcomes of function,                       include studies that evaluated
sensation and prorioception in both                      electrical somatosensory stimulation in
the upper and lower extremity. Only 2                    an effort to improve motor function.
of the included trials assessed                          This technique has been used
sensation in the upper extremity. The                    successfully in animals with
results were ambiguous. A recent                         demonstrated effects on motor cortical
Cochrane review (Doyle et al. 2010)                      function that outlast the period of
included the results from 13 studies                     stimulation. However, it remains
(467 participants) examining a variety                   unknown whether somatosensory
of treatments for sensory impairment                     stimulation associated with training
following stroke and concluded that                      can produce longer-lasting behavioural
there was insufficient high-quality                      gains in patients with subcortical
evidence available to recommend the                      strokes.
use of any of them. Treatments with
preliminary evidence of benefit include                  The results are summarized in Tables
mirror therapy, thermal stimulation                      10.13 and 10.14. Sensorimotor
and intermittent pneumatic                               training involving robotics is included
compression.                                             in a separate section.

Table 10.13 Studies Evaluating Sensorimotor Training or Somatosensory Stimulation
 Author, Year                  Methods                                       Results
 PEDro score
Jongbloed et    90 stroke patients were randomized       No significant differences between the two
al. 1989        to receive either sensorimotor           groups on the three outcome measures
Canada          stimulation by an OT (40 min/day x5      (Barthel Index, meal preparation and 8
5 (RCT)         days/week x 8 weeks) (n=43) or           subsets of the Sensorimotor Integration test).
                functional therapeutic approach
                (n=47). Mean of 40 days post stroke
                to admission to study
Yekutiel and    20 patients received 45 minutes of       Re-training group showed large and significant
Guttman.        systematic sensory re-training 3         gains on all sensory tests – location of touch,
1993            times a week for 6 weeks and were        sense of elbow position, two-point
Israel          compared to 19 non-treatment             discrimination and stereognosis, while no
No Score        controls.                                change occurred in the control group.
Feys et al.     Single blind multi-centre trial of 100   There were no differences in Fugl-Meyer scores
1998            patients randomized to either a          between the groups at 6 wks. Patients in the
Belgium         treatment or control group. 2-5          experimental group scored significantly higher
6 (RCT)         weeks after stroke onset, patients in    at 6 month and 12 month follow-up. No
                the treatment group received             significant differences between the groups in
                additional sensorimotor treatment for    Arm Research Action or Barthel Index scores.
                30 minutes, 5 days a week for 6          Patients in both groups improved significantly
                weeks while patients in the control      over time in FM, ARA and BI scores.
                group received a placebo treatment.
Conforto et al. 8 chronic stroke patients participated   A significant increase in pinch muscle strength
2002            in two 2-hour sessions, randomly         was observed during the MNS sessions;
USA             ordered, in which stimulation was        however, no significant changes were noted in
6 (RCT)         delivered to wrist by (1) median         the CS sessions.
                nerve stimulation (MNS), in which,

10. Upper Extremity Interventions                                                         pg. 30 of 150
                stimulus intensity increased until
                patients reported strong paresthesias
                in the median nerve territory in the
                absence of pain, and (2) control
                simulation, in which, stimulus
                intensity kept immediately below that
                required to elicit paresthesias.
Cambier et al. 23 patients were enrolled in a multi-     While both groups demonstrated significant
2003            centre randomized controlled             improvements in Nottingham Sensory
Belgium         preliminary trial that compared the      Assessment scores from baseline to the end of
7 (RCT)         application of intermittent pneumatic    the study, patients in the treatment group
                compression with a passive treatment     scored significantly higher than the control
                strategy. All patients received NDT      group. Patients in the experimental group also
                therapy. The experimental group          showed significantly higher scores on the Fugl-
                received an additional 30 min of         Meyer Assessment.
                sensorimotor stimulation therapy 5
                days/wk, for 4 wks, while the control
                group received sham short-wave
                therapy for the same amount of time.
Byl et al. 2003 21 subjects who were 6 months to 7       Subjects with right-sided hemiparesis
USA             yrs post stroke and able to walk 100     improved significantly more than those with a
6 (RCT)         feet with or without a cane; partially   left hemiparesis in terms of functional
                open and close the hand; and             independence (p<0.002), sensory
                partially elevate the shoulder and       discrimination (p<0.05), and gait speed
                elbow against gravity. Eligible          p<0.05). Across both groups, more than 20%
                patients were then randomly assigned     (P < 0.01) improvement was measured in
                to Group A (sensory training 4 wks,      functional independence and UE function (fine
                motor training 4 wks) or Group B         motor, sensory discrimination, and
                (motor training 4 wks, sensory           musculoskeletal performance).
                training 4 wks). 18 patients
                completed the study.
Chen et al.     46 acute stroke patients were            29 patients completed the study. The
2005            randomly assigned to standard            performance of Brunnstrom stage and wrist
Taiwan          rehabilitation treatment or standard     extension and sensation were improved
7 (RCT)         treatment plus thermal stimulation       significantly after TS intervention. Recovery
                (TS) for 30 minutes daily for 6 weeks.   rates of 6 measures after TS were significantly
                Outcome measures included:               higher than those of the control, except for
                Brunnstrom stage, modified motor         grasping.
                assessment scale, grasping strength,
                angles of wrist extension and flexion,
                sensation by monofilament, and
                muscle tone by modified Ashworth
                scale. Assessments were performed
                weekly to evaluate sensory and motor
                functional outcomes.
Hummel et al. Crossover trial whereby 6 chronic          Five patients completed the two treatment
2005            stroke patients received 1 session of    sessions. Following treatment there was
Germany         non-invasive, painless cortical          significant improvement in the paretic limb
 6 (RCT)        stimulation by transcranial direct       that received tDCS, but not in the sham
                current stimulation (tDCS) and one       treated arm.
                sham session. Outcome was assessed
                by the Jebsen-Taylor Hand Function
                test (JTT).

10. Upper Extremity Interventions                                                         pg. 31 of 150
Mann et al.     22 patients between 1 and 12 months       At the end of 12 and 24 weeks, there was a
2005            post stroke were randomized to            significant difference in ARAT scores between
UK              receive NMES of the elbow, wrist and      groups, favouring the NMES group, although
5 (RCT)         finger for 30 min a day, 2x/day for 12    there were no differences in sensation
                weeks or passive extension exercises.     assessments between groups.
                Outcomes were assessed before,
                midway and after treatment and at 24
                weeks and included the ARAT for
                motor function and static 2-point
                discrimination for sensation.
Sawaki et al.   7 chronic stroke patients participated    Use-dependent plasticity was more prominent
2006            in 3 training sessions, randomly          with arm stimulation (increased by 22.8%)
Switzerland     ordered and separated by at least 24      than with idle time (increased by 2.9%) or leg
5 (RCT)         hrs between treatments. Use-              stimulation (increased by 6.4%).
                dependent plasticity was tested after
                2 hrs of stimulation of: i) the ulnar,
                median and radial nerves of the
                paretic hand; ii) tibial, superficial
                peroneal and sural nerves in the
                paretic leg, and iii) no stimulation.
                Movement threshold, amplitude of
                motor-evoked potential and training
                kinematics were analyzed.
Wu et al.       9 chronic stroke patients participated    JTHFT time was shorter following hand
2006            in a crossover study with 3 treatment     stimulation compared to either leg stimulation
USA             conditions. Patients received a single    or no stimulation treatment conditions. The
6 (RCT)         2-hour peripheral nerve stimulation       greatest improvement occurred in patients
                session, separated by a 24-hour           with the greatest level of impairment.
                period, of the affected arm and hand
                and no stimulation. Outcome was
                assessed by the Jebsen-Taylor Hand
                Function Test (JTHFT) before and
                after each treatment.
Yozbatiran et   36 acute stroke patients received 1       Patients in both groups improved; however,
al. 2006        h/day neurodevelopmental                  the results of between group comparisons are
Turkey          rehabilitation for 10 days. 18 patients   not reported.
No Score        were assigned to receive additional
                treatment with TENS of the wrist and
                finger extensors for 1 h. Outcomes
                assessed before and after treatment
                included kinaesthesia and position
                sense tests, a hand function test and
                a hand movement scale.
Celnik et al.   9 patients with subcortical stroke with   Reduction in JTHFT time at 1 and 24 hrs
2007            onset of at least 1-year participated     relative to pretest were only significant in the
Germany         in this crossover designed study.         nerve stimulation group. There were no
6 (RCT)         After a familiarization session each      differences among conditions with respect to
                subject returned to take part in          reductions in fatigue, attention or perceived
                sessions 2,3 and 4. These sessions        difficulty.
                each performed on a different day
                started with the Jebsen-Taylor Hand
                Function Test (JTHFT) after which
                subjects received 2 hours of nerve

10. Upper Extremity Interventions                                                            pg. 32 of 150
                stimulation (ulnar and median nerves
                of the affected hand), no stimulation
                or asynchronous nerve stimulation.
                The JTHFT was assessed 1 and 24 hrs
                later. Subjective reports of fatigue,
                attention, perceived difficulty of task
                performance were assessed using
Conforto et al. 11 patients with chronic cortical         Improvement in performance in the JTT after
2007            stroke participated in a crossover        somatosensory stimulation and after motor
Brazil          study whereby subjects received 2         training was significantly greater in the MNS
No Score        hrs of somatosensory stimulation in       session than in the CS session. Patients who
                the form of median nerve stimulation      received MNS in the second session
                at 2 different intensities                maintained the beneficial effects of training 30
                (suprathreshold and subthresthold)        days later.
                and a control condition of sham
                stimulation. The interval between
                treatments was at least 60 days. The
                primary outcome was the Jebsen-
                Taylor test (JTT).
Sullivan &      10 subjects with onset of stroke          6 subjects demonstrated a statistically
Hederman        between 2 and 6 years previously          significant improvement on the ARAT, 5 on
2007            participated in an 8-week,                STREAM and 4 demonstrated a 10% or greater
USA             individualized, home programme of         improvement in spascticity (MAS). Two
No Score        neuromuscular and sensory amplitude       subjects demonstrated significant
                electrical stimulation. All subjects      improvement on all three outcome measures.
                engaged in stimulation-assisted task-     Nine subjects demonstrated sensory deficits at
                specific exercises for 15 minutes, 2 -3   baseline. Among the 6 subjects who were
                times daily. Participants with sensory    tested at the end of the intervention, 4
                deficits received an additional 15        demonstrated improvements in NSA scores.
                minutes of sensory amplitude
                stimulation twice daily. The following
                outcomes were assessed before and
                after treatment: The Action Research
                Arm Test (ARAT), the Stroke
                Rehabilitation Assessment of
                Movement (STREAM) (to examine
                movement quality) and the Modified
                Ashworth Assessment of Spasticity
                (MAS). The Nottingham
                Stereogennosis Assessment (NSA)
                was used to examine sensation.
Volpe et al.    21 chronic stroke patients were           Patients in both groups demonstrated
2008            randomized to receive a course of         improvement over time, which was maintained
USA             intensive upper-extremity treatment       at 3 months; however, there were no
5 (RCT)         that was provided by either a             significant between group differences on either
                therapist or a robotic device             the primary or secondary outcomes.
                (InMotion2). Treatment consisted of 1
                hr sessions, 3x/week for 6 weeks.
                Primary outcome was the Fugl-Meyer
                (FM) score for shoulder/elbow.
                Secondary outcomes were the FM
                wrist/hand and the Motor Power Scale
                for Shoulder/elbow. A variety of other

10. Upper Extremity Interventions                                                           pg. 33 of 150
               outcomes were also assessed.
               Assessments were conducted monthly
               for 3 months.
Byl et al 2008 45 subjects, an average of 2.3 years     Subjects in group III achieved greater gains
USA            post stroke participated in a 6-8 week   than subjects in either groups I or II on all 4
6 (RCT)        learning-based sensorimotor training     primary outcome variables. Across all
               (LBSMT) program of varied dosage:        individual subjects, significant gains were
               group I (n = 18; 1x/week, 1.5            measured on the 4 dependent variables
               hours/visit); group II (n = 19,          (improvement ranging from 9.0% to 38.9%).
               3x/week, 0.75 hours/visit); and group
               III (n = 8; 4x/week, 3 hours/visit).
               All subjects reinforced their training
               with home-based practice. The
               primary outcome measures, assessed
               before and after treatment, were
               functional independence, strength,
               sensory discrimination, and fine
               motor skills.
Hesse et al.   8 sub-acute stroke patients were         The mean initial/final FM scores for the
2008           randomised to receive additional         treatment and control group were 11.3/26.5
Germany        therapy with the Finger Trainer (a       vs. 10.5/18.5. Only 1 subject (experimental
7 (RCT)        device for repeating controlled          group) was able to move any blocks following
               passive movements of paralysed           treatment. Median Modified Ashworth score
               fingers) for 20 min every work day for   increased from 0/5 to 2/5 in the control group,
               four weeks, or the same duration of      but not in the treatment group, 0 to 0. Only
               bimanual group therapy, in addition      one patient, in the treatment group, regained
               to their usual rehabilitation.           function of the affected hand.
               Assessments conducted before and
               after treatment included the Fugl
               Meyer (FM) Assessment, the Box and
               Block test and Modified Ashworth
Barker et al.  42 stroke survivors with severe and      There was a significant treatment effect
2008           chronic paresis were randomized to       associated with both of the SMART arm
Australia      receive: i) Sensorimotor Active          groups, but not for the control group. Median
8 (RCT)        Rehabilitation Training (SMART) Arm      scores on item 6 of the MAS improved from: 0
               (n=10) with electromyography-            to 2 (SMART arm + stim); 1 to 3 (SMART arm
               triggered electrical stimulation;ii)     with no stim) and remained at 1 over the
               SMART Arm alone (n=13) or iii) no        intervention period among subjects in the
               intervention (control)(n=10). Training   control group.
               consisted of 12 1-hour sessions over
               4 weeks. The primary outcome
               measure was "upper arm function,"
               item 6 of the Motor Assessment Scale
               (MAS). Secondary outcome measures
               included impairment measures;
               triceps muscle strength, reaching
               force, modified Ashworth scale; and
               activity measures: reaching distance
               and Motor Assessment Scale.
               Assessments were administered
               before (0 weeks) and after training (4
               weeks) and at 2 months follow-up (12

10. Upper Extremity Interventions                                                         pg. 34 of 150
Klaiput et al.  20 stroke patients with stroke onset      There were no changes in either group in ARAT
2009            less than 6 months previously, who        scores from beginning to end of treatment.
Thailand        could voluntarily pinch the thumb to      Scores were 56.2 in real stimulation group and
8 (RCT)         the index finger were randomized to       56.9 in sham group. There were significant
                to receive 2 hours of real (10Hz- to      between group differences favouring the real
                the level of appreciating paresthesias)   stimulation group in lateral and tip pinch
                or sham (stimulation to the level of      strength. Mean lateral pinch strength of real
                perception only) electrical stimulation   stimulation and sham groups was 1.24 +/-
                over the median and ulnar nerves at       0.54 and 0.20 +/- 0.28 pounds, respectively.
                the wrist. Pinch strength of the          Mean increase tip pinch strength of real and
                thumb pad to tip and to lateral side of   sham groups were 1.00 +/- 0.72 and 0.37 +/-
                the index finger of the paretic hand      0.36 pounds, respectively.
                and the Action Research Arm test
                were tested before and immediately
                after the stimulation.
Conforto et al. 22 patients were pseudo-randomized        At the end of treatment, JTT scores in the
2010            within the second month after stroke      subsensory group had improved significantly
Brazil          to receive application of 2-hour          greater campared with the suprasensory group
No Score        repetitive peripheral nerve sensory       (subsensory: 114 to 49.5 vs. suprasensory: 93
                stimulation (RPSS) at 1 of 2 stimulus     to 61.4, p=0.026). The differences between
                intensities (subsensory and               groups were no longer signifcant at 2-3
                suprasensory) immediately preceding       months. There were no differences between
                motor training, 3 times a week, for 1     groups in terms of pinch force or FIM scores at
                month. All patients received              either the end of treatment, or at follow-up.
                conventional rehabilitation therapies
                on an outpatient basis, once a week.
                Jebsen-Taylor test (JTT), pinch force
                and FIM, were measured before and
                after the end of the treatment month
                and 2-3 months later.

Discussion                                                2003, 2009, Jongbloed et al. 1989),
                                                          which in one case was delivered by a
There was a broad range of                                robotic device (Volpe et al. 2008). The
interventions provided, which                             interventions were evaluated in the
complicated the process of formulating                    acute (Chen et al. 2005), subacute
conclusions. Among the RCTs,                              (Feys et al. 1998, Cambier et al. 2003,
sensorimotor stimulation treatment                        Jongbloed et al. 1989, Hesse et al.
included thermal stimulation (Chen et                     2008, Klaiput et al. 2009) and chronic
al. 2005), intermittent pneumatic                         (Byl et al. 2003, 2009) stages of
compression (Cambier et al. 2003),                        stroke. Several RCTs also evaluated
splinting (Feys et al. 1998), cortical                    the effects of electrical somatosensory
stimulation (Hummel et al. 2005), and                     stimulation.
sensory training programs (Byl et al.

Table 10.14 Summary of Results From RCTs Evaluating Sensorimotor Training or
      Author/          n              Intervention                         Main Outcome(s)
    PEDro Score                                                                 Result
Barker et al. 2008     42     Non-robot training device                       MAS (+)
8 (RCT)                       with stimulation vs. device
                                    only vs. control

10. Upper Extremity Interventions                                                          pg. 35 of 150
Klaiput et al. 2009   20      Real vs. sham electrical             Action research Arm test (-)
8 (RCT)                             stimulation
Cambier et al.        23      Intermittent pneumatic           Nottingham Sensory Assessment (+)
2003                        compression vs. sham short-             Fugl-Meyer Assessment (+)
7 (RCT)                            wave therapy                           Ashworth Scale (-)
                                                                      Visual Analogue Scale (-)
Chen et al. 2005      46        Thermal stimulation                        Brunnstrom (+)
7 (RCT)                                                                       MMAS (-)
                                                                             Grasping (-)
                                                                            Sensation (+)
Feys et al. 1998      100   Sensorimotor stimulation vs.                Fugl-Myer Assessment
6 (RCT)                               control                          (+ at 6 and 12 months)
                                                                    Action Research Arm test (-)
                                                                           Barthel Index (-)
Byl et al. 2008       45    Leaning-based sensorimotor          Functional independence, strength,
USA                          training (3 intensity levels)    sensory discrimination, fine motor skills
6 (RCT)                                                       (+ for highest intensity group compared
                                                                             with other 2)
Byl et al. 2003       18    Sensory training for the first          Functional Independence (-)
USA                           4 wks, followed by motor       (+ patients with right hemiparesis vs. left
6 (RCT)                     training for 4 wks vs. motor                     hemiparesis)
                            training 4 wks, then sensory             Sensory discrimination (-)
                                    training 4 wks           (+ patients with right hemiparesis vs. left
                                                                              Gait speed
                                                             (+ patients with right hemiparesis vs. left
                                                                            Fine Motor (-)
Conforto et al.       8        Single session of medial              Pinch muscle strength (+)
2002                         nerve stimulation vs. sham
6 (RCT)                               stimulation
Wu et al. (2006)      9      Single session of peripheral     Jebsen-Taylor Hand Function Test (+)
6 (RCT)                        nerve stimulation vs. no
Celnik et al. 2007    9      Single session of peripheral     Jebsen-Taylor Hand Function Test (+)
6 (RCT)                        nerve stimulation vs. no
                            stimulation vs. asynchronous
                                  nerve stimulation
Hummel et al.         5     cortical stimulation vs. sham      Jebsen-Taylor Hand Function test (+)
2005 Germany                          stimulation
6 (RCT)
Jongbloed et al.      90       Functional approach vs.                 Barthel Index (-)
1989                           sensorimotor integrative   8 subsets of the Sensorimotor Integration
5 (RCT)                               approach                              test (-)
Volpe et al. 2008     21      Sensorimotor arm training    2 subsections of the FM Assessment (-)
5 (RCT)                      delivered by a therapist vs.
                                    robotic device
- Indicates non-statistically significant differences between treatment groups
+ Indicates statistically significant differences between treatment groups

 Conclusions Regarding Sensorimotor

10. Upper Extremity Interventions                                                         pg. 36 of 150
 There is conflicting (Level 4) evidence      patient characteristics, interventions
 that sensorimotor training delivered by      and outcomes assessed of the included
 a therapist improves upper extremity         studies were sufficiently
 function, compared to traditional            heterogeneous to preclude pooled
 techniques.                                  analysis. The authors were unable to
                                              draw conclusions based on the
 There is strong (Level 1a) evidence that     available evidence and suggested more
 electrical somatosensory stimulation
                                              research is needed.
 improves hand motor function.
                                              More recently, Zimmerman-Schlatter
  It is uncertain whether sensorimotor        et al. (2008) also assessed the efficacy
  training results in improved upper          of motor imagery in recovery post
  extremity functioning.                      stroke. Theses authors included the
                                              results from only 4 RCTs (Liu et al.
10.2.7 Mental Practice                        2004, Page et al. 2001, 2006 2007) in
                                              which the duration and frequency of
The use of mental imagery or mental           treatment lasted from 10 minutes to
practice as a means to enhance                one-hour a day, with 3 to 5 sessions
performance following stroke was              per week for 3 to 6 weeks. Mean time
adapted from the field of sports              of stroke onset ranged from several
psychology were the technique has             days to several years. Three of these
been shown to improve athletic                studies reported improvements in the
performance, when used as an adjunct          mean Action Research Arm Test and
to standard training methods. The             Fugl-Meyer scores. Two of these
technique, as the name suggests,              studies also found higher mean change
involves rehearsing a specific task or        scores than the minimally clinically
series of tasks, mentally. A series of        relevant difference in the ARAT and FM
small trials have adapted and                 scores. These authors concluded that
evaluated the effects of mental               although there was evidence of benefit
practice as a treatment following             of treatment, larger and more rigorous
stroke. The ability of the treatment to       studies are required to confirm these
improve motor function or ADL                 findings. A study protocol which aims
performance is the outcome most               to evaluate the effectiveness and cost-
frequently assessed in these studies.         effectiveness of this treatment has
The most plausible mechanism to               recently been published (Verbunt et al.
explain the success of the technique is       2008).
that stored motor plans for executing
movements can be accessed and                 The details of all studies from this
reinforced during mental practice             review with a sample size greater than
(Page et al. 2001). Mental practice can       2 are summarized in table 10.15.
be used to supplement conventional
therapy and can be used at any stage
of recovery.

A systematic review (Braun et al.
2006) included the results from 10
studies, three of which were RCTs.
Sample sizes ranged from 1 to 46. The

10. Upper Extremity Interventions                                         pg. 37 of 150
Table 10.15 Studies Evaluating the Use of Mental Practice Following Stroke
     Author/                    Methods                                     Results
  PEDro Score
Page et al. 2000 16 chronic stroke patients were        Patients receiving IT demonstrated
USA              randomized to receive OT +             significantly greater improvement in Fugl-
4 (RCT)          imagery training (IT) (n=8) or to      Meyer scores over the treatment period,
                 OT (n=8) and received a ½ hr           compared with controls. The associated effect
                 session 3 x /week for 4 weeks as       size was 1.39.
                 outpatients. Patients in the imagery
                 group received a tape-recorded
                 guided mental imagery session
                 which lasted for 20 min. Patients in
                 the control group listened to tape-
                 recorded message on general stoke
                 information. Therapy was provided
                 on an outpatient basis.
Page et al. 2001 13 stroke patients (stroke onset 4     No inferential statistics were reported. Patients
USA              weeks to 1 year) with stable motor     in the IT group had greater improvements in
5 (RCT)          deficits were randomized to receive    mean Fugl-Meyer and Action Research Arm
                 either OT +imagery training            tests, compared with patients in the control
                 (IT)(n=8) or OT (n=5) and received     group (FM changes: + 13.8 vs. + 2.9; ARA
                 one-hour sessions 3x/week for 6        changes: 16.4 vs. + 0.7).
                 weeks. Therapy sessions focused
                 equally on upper and lower
                 extremities. Therapy was provided
                 on an outpatient basis.
Dijkerman et al. 20 chronic stroke patients           Groups 2 and 3 were pooled for statistical
2004             performed a reach and grab task on   analysis. At the end of 4 weeks, there were no
UK               a daily basis for 4 weeks.           statistically significant differences between
No Score         Additionally, 10 patients performed  group on: BI scores, Hospital Anxiety and
                 the task mentally (group 1). Five    Depression scale, or Recovery of Locus
                 patients practiced a visual imagery  Control. While there was a significant
                 task, recalling a set of pictures    difference in the mean scores of Functional
                 (group 2) and 5 patients did not     Limitations Profile, this was due to
                 practice any mental imagery (group   deterioration within the control group and not
                 3).                                  to improvements among patients in the
                                                      treatment group. There was significant
                                                      between group improvement only on the
                                                      performance of the practiced motor reaching
Liu et al. 2005   46 acute stroke patients were       At the end of weeks 2 and 3, patients in the
Hong Kong         randomized to receive either 15     mental imagery group had higher scores on
7 (RCT)           sessions (1hr/day x 3 weeks) of     the ADL tasks and at one-month follow-up.
                  either a mental imagery program or There were no significant differences in mean
                  functional training designed to     FM or CTT between groups.
                  improve performance of ADLs.
                  During the 3 weeks, patients were
                  trained to perform 3 sets (5 items
                  each) of daily tasks. Patients also
                  received 1-hour of PT daily. The
                  ability to perform tasks was

10. Upper Extremity Interventions                                                         pg. 38 of 150
                 assessed on a 7-point Likert scale,
                 where the higher values were
                 associated with increasing
                 independence. Fugl-Meyer (FM)
                 and Colour Trails Test (CTT) were
                 also assessed at the end of
Page et al. 2005 11 chronic stroke patients received    Patients in the mental practice group had a
USA              30 minute therapy sessions twice a     greater mean change score of ARA test results
6 (RCT)          week for 6 weeks. Patients were        (10.7 vs. 4.6, p=0.004). Patients in the MP
                 randomized to an intervention          group also increased the amount of use in
                 consisting of either mental practice   their affected upper limb and the quality of the
                 of ADL activities or to sessions       movements improved to a greater degree.
                 which focused on relaxation
                 techniques. Outcome measures
                 included the Motor Activity Log and
                 Action Research Arm tests.
Butler et al.    4 patients received a 2-week           After 2-weeks CIMT or CIMT + MP (in one
2006             regimen of either mental practice      case) resulted in modest decreases in
USA              (MP) + constraint-induced              impairment and functional improvement. MP
No Score         movement therapy (CIMT)(n=2),          alone did not result in a clinically meaningful
                 CIMT only (n=1) or MP only (n=1).      improvement in upper-limb impairment.
Ertelt et al.    15 chronic stroke patients with        Patients in the treatment group showed
2007             moderate motor deficits of the         significantly greater improvements from pre-
Germany          upper extremity were randomly          test to post-test on FAT, WMFT and SIS at
5 (RCT)          assigned to receive either action      post-treatment (4 weeks) compared to
                 observation therapy (treatment) or     patients in the control group. The
                 traditional therapy (control). The     improvements were sustained at 8 weeks
                 treatment group underwent 18           following the intervention.
                 consecutive daily sessions lasting
                 90 min each, in which patients
                 watched 6 min videos of sequences
                 of arm and hand movements and
                 then performed the movements for
                 6 mins following the video. The
                 control group received the same
                 therapy however they watched a
                 video with geometric shapes.
                 Outcome measures included
                 Frenchay Arm Test (FAT), the Wolf
                 Motor Function Test (WMFT) and
                 the Stroke Impact Scale (SIS).
Page et al. 2007 32 chronic stroke patients were        The MP group improved significantly on the FM
USA              randomly assigned to receive 30-       compared to the control group (+ 6.7 vs. +
6 (RCT)          min mental practice (MP) sessions      1.0, p<0.0001) and the ARA (+ 7.8 vs.+ 0.44,
                 (n=16) or a sham intervention          p< 0.001).
                 consisting of 30 min of relaxation
                 exercises (n=16), 2 days/wk for 6
                 weeks and were preceded by 30
                 min of standard therapy. Outcomes
                 included the upper-extremity
                 portion of the Fugl-Meyer
                 Assessment (FM) and the Action

10. Upper Extremity Interventions                                                         pg. 39 of 150
                 Research Arm test (ARA).
Page et al. 2009 10 chronic stroke subjects               Subjects in the mCIT+MP group exhibited
USA              exhibiting stable, affected arm          significantly greater gains on both outcomes
4 (RCT)          motor deficits were recruited from       after intervention. 1) ARAT: +15.4 vs. 8.4,
                 the community and received               (p<0.001); 2) FM: +7.8 vs. +4.1, p=0.01).
                 modified constraint-induced therapy      These changes were sustained 3 months after
                 (mCIT), consisting of structured         intervention.
                 therapy emphasizing affected arm
                 use in functional activities 3
                 days/week for 10 weeks and less
                 affected arm restraint 5 days/week
                 for 5 hours. 5 subjects were
                 randomly assigned to receive an
                 additional co-intervention-mental
                 practice (MP) for 30 minutes/day
                 which required daily cognitive
                 rehearsal of the activities of daily
                 living practiced during mCIT clinical
                 sessions. Outcomes assessed the
                 Action Research Arm test (ARAT)
                 and Fugyl-Meyer (FM)
Riccio et al.    36 stroke patients underwent a 6         There were statistically significant differences
2009             week inpatient rehabilitation            favouring the group that received CV+MP at
Italy            program. Patients were randomly          the 3-week crossover point on all outcomes
5 (RCT)          assigned to receive 3 weeks of           assessed. There were no significant differences
                 mental practice (MP) +                   between groups at the end of treatment
                 conventional rehabilitation (CV)(3       period, after which all patients had received
                 hrs/day x 5 days/week) followed by       MP +CV.
                 3 weeks of CV or CR for 3 weeks
                 and then CV + MP for the next 3
                 weeks. Outcome assessments were
                 conducted at baseline, 3 weeks
                 (crossover point) and at 6 weeks
                 and included Motricity Index (upper
                 etremity subscore), Arm Function
                 Test-Functional Ability Scale and
Bovend'Eerdt et 50 patients undergoing either             Patients in both groups improved over time,
al. 2010         inpatient or outpatient rehabilitation   but there were no significant differences
The Netherlands following stroke were randomized to       between groups on any of the outcomes
8 (RCT)          receive a 6-week program                 assessed. Compliance with advised treatment
                 consisting of conventional therapy       was poor in 85% of the therapists and in 72%
                 + mental practice (total of 6.5 hurs)    of the patients.
                 or to conventional therapy only.
                 Assessments were conducted at
                 baseline, after 6 weeks
                 (postintervention), and after 12
                 weeks (follow-up). Outcomes
                 included Goal Attaintment Scaling,
                 BI, Rivermead Mobility Index,
                 Nottingham Extended ADL, Action
                 Research Arm Test and Timed up &

10. Upper Extremity Interventions                                                          pg. 40 of 150
Discussion                                   achieve the improvements in motor
                                             function. Mental practice also appeared
Page et al. (2000, 2001, 2005, 2007)         to provide additive benefit when
used a tape-recorded (guided) imagery        combined with the co-intervention of
intervention to enable mental practice,      modified constraint-induced therapy
whereby patients would sit in a room         (Page et al. 2009). Bovend'Eerdt et al.
quietly and listen to a male voice           (2010) suggested that the poor
encouraging them to first relax (warm-       compliance with the therapy was
up) and then to mentally perform a           instrumental in the failure of patients
series of tasks (reaching for a cup).        to achieve significantly better
Patients mentally practised both at          outcomes.
home and during supervised therapy
sessions. Patients in both the control        Conclusions Regarding Mental Imagery
and intervention groups also received
occupational therapy. Page et al.             There is strong (level 1a) evidence that
reported significant improvement in           mental practice may improve upper-
Fugl-Meyer scores (2000; 2007) and            extremity motor and ADL performance
Action Research Arm tests (2005;              following stroke.
2007) between treatment and control
subjects. Patients in the study by             Mental practice may result in improved
Dijkerman et al. (2004) were asked to          motor and ADL functioning after
read a set of instructions directing           stroke.
them through a series of tasks. A
placebo mental imagery condition was         10.2.8 Hand Splinting
also used where patients were asked
to describe a series of pictures, which      There are many aims when applying
had been presented previously. There         splints. These include: reduction in
was no difference between groups with        spasticity, reduction in pain,
respect to ADL performance, measured         improvement in functional outcome,
by the Barthel Index. The study by Liu       prevention of contracture, and
et al. (2004) had patients in the            prevention of edema (Lannin &
mental practice group practice a             Herbert, 2003). In a systematic review
different series of mental tasks each        of hand splinting for adults with stroke,
week (ie. week 1: wash the dishes,           Lannin and Herbert (2003) concluded
prepare tea, fold laundry) while             that there was insufficient evidence to
patients in the control group                either support or refute the
(functional training program) practiced      effectiveness of hand splinting for
the same tasks having the therapist          adults following stroke. In a
first demonstrate the task. This study       systematic review by Steultjens et al.
reported a benefit of treatment in           (2003), the authors also concluded
terms of improvement in ADL                  that based on the results of 2 RCTs
performance. Page et al. (2005) also         (Langlois et al.1991, Rose et al. 1987),
demonstrated that mental practice,           2 case-controlled trails (McPherson et
where patients cognitively rehearse          al. 1982, Poole et al. 1990) and one
ADL activities, improved motor               uncontrolled trial (Gracies et al. 2000)
function in the affected upper limbs of      that there was insufficient evidence
chronic stroke patients. The author          that splinting was effective in
suggests that the technique induces          decreasing muscle tone. Five RCTs
use-dependent brain reorganization to        were included in our review on the

10. Upper Extremity Interventions                                          pg. 41 of 150
effectiveness of splinting with a variety              of clinical aims.

Table 10.16 Splinting the Upper Extremity
Author/                         Methods                                       Outcomes
Pedro Score
Rose et al.     30 patients with spastic wrist flexors      The patients in the intervention groups had
1987            resulting from a diagnosed stroke, no       significant increases in passive range of wrist
4 (RCT)         more than 6 months post-stroke.             extension and a decrease in hypertonus
                Subjects were randomly assigned to          compared with the control group. No
                either of two experimental groups           significant difference in passive range of
                (predominantly dorsal or predominantly      motion or resistance to passive extension was
                volar static orthosis) or to the control    found between the dorsal and volar splinting
                group (no orthosis). Both of the            groups. There was a significant difference
                intervention groups had hand splinting      between spontaneous flexion between dorsal
                in the functional position for 2 hours      and control groups, but not volar and control
                using either volar or dorsal splints.       groups.
Poole et al.    18 patients with hemiplegia were            No difference in motor function (Fugl-Meyer)
1990            matched according to upper extremity        in the wrist and hand after wearing the
5 (RCT)         motor scores on the Fugl-Meyer. Within      inflatable pressure splint.
                each pair, subjects were then randomly
                assigned to either a non-splint or splint
                condition. The intervention group wore
                an inflatable pressure splint which
                positioned the shoulder in 90 degrees of
                flexion and maximum external rotation
                with full elbow extension (hand and
                wrist not enclosed in the splint) for 30
Langlois et al. 9 patients within 12 months of acute        No significant differences were found in
1991            stroke were randomly allocated to one       spasticity reduction between groups,
3 (RCT)         of 3 groups: wearing a finger spreader      however, all 3 groups demonstrated a
                splint for 6 hrs, 12 hrs or 22 hrs per      reduction in spasticity. No significant
                day for a period of 2 weeks.                differences were found between groups on
                                                            measures of expectation or satisfaction and
                                                            the reduction of spasticity (Expectation and
                                                            Satisfaction Questionnaire that was developed
                                                            specifically for this study), or on reported
                                                            compliance and prescribed wearing schedule
                                                            at 2 weeks. There was a significant
                                                            association between expectation and
Lannin et al.   28 rehabilitation patients were             No difference in contracture formation in the
2003            randomized to either control or             wrist and finger flexor muscles between
Australia       experimental groups. Subjects in both       groups.
8 (RCT)         groups participated in routine therapy
                for individual motor training and upper
                limb stretches 4 days a week. In
                addition, patients in the experimental
                group wore an immobilizing hand splint
                on a daily basis, for a maximum of 12
                hours each night, for 4 weeks.
Lannin et al.   63 stroke patients within 8 weeks of        There were no significant differences between

10. Upper Extremity Interventions                                                          pg. 42 of 150
2007            stroke onset were randomly allocated to     groups or within groups. Splinting did not
Australia       receive 1 of 3 therapies: i.) no splint     reduce wrist contractures.
7 (RCT)         control group (n=21), ii.) a neutral
                splint group (n=20), or iii.) an
                extension splint group (n=21). All
                patients received routine rehabilitation.
                Splints were worn 12 hours overnight
                for the 4-week treatment period. The
                Primary Outcome was muscle
                extensibility of the wrist and fingers,
                assessed before/after treatment and at
                6 weeks.
Bürge et al.    30 hemiparetic patients with subacute       At baseline, 2 patients in each group
2008            hemiparesis and severe upper limb           complained about a painful hand. After 13
Switerland      deficits were randomized to 1) a            weeks, 8 subjects in the control group and 1
5 (RCT)         standard rehabilitation program without     subject in the orthosis group complained of
                orthosis 2) an experimental orthosis in     hand pain. The number of patients presenting
                addition to their standard rehabilitation   with a loss of wrist mobility increased in the
                program. The orthosis group wore the        control group (from 1 at baseline to 8 at 13
                neutral functional realignment orthosis     weeks) while remaining unchanged in the
                for at least 6 hours daily. Outcome         orthosis group (which remained at 4). In
                measures assessed before, and at 13         terms of hand edema, it remained unresolved
                weeks at the end of treatment included      in the one subject in the orthosis group while
                hand pain at rest (visual analog scale),    the number of subjects with hand edema
                wrist range of motion (Fugl-Meyer           increased from 1 to 2 in the control group.
                Assessment subscale), and edema of
                hand and wrist (circumferences).

Of the six studies, two were of “good”
quality, two were of “fair” quality, and
two were of “poor” quality.

Table 10.17 Summary of Splinting the Upper Extremity
   Author/      n            Intervention                                   Main Outcome(s)
PEDro Score                                                                      Result
Lannin et al.   28    Routine motor training +                          Contracture formation
2003                    upper limb stretches +                  (- in wrist and finger flexor muscles)
8 (RCT)                immobilizing hand splint
                      vs. routine motor training
                        + upper limb stretches
Lannin et al.   63         Neutral splint vs.                             Muscle Extensibility
2007                    extension splint vs. no                 (- in wrist and finger flexor muscles)
7 (RCT)                          splint
Poole et al.    18    Inflatable pressure splint                              Fugl-Meyer
1990                         vs. no splint                            (- in the wrist and hand)
5 (RCT)
Bürge et al.    30        Neutral functional                                   Pain (-)
2008                   realignment orthosis vs.                             Fugl-Mayer (-)
5                              no splint                                      Edema (-)
Rose et al.     30    Dorsal static orthoses vs.                  Passive Range of Wrist Extension
1987                  volar static orthosis vs. no          (+ for both volar and dorsal orthoses groups)

10. Upper Extremity Interventions                                                           pg. 43 of 150
4 (RCT)                        orthosis                                    Hypertonus
                                                    (+ decrease for both volar and dorsal orthoses groups)
                                                   Passive range of motion or resistance to passive extension
                                                              (- between dorsal and volar groups)
                                                                      Spontaneous Flexion
                                                             (+ between dorsal and control groups)
                                                             ( - between volar and control groups)
Langlois et      9      Finger spreader splint for                   Spasticity reduction (-)
al. 1991                 6 hrs vs. finger spreader      Expectation and Satisfaction Questionnaire (-)
3 (RCT                  splint for12 hrs vs. finger Reported compliance and prescribed wearing schedule
                        spreader splint for 22 hrs                       (- at 2 weeks)
                                   per day
- Indicates non-statistically significant differences between treatment groups
+ Indicates statistically significant differences between treatment groups

                                                       difficult to disassociate from the
 Conclusions Regarding Treatment of                    contribution of constraining the good
 Spasticity: Hand Splinting                            limb, this form of treatment shows
                                                       promise, especially for survivors with
 There is strong (Level 1a) evidence that              moderate disability following stroke.
 hand splinting does not improve motor
 function or reduce contracture                        Several reviews have been published
                                                       on the effectiveness of CIMT (Taub &
                                                       Morris, 2001, Barreca et al. 2003,
  Hand splinting does not improve                      Hakkennes & Keating 2006, Bonaiuti et
  motor function or reduce contractures                al. 2007) and while the results have
  in the upper extremity.                              been generally positive, uncertainty of
                                                       its effectiveness remain due to the
10.2.9 Constraint-Induced Movement                     small number of trials published, the
Therapy                                                small sample sizes of the studies and
                                                       heterogeneous patient characteristics,
Constraint-Induced Movement Therapy                    duration and intensity of treatment,
(CIMT) refers to a new set of                          and outcomes assessed. A meta-
rehabilitation techniques designed to                  analysis conducted by Van Peppen et
reduce functional deficits in the more                 al. (2004) concluded that CIT was
affected upper extremity of stroke                     associated with improvements in
survivors. The two key features of                     dexterity, measured by the Arm Motor
CIMT are restraint of the unaffected                   Activity Test or the Action Research
hand/arm and increased practice /use                   Arm test, but not in performance of
of the affected hand/arm (Fritz et al.                 ADL, measured by FIM or Barthel
2005). Since stroke survivors may                      Index scores. A more recent review by
experience “learned non-use” of the                    Hakkenes and Keating (2006) included
upper extremity within a short period                  the results from 14 RCTs and
of time (Taub 1980), CIMT is designed                  concluded that there was a benefit
to overcome learned non-use by                         associated with treatment although
promoting cortical reorganization                      larger well-designed studies are still
(Taub et al. 1999). While the                          required. Several treatment contrasts
biological mechanism(s) responsible                    were examined including traditional
for the benefit are unknown and the                    CIT vs. alternative therapy or control,
contribution from intense practice is                  modified CIT vs. alternative therapy or

10. Upper Extremity Interventions                                                       pg. 44 of 150
control and traditional CIT vs. modified      modifiable.” (Taub et al. 2003). Van
CIT, although pooled estimates of the         der Lee (2001) suggests that the
treatment sizes for the subgroups were        positive results attributed to CIMT may
not provided. The associated pooled           simply reflect a greater intensity of
effect sizes for all of the included RCTs     training of the affected arm and
were: Action Research Arm test 1.51           questions the concept of non-use
(95% CI 0.27, 2.74), Fugl Meyer               implying that it may not be a distinct
Assessment 1.16 (95% CI –0.18, 2.52)          entity, but rather the result of sensory
and the Wolf Motor Function Test 0.50         disorders or hemineglect.
(95% CI –0.28, 1.27).
                                              According to Dromerick et al. (2000),
Taub et al. (2003) noted that                 constraint of the unaffected arm by
constraint-induced movement therapy           use of a mitten (6 hours per day for 14
has limitations in that the                   days), and ‘forced use’ of the affected
improvement seen does not restore             arm soon after stroke (mean six days),
the stroke patients’ movement to their        is feasible. However, trials reporting
motor status prior to the stroke. The         small, but significant reductions in arm
same authors note that constraint-            impairment, especially for patients
induced movement therapy “produces            with sensory disorders and hemi-
a variable outcome that depends on            neglect (van der Lee et al. 1999,
the severity of initial impairment. If        Ploughman & Corbett 2004), have also
patients with residual motor function         reported a high number of deviations
are categorized on the basis of their         from the randomized treatment
active range of motion, the higher            schedule, due to patients’ non
functioning individuals tend to improve       compliance. This has led to trials
more than persons who are more                investigating the effectiveness of
disabled (Taub et al. 1999)…. For             modified or shorter periods of
patients with the lowest motor                constraint induced therapy treatment.
functioning, constraint-induced therapy
does improve movement at the                  There is promising evidence that the
shoulder and elbows. Because these            drawbacks to stroke patient
people have little or no ability to move      participation in CIMT (i.e., required
the fingers, there is no adequate motor       practice intensity and duration of
basis for carrying out training of hand       restraint) may be overcome through
function. Consequently, because most          modifications to the basic procedures.
daily activities that are carried out by      They include a less intense, modified
the upper extremity are performed by          CIMT (mCIMT) that combines
the hand, there is relatively little          structured functional practice sessions
translation of the therapy induced            with restricted use of the less affected
movement in proximal joint function           upper limb (Page et al., 2004), and
into an increase in the actual amount         also forced-use therapy (FUT) which
of use of the more affected extremity         employs constraint without intensive
in the real life situation… Thus,             training of the affected arm
constraint-induced therapy is clearly         (“shaping”) (Ploughman & Corbett,
not a complete answer to motor                2004). Page et al. (2002, 2004 and
deficits after stroke. The work so far        2005) provide one example of the
does show that motor function in a            distinction between CIMT and mCIMT:
large percentage of patients with             CIMT is defined by the i) restriction of
chronic stroke is substantially               a patient’s less affected upper-limb

10. Upper Extremity Interventions                                          pg. 45 of 150
during up to 90% of waking hours               The results from the largest and most
during a 2-week period and ii)                 rigorously conducted trial-The
participation in an intensive upper-           Extremity Constraint Induced Therapy
extremity therapy program for 6 hours          Evaluation (EXCITE), may provide the
per day, using the affected limb during        strongest evidence of a benefit of CIMT
the same 2-week period. In contrast,           treatment, to date. The study recruited
mCIMT involves the restriction of the          222 subjects with moderate disability 3
unaffected limb for periods of 5 hours         to 9 months following stroke, over 3
a day, 5 days a week for 2 weeks               years from 7 institutions in the US.
combined with structured, ½ hour               Treatment was provided for up to 6
therapy sessions, 3 days a week.               hours a day, 5 days a week for 2
However, other criteria for defining           weeks. Patients were reassessed up to
mCIMT have also been used, which               24 months following treatment. At 12
overlap with CIMT, blurring the                months, compared with the control
distinction. Lin et al. (2007) cite            group who received usual care,
mCIMT as providing 2 hours of therapy          subjects in the treatment group had
a day for 10-15 consecutive weekdays           significantly higher scores on sections
+ restraint for 6 hours per day. There         of the Wolf Motor Function test and the
are also examples of trials, presented         Motor Activity Log. At 24 months these
in the following tables, which provided        gains were maintained. While these
the intervention for periods of up to 10       results are encouraging, as Cramer
weeks.                                         (2007) points out, the number of
                                               patients for whom this treatment may
There is some evidence, too, for the           be suitable for, remains uncertain. In
beneficial effect of donepezil, a              the EXCITE trial, only 6.3% of patients
primarily central acetylcholinesterase         screened were eligible. While larger
inhibitor, as an adjuvant therapy              estimates of 20-25% have been
(Nadeau et al., 2004; Richards et al.          suggested, it remains uncertain if
2006). Taub et al. (2005) recently             subjects with greater disability would
reported that the benefits associated          benefit from treatment.
with CIMT could be achieved with the
use of an automated device                     A recent Cocharne review (Sirtori et al.
(AutoCITE).                                    2009) examined the benefit of all
                                               forms of CIMT including studies that
The optimal timing of treatment                used the traditional protocol as
remains uncertain. While there is              described by Taub, in addition to trials
evidence that patients treated in the          of modified CIMT and forced use. The
acute phase of stroke may benefit              review included the results from 19
preferentially (Taub & Morris 2001),           trials involving 619 subjects. The
there is also evidence that it may, in         primary outcome was disability. The
fact, be harmful (Dromerick et al.             authors reported that there was a
2009). Grotta et al. (2004) suggest            significant improvement in arm motor
that the greatest benefit is likely to be      function, assessed immediately
conferred during the chronic stages of         following the intervention, but not at
stroke and that the treatment has              3-6 months post-intervention. A
been shown to be harmful in animal             subgroup analysis compared the
studies of “forced use” immediately            benefit of CIMT in terms of time since
post stroke.                                   stroke onset (0-3 months and >9
                                               months). No studies were included that

10. Upper Extremity Interventions                                          pg. 46 of 150
measured disability 3-9 months
following stroke. The associated effect              Studies examing modified CIMT and
sizes were not statistically significant             CIMT, as well as several studies that
for either subgroup. The authors                     examined forced use therapy are
caution that the findings cannot be                  presented in Table 10.18.
considered robust due to the small
sample sizes and poor methodological
quality of the primary studies.

Table 10.18 Studies Evaluating Constraint-Induced Movement Therapy
    Author,                    Methods                                     Outcomes
 Pedro Score
Wolf et al.    21 patients had their uninvolved upper    Significant changes in force or speed seen in
1989           extremity kept within a hand-enclosed     19 of the 21 tasks with most persisting at the
USA            sling during walking hours over a 2-      1-year follow up.
No Score       week period.
Taub et al.    9 patients randomized to either have    Restraint group showed significantly greater
1993           their unaffected upper extremity        improvement in quality of movement and
USA            restrained in a sling during waking     functional ability compared to control on
6 (RCT)        hours for 14 days with 10 of those 10   Emory Test and the Arm Motor Activity Rest
               days patients given 6 hours of practice test at the end of treatment. Motor Activity
               in using impaired upper extremity or to Log indicates that the restraint group showed
               receive several procedures designed to  a marked increase in their ability to use their
               focus attention on use of the impaired  affected upper extremity. Gains made during
               upper extremity (control).              treatment period were maintained during 2
                                                       year follow up.
Kopp et al.    4 patients had their unaffected upper   All behavioural measures showed significant
1999           extremity restrained by a resting hand- improvement in the use of the affected
USA            splint placed in a sling for 90% of     extremity. Study wide effect size amounted to
No Score       waking hours for two weeks and they     2.38 pre- to post-treatment and 1.92 pre-
               were given behaviour training of the    treatment to follow up.
               affected limb (6h/day).
van der Lee et In an observer blind trial, 66 patients Mean improvement on Action Research Arm in
al. 1999       were randomized to receive either       patients with sensory disorder was
Netherlands    forced use therapy with immobilization significantly greater in those receiving force
7 (RCT)        of the unaffected arm combined with     use rather than bimanual training.
               intensive treatment or to receive       During treatment, force use patients also
               intensive bimanual training based on    showed greater clinical significant
               Neuro-Development Treatment.            improvement on Motor Activity Log than
                                                       bimanual training patients.
Dromerick et 20 acute, stroke inpatients were          Total Action Research Arm scores were
al. 2000       randomized to receive either CI         significantly higher in patients who received
USA            movement therapy or traditional upper CI treatment. FIM score for upper body
6 (RCT)        extremity therapies for two weeks.      dressing was also significantly improved.
Sterr et al.   15 stroke patients with chronic         Assessments were performed at baseline,
2002           hemiparesis were randomized to 2        pretreatment, post-treatment and weekly
UK             training groups for 14 days -group 1:   follow-up for 4 weeks. MAL: Both groups
4 (RCT)        standard receive constraint–induced     improved over time (p<0.01), with a greater
               movement therapy (CIMT) plus 6 hours treatment effect for the 6 hrs group,
               of daily ‘shaping procedure’ 2: CIMT    compared to the 3 hr group. (p<0.05).
               plus ‘shaping procedure’ for 3hrs/day.

10. Upper Extremity Interventions                                                       pg. 47 of 150
                 Assessments of Motor Activity Log
                 (MAL) and Wolf Motor Function Test
Page et al.      14 stroke patients an average of 4.4        After the intervention mCIT patients had
2002             months post stroke were randomized to       significantly improved Fugl-Meyer scores
USA              receive one of three treatments: 1)         (+11, p=.02). Patients in the mCIT group also
5 (RCT)          modified constraint-induced therapy         improved 11 points on the Action Research
                 (mCIT): 30 min each of physical and         Arm test. Patients in the TR and CON groups
                 occupational therapy 3x per week for        did not demonstrate significant improvement.
                 10 weeks + mCIT program, 2)
                 traditional rehab (TR) :30 min each of
                 physical and occupational therapy for
                 10 weeks or 3) Control (CON): no
Wittenberg et    16 stroke patients more than 1 year         There was no significant difference between
al. 2003         post stroke with significant impairment     groups on the Wolf Motor function test, the
USA              indicated by the Motor Activity Log were    Assessment of Motor and Process Skills,
5 (RCT)          randomized to receive intense or less       transcranial Magnetic stimulation or Positron
                 intense CIMT. The intense group             emission tomography. However, both groups
                 received CIMT for 10 continuous             did show statistically significantly
                 inpatient days for 6 hrs/day (4 hrs/day     improvements in the above outcome
                 on weekends). The less intense group        measurements. There was however a
                 received CIMT 3 hrs/day on weekdays         significant difference in favour of the more
                 only. Outcome measures included: Wolf       intense CIMT group in the Motor activity Log
                 Motor function test, Motor activity Log,    scores.
                 Assessment of Motor and Process Skills,
                 transcranial Magnetic stimulation and
                 Positron emission tomography.
Alberts et al.   10 right-handed stroke patients (3-9        Overall change in WMFT median time was
2004             months post stroke) from the ongoing        nearly significant (p=0.07). The immediate
USA              EXCITE trial (with 222 patients in total)   CIT group reduced movement time to perform
6 (RCT)          were randomized to 1 of 2 groups: 1)        the key-turning task to 47% after CIT,
                 immediate constraint-induced therapy        whereas the delayed group required 15%
                 (CIT): patients began CIT                   more time after CIT. The maximum precision
                 approximately 3 days after pre-             grip force and maximum strength from
                 intervention evaluations or 2) delayed      pretest to posttest were statistically
                 CIT: patients received CIT                  significant group-by-time interactions for the
                 approximately 1 year after pre-             intermediate group.
                 intervention evaluations; therapy took
                 place over a 2 week period and patients
                 wore a soft mitt for a goal of 90% of
                 waking hours.
Atteya et al.    6 stroke patients (<6 months post           The CIT group showed substantial
2004             stroke) were randomly divided into 3        improvements on the Fugl-Meyer Assessment
Saudi Arabia     groups of 2: 1) constraint induced          of Recovery After Stroke, the Wolf Motor
3 (RCT)          therapy (CIT): patients received 30 min     Function Test and the Motor Activity Log from
                 of physical therapy and occupational        pre to post treatment. No tests of statistical
                 therapy 3x per week for 10 weeks +          significance were conducted.
                 mCIT program, 2) traditional rehab
                 (TR): 30 min of physical and
                 occupational therapy for 10 weeks or 3)
                 Control (CON): no therapy
Page et al.      17 patients with chronic stroke (1 year     The mCIMT patients had significantly greater

10. Upper Extremity Interventions                                                           pg. 48 of 150
2004           post stroke) were randomly assigned to     motor changes on the Fugl-Meyer scores and
USA            one of 3 groups: 1) modified constraint-   on the Action Research Arm test than those in
6 (RCT)        induced therapy (mCIT): 30 min each        the two other groups.
               of physical and occupational therapy 3x
               per week for 10 weeks + mCIT
               program, 2) traditional rehab (TR) :30
               min each of physical and occupational
               therapy for 10 weeks or 3) Control
               (CON): no therapy.
Ploughman &    23 stroke patients were randomized        FUT participants had greater recovery of
Corbett 2004   into a forced-use therapy (FUT;           postural control (Chedoke McMaster
Canada         constraint without shaping) group         Impairment Inventory, CMII) (p=.04), and
5 (RCT)        (n=10) and a control group                more shoulder pain, than did controls. CMII
               (conventional treatment for the upper     mean scores suggested greater clinical
               extremity; n=13).                         recovery for the arm and hand for FUT
                                                         participants, but the results were not
                                                         statistically significant.
Suputtitada et 69 chronic stroke patients were           ARA scores, pinch strength of affected upper
al.            allocated either to constraint-induced    extremities were significantly higher for CIMT
2004           movement technique (CIMT) (n = 33)        patients compared to the control group. There
Thailand       or conservative treatment (n = 36). The were no statistically significant differences
6 (RCT)        CIMT group received 6 hours of daily      between the groups in hand grip strength.
               affected-upper-extremity training and
               restrained unaffected upper extremities
               for 5 days per week, totally 2 weeks.
               The control group received bimanual-
               upper-extremity training by
               conservative neurodevelopmental
               technique without restrained unaffected
               upper extremities for 2 weeks.
               Assessments included the Action
               Research Arm Test (ARA test), hand
               grip strength, pinch strength of affected
               upper extremity
Bonifer et al. 20 chronic patients, > 1 yr post stroke There were significant improvements in: Fugl-
2005           with moderate to severe levels of         Meyer Assessment scores, Graded Wolf Motor
USA            impairment received CIT therapy for 3 Function Tests and Motor Activity log at end
No Score       weeks, including 6 hours of upper         of treatment and one-month follow-up.
               extremity training/day and wore a soft
               mitt and sling for a goal of 90% of
               waking hours.
Brogårdh &     16 chronic stroke patients underwent 2 There was significant improvement for the
Bengt          weeks of constraint-induced movement Modified Motor Assessment Scale (p=0.003),
2006           therapy in groups of 2-3 patients per     the Sollerman Hand Function Test (p=0.037),
Sweden         therapist for 6 hours/day. Each wore a and the Motor Activity Log (p<0.001) after
7 (RCT)        mitt on the unaffected hand for a target the 2 week CIMT treatment period. However,
               of 90% of waking hours. Patients were there were no significant differences between
               randomly assigned into 1 of 2 groups:     groups at the end of 3 months.
               1) Using the mitt at home for an
               additional 3 months every other day, 2)
               no further treatment.
Page et al.    10 acute stroke patients with upper       Improvements in both Action Research Arm
2005           limb hemiparesis and within 14 days of test and Fugl Meyer Scale from pre to post

10. Upper Extremity Interventions                                                       pg. 49 of 150
USA             stroke were randomized to receive            treatment. No between group differences
5 (RCT)         constraint-induced movement therapy          were reported.
                (CIT) or regular rehabilitation. Patients
                in the CIT group received ½ hr therapy
                sessions, 3X/week for 10 weeks. During
                this time patients’ unaffected arm and
                hand was restrained everyday for 5
                hours. Patients in the regular
                rehabilitation group received standard
                therapy for their affected limbs. Fugl-
                Meyer, Action Research Arm test and
                Motor Activity Log tests were conducted
                at week 10.
Tarkka et al.   27 chronic stroke patients, mean time        There was a significant improvement in the
2005            post stroke of 3.6 years received two-       mean Wolf Motor Function test after
Finland         weeks of CIMT program including 7            treatment, which remained a five-month
No Score        hours per day of therapy. The                follow-up.
                unaffected arm was always restrained.
Page & Peter 6 patients <1 yr following stroke with          Following ETMS patients showed no functional
2006            upper-limb hemiparesis underwent             changes. After EMTS and mCIMT patients had
USA             electromyography-triggered                   an improved ability to perform FMA wrist
No score        neuromuscular stimulation (ETMS) 2           items. A new ability to finish FMA hand items
                days/week for 35 min over 8 weeks. A         was seen in patients following mCIMT. No
                week later patients received modified        significant differences were reported.
                constraint-induced movement therapy
                (mCIMT) including ‘shaping’ procedures
                3 days/week for 10 weeks. The less
                affected arm was restrained.
Taub et al.     41 chronic stroke patients (>1 yr post-      There were significant improvements in the
2006            stroke) were assigned either to a            Wolf Motor Function Test, the Motor Activity
USA             constraint-induced movement therapy          Log, and the Actual Amount of Use from pre-
No score        for patients with mild to moderate           treatment to post-treatment. In addition,
                deficit following stroke (n=21) or a         females showed greater gains on MAL then
                placebo control group who underwent a        males in the CI therapy group (p=0.02). At 2-
                general fitness program (n=20). CI           year follow-up significant improvements were
                therapy received intensive training of       still seen on the MAL relative to post-
                the more affected upper extremity for 6      treatment for the treatment group.
                hours/day, for 10 consecutive days
                which included ‘shaping’ procedures,
                restraint of less affected extremity for a
                target of 90% waking hours during a 2-
                treatment period.
Richards et al. 39 chronic unilateral stroke individuals     The CIMT-6 group showed significant
2006            with hemiparesis were randomly               improvement in use (P<0.001) and
USA             assigned to receive traditional CIMT         movement quality (p<0.004) compared to the
7 (RCT)         plus donepezil or placebo (CIMT-6),          CIMT-1 group. However, after 2 weeks of
                who underwent the standard 6 hours of        therapy motor skill gains for both groups were
                therapist-guided, in-clinic task practice    equivalent and at six months both groups did
                (n=16) or shortened CIMT along with          not maintain the gains made.
                repetitive transcranial magnetic
                stimulation or sham stimulation (CIMT-
                1) who underwent a single hour of in-
                clinic task practice (n=19). Treatment

10. Upper Extremity Interventions                                                          pg. 50 of 150
               continued for 2 weeks. Outcome
               Measures included the Wolf Motor
               Function Test and the Motor Activity
               Log and were assessed before/after
               treatment and at 6 months.
Wolf et al.    222 patients between 3 to 9 months           The CIMT group significantly improved in the
2006           post stroke received either CIMT             WMFT (log performance time, functional
USA            (n=106) or usual care (no treatment,         ability 0-5 scale (p<0.001), the MAL Amount
8 (RCT)        home care or outpatient programs)            of Use (p<0.001) and the MAL Quality of
EXCITE Trial   (n=116). The CIMT group wore a mitt          Movement (p<0.001) and caregiver MAL.
               on the less-affected hand while              (Group x time interaction).
               performing repetitive task practice and
               behavioural shaping with the hemiplegic
               hand). Outcome measures included the
               Wolf Motor Function Test (WMFT),
               Motor Activity Log (MAL), functional
               ability measures, a measure of the
               quality and frequency of the
               performance of 30 standard daily
               activities. Assessments were conducted
               before/after treatment and at 4, 6 and
               12 months.
Underwood et The results from 41 individuals                Fatigue and pain scores for all participants
al. 2006       recruited from a single centre of the        were low. Patients receiving CIMT did not
USA            EXCITE trial were evaluated. Outcome         have significantly increased pain or fatigue
8 (RCT)        measures were assessed before and            compared with those in the control group.
               after treatment including the pain scale
               of the Fugl-Meyer test for upper
               extremity and the WMFT.
Uswatte et al. 18 chronic stroke patients were              One form of treatment did not emerge as
2006           consecutively assigned to 1 of 4             superior. For all groups combined there was a
USA            groups: i.) Sling and Task-practice          significant improvement from pre-treatment
No Score       (n=4), ii.) Sling and Shaping (n=5), iii.)   in MAL scores (1.6 vs. 3.1 at 2 years,
               Half-glove and Shaping, iv.) Shaping         p<0.05).
               Only (n=4). Task-practice occurred 6
               hr/day for 2 weeks and incorporated
               using the affected arm to carry out
               repetitive arm training on functional
               tasks. Patients who underwent shaping
               treatment were given feedback
               immediately after tasks preformed and
               tasks became increasingly more
               demanding over the 2 week period.
               Outcome measures included Motor
               Activity Log and the Wolf Motor
               Function Test and were assessed
               before/after treatment and at 1-month
               and 2 years.
Boake et al.   23 patients within 2 weeks of stroke         Patients in both groups improved but there
2007           were randomly assigned to receive            were no statistically significant differences
USA            either 2 weeks of CIMT (n=10) or             between groups, although there was a trend
5 (RCT)        traditional rehabilitation (n=13). Both      in favour of CIMT over traditional therapy.
               given at the same frequency of up to 3

10. Upper Extremity Interventions                                                           pg. 51 of 150
                hours daily. Outcome measures include
                Fugl-Meyer (FM) Assessment, Grooved
                Pegboard Test, Motor Activity Log
                (MAL), assessed before/after treatment
                and at 3 months.
Burns et al.    In an A-B-A designed trial, 10 subjects     A mean improvement in ARAT scores of 4.0
2007            with a mean onset time of 6.1 months        points (95% confidence interval 1.7 to 6.2;
UK              post stroke wore a constraint mitten on     P=00.016) was found following intervention
No Score        the unaffected upper limb for 9 waking      (correcting for background recovery). The
                hours/day for two weeks to encourage        improvement was just below that associated
                use of the hemiplegic arm. Existing         with a clinically significant improvement.
                levels of therapy continued during the      Mean compliance was 6.7 hours/day
                entire study. The primary outcome
                measure was the Action Research Arm
                Test (ARAT), which was assessed twice
                at baseline and at 4 and 6 weeks
                following intervention.
Lin et al. 2007 32 patients with a mean post-stroke         There were moderate and significant effects
Taiwan          onset time of 16.3 months were              of modified CIMT compared with the control
7 (RCT)         randomized to receive modified CIMT         group on the following kinematic variables:
                (restraint of the unaffected limb or        reaction time, percentage of movement time
                traditional rehabilitation for three        where peak velocity occurs. The mCIMT group
                weeks). Kinematic analysis was used to      also showed significantly improved functional
                assess motor control characteristics as     performance on the MAL (amount of use:
                patients reached to grasp a beverage        2.04 vs. 0.93 and quality of movement: 2.30
                can. Functional outcomes were               vs. 0.99) and FIM scores (113 vs. 106).
                evaluated before and after treatment
                using the Motor Activity Log (MAL) and
Wu et al.       26 stroke patients (0.5 to 31 months        Significantly greater improvements were seen
2007a)          post stroke) were randomized to             for the mCIMT group in FM, FIM, MAL and 3
Taiwan          receive 3 weeks of either modified CIMT     components of the SIS.
5 (RCT)         (mCIMT) (n=13) or traditional
                occupational therapy (n=13). Both
                groups received 2-hour therapy
                sessions 5 days a week. The patients in
                the mCIMT group received intensive
                therapy combined with the use of a
                restraining mitt on the unaffected hand.
                Outcome Measures: Fugl-Meyer (FM)
                Assessment, FIM instrument, Motor
                Activity Log (MAL), and Stroke Impact
                Scale (SIS) were assessed before and
                after treatment.
Wu et al.       30 stroke patients (mean of 18 mos          Patients receiving mCIMT showed greater
2007 b)         post stroke) were randomized to             improvements in reaching kinematics,
Taiwan          receive a course of modified CIMT with      although the effect was modest. MAL scores
6 (RCT)         intensive therapy for 2 hrs, 5 x/wk for 3   were also significantly higher for both amount
                weeks + wearing a mitt for 6 hours a        of use (2.32 vs. 1.45) and quality of
                day or traditional rehabilitation.          movement (2.32 vs. 1.63). FIM scores were
                Kinematic analyses and clinical             significantly higher in the treatment group
                assessments including the FIM and the       (107 vs. 98).
                Motor Assessment Log (MAL) were

10. Upper Extremity Interventions                                                          pg. 52 of 150
                measured before and after treatment.
Wu et al.       47 patients with stroke onset of 3        After treatment, the CIMT group showed
2007 c)         weeks to 37 months (mean 12 months)       better strategies of reaching control than the
Taiwan          were randomized to receive either CIMT    control group as demonstrated by a bell-
6 (RCT)         (restraint of the less affected hand      ringing test. MAL scores were also
                combined with intensive training of the   significantly higher in the mCIMT group for
                more affected upper extremity) or         both amount of use (1.85 vs. 0.81) and
                traditional intervention (control         quality of movement (1.85 vs. 1.08). There
                treatment) during the study. The          were no significant differences in FMA scores
                treatment intensity was matched           (47 vs. 45).
                between the 2 groups (2h/d, 5d/wk for
                3wk) + the treatment group wore a
                mitt for 6 hours a day during the study
                period. Outcomes assessed before and
                after treatment included kinematic
                variables of reaching movement used to
                describe the control strategies for
                reaching, the Fugl-Meyer Assessment
                (FMA) and the Motor Activity Log (MAL).
Dahl et al.     30 patients with mean stroke onset of     At the end of treatment, the CIMT group
2008            17 months were randomized to receive      showed a significantly shorter performance
Norway          CIMT training: 6 hours of arm therapy     time (4.76 seconds versus 7.61 seconds, p=
8 (RCT)         for 10 consecutive weekdays, while        0.030) and greater functional ability (3.85
                using a restraining mitten on the         versus 3.47, P= 0.037) than the control group
                unaffected hand for 90% of waking         (n=12) on the WMFT. At follow-up the CIMT
                time (n=18) or community-based            group maintained their improvement, but as
                rehabilitation (n=12). Primary outcome    the control group improved even more, there
                assessed at the end before and after      were no significant differences between the
                treatment and at 6 month follow-up        groups on any measurements. There were no
                was the Wolf Motor Function Test          significant differences between the groups at
                (WMFT). Secondary outcomes were the       either the end of treatment, or at follow-up on
                Motor Activity Log (MAL), FIM and         any of the secondary outcomes.
                Stroke Impact Scale (SIS).
Wolf et al.     Further results from the EXCITE trial,    The drop out rate was 34% at 24 months.
2008            which assessed outcomes at 24             The effects at 24 months either improved or
USA             months. Only the 106 patients who         remained stable compared with those at 12
8 (RCT)         were randomized to receive immediate      months for all domains of the WMFT, the MAL
                CIMT were included in this analysis.      and for all domains of the SIS scale, except
                                                          memory and thinking.
Sawaki et al.   30 subjects (>3 and <9 months             Both experimental and control groups
2008            poststroke) were randomized into 2        demonstrated improved hand motor function
USA             groups: 1) the experimental group         2 weeks after baseline. The experimental
3 (RCT)         received CIMT immediately after           group showed significantly greater
                baseline evaluation, 2) the control       improvement in grip strength after the
                group where subjects received CIMT        intervention and at follow-up, but not on the
                after 4 months. The primary outcome       overall WMFT or on the force-based measure.
                measure, the Wolf Motor Function Test
                (WMFT) was assessed at baseline, 2
                weeks after baseline, and at 4-month
Myint et al.    43 patients, 2-16 weeks post stroke       The intervention group scored higher over the
2008            with hemiparesis of the affected limb,    study period on both subscales of the MAL
Hong Kong       were randomized to receive a program      test and on total ARAT scores. They also

10. Upper Extremity Interventions                                                        pg. 53 of 150
7 (RCT)         of 10 days upper extremity training (4       scored higher on all of the subscales of the
                hours per day) with the unaffected limb      ARAT at the end of intervention: grasp, grip,
                being restrained in a shoulder sling         pinch and gross, although only grip subscale
                (intervention group, n=23) or to a           scores were significantly better at 12 weeks
                control group which received an              follow-up.
                equivalent duration of conventional
                rehabilitation therapy (n=20). Primary
                outcomes were assessed at baseline,
                post-intervention and at 12 weeks
                follow-up and included the Motor
                Activity Log (MAL), Action Research
                Arm Test (ARAT) and modified Barthel
                Index (BI).
Page et al.     35 patients with chronic stroke (onset       There was no significant treatment effect for
2008            >12 mos) were randomized to 1 of 3           FM scores. Controlling for pre-intervention
USA             groups for a 10 week course of               scores there was a significant treatment effect
5 (RCT)         treatment: i) mCIMT group (n=13), ii)        on ARAT scores favouring the mCIMT group
                a traditional rehab group (n=12) and         (40 vs. 29. vs. 25)
                iii) a no treatment control group
                (n=10). Outcome measures were
                assessed before and after treatment
                and included the Fugl Meyer (FM) scale
                scores and the Action Research Arm
                Test (ARAT)
Lin et al. 2008 22 chronic stroke patients (mean time        The mean improvement was greater for
Taiwan          post onset of stroke = 18.9 months)          subjects in the CIMT group in terms of FMA
5 (RCT)         were randomized to receive CIMT              and FIM scores. There were no significant
                (restraint of the less affected limb         differences between groups on the outcomes
                combined with intensive training of the      of MAL and NEADL, with the exception of the
                affected limb) or traditional intervention   mobility subsection.
                (control treatment) for 2h/d, 5d/wk for
                3 wk. Both groups of patients received
                restraint of the less affected limb
                outside rehabilitation. Outcomes
                assessed before and after treatment
                included Fugl-Myer Assessment (FMA),
                the Motor Activity Log (MAL), FIM and
                the Nottingham extended activities of
                daily living scale (NEADL).
Barzel et al.   A retrospective analysis of 7 patients       Patients from both groups showed similar,
2009            who received a modified (home-based)         non-statsitcally significant improvement in
Germany         version of CIT who were matched with         both outcomes assessed at the end of
No Score        7 patients who received traditional CIT.     treatment and follow-up.
                Patients in the modified group were
                supervised by an instructed family
                member for 2 hours a day for 4 weeks
                and had there unaffected hand
                constrained for a target of 60% of
                waking hours. Patients in the traditional
                CIT group were supervised by a
                physiotherapist and received therapy
                for 6 hours a day for 2 weeks (10 days-
                w/e off) and wore a mitt for a target of
                90% of waking hours. Outcomes

10. Upper Extremity Interventions                                                           pg. 54 of 150
               assessed pre-, post-treatment and 6-
               month follow-up included Wolf Motor
               Function Test (WMFT) and Motor
               Activity Log (MAL).
Hammer &       30 patients, between 1 and 6 months        Subjects in both groups received similar
Lindmark       post stroke, were randomized to a          amounts of therapy. The mean duration of
2009a          forced use or conventional therapy         sling wear was 37.4 hours. There were no
Sweden         group. The patients of both groups         significant differences between groups. There
6 (RCT)        participated in two weeks of daily         was a trend towards higher scores in the
               training on weekdays. In addition, the     forced-used group immediately post-
               forced use group wore a restraining        intervention, but these small differences also
               sling on the non-paretic arm for up to 6   leveled out up to the three-month follow-up,
               hours per weekday. The primary             with both groups earning an approximately
               outcome measure was the Motor              1.0 score point on both scales of the MAL
               Activity Log (MAL), assessed one and       (amount of use and quality of use).
               three months after intervention.
Hammer &       Additional outcomes reported. The Fugl-    There were no significant between group
Lindmark       Meyer (FM) test, the Action Research       differences on any of the outcomes assessed.
2009b          Arm Test, the Motor Assessment Scale       Subjects in both groups demonstrated
Sweden         (MAS) (sum of scores for the upper         improvements over the study period.
6 (RCT)        limb), a 16-hole peg test (16HPT), a
               grip strength ratio (paretic hand to
               nonparetic hand), and the Modified
               Ashworth Scale were used to obtain
Brogårdh &     4-year follow-up from 2006 study. 14 of    There was no significant change in the median
Bengt          the original 16 subjects participated.     score of the Sollerman hand function test
2009a          However, the 2 original study groups       (50.1 points after CIT therapy, with a gain of
Sweden         (continued mitt use for 21 days vs. no     2.3 points 4 years later). Both components of
7 (RCT)        mitt use) following 12 days of CIT were    the Motor Activity Log scale had decreased
               collapsed to one.                          significantly over time: Median Amount of use
                                                          MAL score before CIT was 2.8, which
                                                          decreased by 0.6. Median Quality of
                                                          Movement score was 2.6 after CIT and
                                                          decreased by 0.4 points.
Brogårdh et     24 subjects, an average of 7 weeks        Patients in both groups showed significant
al. 2009b       post stroke with mild to moderate         improvements in arm and hand motor
Sweden          impaired hand function, were              performance and on self-reported motor
5 (RCT)         randomized to mitt use or no mitt use     ability after 2 weeks of therapy and at 3
                on the less affected hand for 90% of      months follow-up. However, no statistically
                waking hours for 12 days. All patients    significant differences between the groups
                received 3 hours of arm and hand          were found in any measures at any point in
                training per day for 2 weeks.             time.
                Assessments included the modified
                Motor Assessment Scale, the Sollerman
                hand function test, the 2-Point
                Discrimination test and Motor Activity
                Log test.
Lin et al. 2009 32 patients within 6 to 40 months after   The mean improvement was greater for
Taiwan          onset of a first stroke were randomized   subjects in the CIMT group oin terms of FMA,
5 (RCT)         to 2 groups: CIT (restraint of the less   FIM, NEADL, and total SIS scores. There were
                affected limb combined with intensive     no significant differences between groups on
                training of the affected limb) for 2      MAL.

10. Upper Extremity Interventions                                                        pg. 55 of 150
                hours daily 5 days per week for 3
                weeks and restraint of the less affected
                hand for 5 hours outside of the
                rehabilitation training) or a conventional
                intervention with hand restraint for the
                same duration (similar protocol to
                2008a study). Outcomes assessed
                before and after treatment included
                Fugl-Myer Assessment (FMA), the Motor
                Activity Log (MAL), FIM and the
                Nottingham extended activities of daily
                living scale (NEADL) and Stroke Impact
                Scale (SIS).
VECTORS         Very Early Constraint-Induced                All groups improved with time on the total
Dromerick et Movement during Stroke Rehabilitation           ARAT score. There was a significant time x
al. 2009        (VECTORS). 52 subjects were                  group interaction. Subjects in the standard
USA             randomized to one of 3 groups an             CIMT and control treatment groups achieved
6 (RCT)         average of 9.7 days following stroke: 1)     similar gains in total ARAT score (24.2 and
                standard CIMT received 2 hours of            25.7, respectively), while subjects in the
                shaping therapy and wore a mitt for 6        high-intensity CIMT group gained an average
                hours per day; 2) high-intensity CIMT,       gain of only 12.6 points.
                3 hours of shaping exercise /day +
                wearing mitt 90% of waking hours; or
                3) control treatment consisting of 1
                hour of ADL training and 1 hour of UE
                bilateral training exercises. All
                treatment was provided for 2 weeks.
                The primary endpoint was the total
                Action Research Arm Test (ARAT) score
                on the more affected side at 90 days
                after stroke onset.
Azab et al.     27 stroke subject an average of 81-87        At the end of treatment the CIMT group had
2009            days post stroke received outpatient         achieved higher BI scores compared with the
Jordan          therapy for 4 weeks (40 minutes              control group (96 vs. 79, p<0.05).
No Score        3x/week). Subjects received either
                conventional or CIMT therapy. The
                experimental/treatment group received
                traditional therapy with the CIMT where
                the intact contralateral upper limb was
                placed in a removable cast for 6 hours a
                day during waking hours. The control
                group received traditional therapy only.
                Both groups were assessed using the
                Barthel Index on admission and on
                discharge from rehabilitation.
Lin et al. 2010 13 patients at least 3 months post           Patients in the CIT group had significantly
Taiwan          stroke were randomized to 2 groups:          higher scores on both the FMA and MAL at the
5 (RCT)         CIT (restraint of the less affected limb     end of treatment compared with the control
                combined with intensive training of the      group. The fMRI data showed that distributed
                affected limb) for 2 hours daily 5 days      form of constraint-induced therapy
                per week for 3 weeks and restraint of        significantly increased activation in the
                the less affected hand for 5 hours           contralesional hemisphere during movement
                outside of the rehabilitation training)      of the affected and unaffected hand. The
                (n=5) or a conventional intervention         control intervention group showed a decrease

10. Upper Extremity Interventions                                                         pg. 56 of 150
               with hand restraint for the same          in primary sensorimotor cortex activation of
               duration (similar protocol to 2009        the ipsilesional hemisphere during movement
               study)(n=8). Outcome measures             of the affected hand.
               included the Fugl-Meyer Assessment
               (FMA), the Motor Activity Log (MLA),
               and functional magnetic resonance
               imaging (fMRI) examination assessed
               before and after treatment.
Wolf et al.    Further results reported from the         106 subjects received early treatment and 86,
2010           EXCITE trial whereby the outcomes of      delayed. The earlier CIMT group showed
USA            subjects who received threatment          significantly greater improvement compared
8 (RCT)        immediately following randomization (3    with the delayed group on the WMFT and the
               to 9 months) were compared with those     MAL. SIS Hand and Activities domains scores
               who received delayed treatment (15 to     were also significantly higher among subjects
               21 months). The primary outcomes          in the early group. Early and delayed group
               were the Wolf Motor Function Test         comparison of scores on these measures 24
               (WMFT) and the Motor Activity Log         months after enrollment showed no
               (MAL). The secondary outcome was the      statistically significant differences between
               Stroke Impact Scale (SIS). Outcomes       groups.
               were assessed before and after
               treatment and at 4, 8, and 12 months

Discussion                                            the stroke (acute vs. chronic). We
                                                      used the author’s own declaration of
To enable better examination of the                   the type of therapy that was provided
included studies, they were classified                (ie. mCIMT or CIMT).The results are
according to type of treatment (CIMT                  summarized in tables 10.19 to 10.22.
or modified CIMT) and to chronicity of

Table 10.19 Summary of RCTs Evaluating CIMT in the Acute Phase Following Stroke
    Author/           Intervention             Intensity/Duration              Main Outcome(s)
 PEDro Score                                                                        Result
Dromerick et    CIMT vs. traditional upper 2 hrs/day x 5 days/wk x Total Action Research Arm and
al.                  extremity therapy                   2 wks               pinch subscore (+)
2000                                                                            FIM score (-)
6 (RCT)                                                                    Upper body dressing (+)
Boake et al.        CIMT vs. traditional        3 hrs/day x 6 days/wk x   Fugl-Meyer Assessment of
2007                    rehabilitation                   2 wks               Motor Recovery (-)
5 (RCT)                                                                   Grooved Pegboard Test (-)
                                                                           Motor Activity Log (+ for
Dromerick et     Standard CIMT vs. high- 2 hrs/day x 5 days/wk x Total Action Research Arm (-)
al.                  intensity CIMT vs.                  2 wks
2009                  traditional upper
6 (RCT)              extremity therapy
- Indicates non-statistically significant differences between treatment groups
+ Indicates statistically significant differences between treatment groups

Dromerick et al. (2000) reported                      Action Research Arm test scores and
significant improvements in total                     on the FIM subset of upper extremity

10. Upper Extremity Interventions                                                       pg. 57 of 150
dressing. Extrapolation from animal                   demonstarting an inverse dose-
studies suggests that CIMT therapy is                 response curve. This result ran counter
most appropriate in the early recovery                to the authors’ hypothesis, predicting
following stroke to minimize the effects              the greatest gains in the most
of learned non-use and could prevent                  intenseive group. The authors
shrinkage. However, the findings of the               proposed too soon timing of the
Boake study do not support these                      intervention following stroke,
conclusions. These authors reported                   overtraining and a practice schedule
that patients receiving CIMT                          that better resembled a blocked,
experienced no greater motor function                 rather than distributed one as possible
recovery compared with patients                       explanations for their findings.
receiving inpatient (followed by
outpatient) rehabilitation at follow-up               Since subjects in both groups received
of 3-4 months. Since the authors                      task-specific therapy directed at the
reported a trend towards greater                      hand and and arm in the study
improvement in the CIMT group, it is                  authored by Brogårdh et al. (2009),
unclear if the study was simply                       we considered it a study examining
underpowered to detect a significant                  forced-use rather than CIT.
difference. In a more recent study
(Dromerick et al. 2009) including 2                   A summary of the results from RCTs
CIMT groups (standard and high                        that evaluated CIMT in the subacute or
intensity), subjects in the higher-                   chronic stages of stroke is presented in
intensity group fared, on average,                    Table 10.19. The author’s own
worse than those in either the control                declarations of whether mCIMT or
group or the standard CIMT group,                     CIMT were used to classify studies.

Table 10.20 Summary of RCTs Evaluating CIMT in the Chronic Phase Following Stroke
    Author/             Intervention            Intensity/Duration          Main Outcome(s)
  PEDro Score                                                                    Result
Wolf et al.       CIMT + a mitt on the un-  6 hrs/day x 5x/wk x 2    Wolf Motor Function Test (+)
2006, 2008        affected hand + ‘shaping            wks                  Motor Activity Log
8 (RCT)           procedure vs. usual care                           (+ amount of use and quality
EXCITE                                                                       of movement)
                                                                    Functional ability measures (-)
                                                                          Quality/frequency of
                                                                       performance of 30 daily
                                                                              activities (-)
Dahl et al.         CIMT vs. community-     6 hrs/day x 5x/wk x 2      Wolf Motor Function Test
2008                 based rehabilitation             wks                End of treatment (+)
Norway                                                                        6 months (-)
8 (RCT)                                                                  Motor Activity Log (-)
                                                                                 FIM (-)
                                                                                 SIS (-)
Sawaki et al.      Early vs. delayed CIMT    14 consecutive days       Wolf Motor Function Test
2008                                       (wearing mitt for 90% of        (+ grip strength)
3 (RCT)                                             the day)           (-total score, lift weight)
Underwood et       Subgroup from EXCITE 6 hrs/day x 5 day/wk x 2 Pain scale of Fugl-Meyer test
al. 2006                                              wks               for upper extremity (-)
8 (RCT)                                                               Wolf Motor Function test (-)
Richards et al.     Traditional CIMT plus    CIMT-6: 6 hrs/day in     Wolf Motor Function Test (-)
2006                donepezil or placebo    clinic x 5 days/wk x 2    Motor Activity Log (+ in use

10. Upper Extremity Interventions                                                     pg. 58 of 150
7 (RCT)           (CIMT-6) vs. shortened                   wks             and movement quality for
                 CIMT along with repetitive      CIMT-1: 1 hr/day in              CIMT-6)
                   transcranial magnetic        clinic + 5 hours home
                   stimulation (CIMT-1)         practice x 5 days/wk x
                                                   All groups wore a
                                                    padded mitt on
                                               unaffected arm for 90%
                                                    of waking hours
van der Lee et       Intensive forced use      6 hrs/day x 5 days/wk x       Action Research Arm
al. 1999          therapy + immobilization               2 wks              (+ at end of treatment)
7 (RCT)             of the unaffected arm                                     Motor Activity Log
                     (n=33) vs. intensive                                    (+ during treatment)
                   bimanual training based
                        on NDT (n=33)
Wu et al. 2007    CIMT (n=24) vs. regular      2 hrs/day x 5 days/wk x       Motor Activity Log (+)
c)                  interdisciplinary rehab             3 wks              Fugl Meyer Assessment (-)
6 (RCT)                     (n=23)
Alberts et al.    Immediate CIT (n=5) vs.      6 hrs/day x 5 days/wk x    Maximum precision grip (+)
2004                  delayed CIT (n=5)                 2 wks             Wolf Motor Function Test (-)
6 (RCT)                                                                    Arm and Hand Section (-)

Suputtitada et          CIMT (n=33) vs.     6 hrs/day x 5 days/wk x         Action Research Arm (+)
al.              bimanual-upper-extremity       14 days or daily                 Pinch test (+)
2004                training based on NDT   weekday therapy for an
6 (RCT)                 approach (n=36)      unspecified time for 2
Taub et al.             Unaffected upper    6 hrs/day x 5 days/wk x                 Emory Test
1993              extremity restrained in a          2 wks                (+ at end of treatment and 2
6 (RCT)              sling + practice using                                             yr)
                 impaired upper extremity                                  Arm Motor Activity Rest test
                     (n=4) vs. procedures                                 (+ at end of treatment and 2
                        designed to focus                                               yr)
                 attention use of impaired                                      Motor Activity Log
                 upper extremity (control)                                 (+ increase in ability to use
                             (n=5)                                          affected upper extremity)
Wittenberg et      Intense CIMT (n=9) vs.      6 hrs/day (4hrs on             Motor Activity Log (+)
al. 2003          less intense CIMT (n=7)   weekends) or 3 hrs/day         Wolf Motor Function Test (-)
USA                                         on weekdays only) x 10          Assessment of Motor and
5 (RCT)                                               days                       Process Skills (-)
                                                                              (All at end of therapy)
Lin et al.        CIT vs. traditional therapy 2 hrs/day x 5days/week              Fugl-Meyer (+)
2008/09/10          (neurodevelopmental)                x 3 weeks                     FIM (+)
5 (RCT)                                                                       Motor Activity Log (-)
Sterr et al.       Longer CIMT + ‘shaping 6 hrs/day for a target of             Motor Activity Log
2002                 procedure’ (n=7) vs.         90% of waking time or     (+ after treatment and at
4 (RCT)            shorter CIMT + ‘shaping          3hrs/day x 2 wks.      weekly follow-up for 4 wks)
                       procedure’ (n=8)                                     Wolf Motor Function Test
                                                                            (+ after treatment and at
                                                                           weekly follow-up for 4 wks)
- Indicates non-statistically significant differences between treatment groups
+ Indicates statistically significant differences between treatment groups

10. Upper Extremity Interventions                                                          pg. 59 of 150
Thirteen RCTs evaluated the benefit of                                               those individuals with sensory deficits
CIMT in the subacute or chronic phase                                                and neglect consistent with a “disuse”
of stroke. There was great variability                                               concept. The selective benefit within
in the chronicity of stroke. The median                                              certain subsets of stroke patients
time of stroke onset in the study by                                                 raises concerns as to the treatment’s
Taub et al. (1993) was 4+ years, while                                               generalizability. Promising research
patients in the EXCITE Trial, the                                                    trends are that more recent studies
largest and most methodologically                                                    have included improved control
rigorous study recruited patients within                                             treatments that have helped determine
3 to 9 months following stroke. The                                                  the specificity of treatment effects to
results from these RCTs reported a                                                   CIMT, and patients whose levels of
positive treatment effect for the                                                    impairment and disability are more
patients receiving constraint-induced                                                typical of all who participate in stroke
movement therapy (Figure 10.3).                                                      rehabilitation. Results from the Lin et
However, functional benefit appears to                                               al (2010) trial demonstrated that the
be largely confined to those individuals                                             apparent benefits of CIMT could be
with some active wrist and hand                                                      attributed to plastic reorganization, as
movement. It is particularly useful for                                              evidenced by fMRI data.

                                          Figure 10.3 Action Research Arm (ARA) for Forced Use vs. Bimanual
                                                                  (Van der Lee 1999)
     ARA value (range 0-57 points)

                                                                                           38              38.5
                                                                     30                    30.8             30.7
                                                 0 weeks               3 weeks                  6 weeks           1 year

                                                                              Forced Use     Bimanual

The majority of studies included                                                     severe upper extremity paresis and
patients with less severe levels of                                                  reported significant improvement in
impairment, typically characterised by                                               functional ability, although noting that
a patients’ ability to demonstrate at                                                there were issues of compliance with
least 200 of wrist extension and 100 of                                              some patients.
each metacarpophalangeal and
interphalangeal joint of the involved                                                Two studies assessed the benefit of
upper extremity. Bonifer et al. (2005)                                               modified CIMT provided in the acute
included patients with moderate-to                                                   phase of stroke (Table 10.21).

10. Upper Extremity Interventions                                                                                   pg. 60 of 150
Table 10.21 Summary of RCTs Evaluating Modified CIMT in the Acute Phase
Following Stroke
   Author/           Intervention             Intensity/Duration             Main Outcome(s)
PEDro Score                                                                       Result
Myint et al.    mCIMT vs. traditional        4 hrs/day x 10 days         Total ARAT and subscale
2008               rehabilitation                                               scores (+)
7 (RCT)                                                                          MAL (+)
Page et al.         mCIMT (n=5) vs.               Less affected arm     Between group comparisons
2005            traditional motor rehab        restrained for 5 hrs/day        not reported
5 (RCT)           (n=5) , consisting of       x 5 days/wk x 10 weeks
                   structured therapy         or motor rehabilitation of
                   emphasizing more            the upper extremity for
              affected arm use in valued 0.5hr x 3days/wk x 10
               activities strategies with               weeks
               the unaffected limb. The
               TR regimens occurred 3
                 d/week for 10 weeks.
- Indicates non-statistically significant differences between treatment groups
+ Indicates statistically significant differences between treatment groups

          Figure 10.4 Effect of Constraint-Induced Movement
               Therapy (CIMT) vs. Usual Care on WMFT
           Performance Time at Baselin e Through 12-Month
                                                                   p<0.001 Change from
                                                                    baseline to 12-mo
 12-mo Follow-up

  8-mo Follow-up                                                                         Usual Care
  4-mo Follow -up



                     0           5          10         15          20              25
                                     Performance Time (s)

Page et al. (2005) reported that the                    small group of Chinese stroke patients
mean change in Fugl-Meyer and                           where rest and recuperation have
Action Research Arm tests for patients                  been traditionally favoured following
randomized to the m-CIMT group were                     an acute illness compared with
greater than previously reported for                    intensive therapy.
patients receiving m-CIMT therapy in
the subacute period of stroke. The                      Six RCTs evaluated the effectiveness
treatment was well tolerated and                        of mCIMT in the subacute and chronic
there were no losses at follow-up.                      phases of stroke (Table 10.22). While
Myint et al. (2008) reported a                          all the studies reported improvements
beneficial effect of treatment in a                     in functional outcomes, the sample

10. Upper Extremity Interventions                                                          pg. 61 of 150
sizes were small. Szaflarski et al.                    reorganization, as demonstrated
(2006) suggested that mCIMT is                         through changes in fMRI.
beneficial as it induces cortical

 Table 10.22 Summary of RCTs Evaluating Modified CIMT in the Chronic Phase
Following Stroke
   Author/         Intervention              Intensity/Duration            Main Outcome(s)
PEDro Score                                                                     Result
Lin et al.      Modified CIMT vs.         6 hrs/day x 5 hrs/day x             MAL (+)
2007            traditional rehab                 3 wks.                       FIM (+)
7 (RCT)
Wu et al.   Modified CIMT vs. regular     2 hrs/day x 5 days/wk x              MAL (+)
2007 b)       occupational therapy                 3 wks                       FIM (+)
6 (RCT)
Page et al.  Modified CIT + physical      5 hrs/day x 5 days/wk x            Fugl-Meyer
2004        and occupational therapy      10 wks. or 1 hr therapy    (+ in modified CIMT group
6 (RCT)       (n=7) vs. traditional       3x/wk x 10wks or 0 hrs       only after intervention)
               rehab (n=4) vs. no                of therapy.          Action Research Arm test
                 therapy (n=6)                                       (+ in modified CIMT group
                                                                       only after intervention)
Page et al.     Modified CIT + physical   5 hrs/day x 5 days/wk x            Fugl-Meyer
2002           and occupational therapy 10 wks. or 1 hr therapy      (+ in modified CIMT group
5 (RCT)          (n=4) vs. traditional    3x/wk x 10wks or 0 hrs       only after intervention)
              rehabilitation (n=5) vs. no        of therapy.          Action Research Arm test
                    therapy (n=5)                                    (+ in modified CIMT group
                                                                       only after intervention)
Page et al.   Modified CIMT (n=13) vs. 5 hrs/day x 5 days/wk x              Fugl-Meyer (-)
2008             conventional therapy       10 wks. or 1/2 hr       Action Research Arm Test (+)
5 (RCT)         (n=12) vs. no therapy    therapy 3x/wk x 10wks
                        (n=10)             or 0 hrs of therapy.
Wu et al.     Modified CIMT (mCIMT) + 2 hrs/day x 5 days/wk x           Fugl-Meyer Assessment (+)
2007           a restraining mitt on the          3 wks                     FIM instrument (+)
5 (RCT)        unaffected hand (n=13)                                      Motor Activity Log (+)
                 or traditional therapy                                     Stroke Impact Scale
                        (n=13).                                        (+ improvement in strength,
                                                                          ADLS/IADLs, and stroke
- Indicates non-statistically significant differences between treatment groups
+ Indicates statistically significant differences between treatment groups

In addition to the studies that                        Among the 2 RCTs examining this
examined either modified CIMT or                       intervention, neither demonstrated a
CIMT, 4 studies were included that                     statistically significant benefit
assessed the use of forced-use                         (Ploghman & Corbett 2004, Hammer &
therapy, in which the unaffected arm                   Lindmark, 2009 a,b), although both
was restrained without a shaping, or                   reported trends in favour of the
more intense exercise component                        treatment group. The sample sizes in
(Wolf et al. 1989, Ploughman &                         both of these studies were small. A
Corbett 2004, Burns et al. 2008,                       fifth study, evaluated the effectiveness
Hammer & Lindmark 2009 a,b).                           of the continued use of a mitt

10. Upper Extremity Interventions                                                        pg. 62 of 150
following a 2-week course of CIMT                         mitt for several weeks either 3 months
(Brogårdh & Bengt 2006, 2009). There                      or 4 years following the initial course
was no additional benefit of therapy                      of treatment.
associated with the continued use of a

 Conclusions Regarding Constraint-                         patients with some active wrist and
 Induced Movement Therapy                                  hand movements, particularly those
                                                           with sensory loss and neglect.
 There is conflicting (Level 4) evidence
 of benefit of CIMT in comparison to                        Constraint-induced movement
 traditional therapies in the acute stage                   therapy is a beneficial treatment
 of stroke.                                                 approach for those stroke patients
                                                            with some active wrist and hand
 There is strong (Level 1a) evidence of                     movement.
 benefit of CIMT and mCIMT in
 comparison to traditional therapies in
 the chronic stage of stroke. Benefits
 appear to be confined to stroke

10.2.10 Mirror Therapy

Mirror therapy is a technique that uses                   the unaffected limb, blocking their
visual feedback about motor                               view of the affected limb, creating the
performance to improve rehabilitation                     illusion that both limbs are working
outcomes. It has been adapted from                        normally. It is believed that by
its original use for the treatment of                     viewing the reflection of the
phantom limb pain as a method to                          unaffected arm in the mirror that it
“re-train the brain” as a means to                        may act as a substitute for the
enhance upper-limb function following                     decreased or absent propreoceptive
stroke and to redcue pain. In mirror                      input.
therapy, patients place a mirror beside

Table 10.23 Studies Evaluating Mirror Therapy
 Author, Year                  Methods                                        Results
 PEDro Score
Altschuler et   9 subjects with stroke onset of > 6       Both raters agreed that 7/9 patients in the
al. 1999        mos were randomly assigned to spend       control group did not improve. Two patients
USA             the first 4 weeks using either a mirror   in the control group improved by 0.5 or 1
(RCT)           or transparent plastic then crossed       point. In the mirror group, at least one of the
                over to the other treatment for the       raters reported that every patient had
Letter-         next 4 weeks. Patients practiced for 15   improved by at least 0.5 points.
insufficient    min 2x/day 6 days a week, moving the
info to score   paretic hand as much as they were
                able while watching the unaffect arm
                in the mirror, or the paretic arm
                through the plastic. 2 Neurologists
                assessed change from baselinein
                movement abiliaty in terms of range of
                motion, speed and accuracy, using a -

10. Upper Extremity Interventions                                                          pg. 63 of 150
                 3 to + 3 scale (0 is no change)
Yavuzer et al.   40 inpatients all within one-year of        The scores of the Brunnstrom stages for the
2008             stroke were randomized to a program         hand and upper extremity and the FIM self-
Turkey           of either 30 minutes of mirror therapy      care score improved more in the mirror group
7 (RCT)          (n=20) a day consisting of wrist and        than in the control group after 4 weeks of
                 finger flexion and extension                treatment (by 0.83, 0.89, and 4.10,
                 movements or sham therapy (n=20) in         respectively; all P<.01) and at the 6-month
                 addition to conventional stroke             follow-up (by 0.16, 0.43, and 2.34,
                 rehabilitation program, 5 days a week,      respectively; all P<.05).There were no
                 2 to 5 hours a day, for 4 weeks.            significant differences in change scores
                 Outcomes including the modified             between the groups at either the end of
                 Ashworth Scale (MAS) and the                treatment or at follow-up. (4 week change
                 Brunnstrom stages of motor recovery         MAS: 0.12 vs. 0.11, p=0.89; 6 months: 0.18
                 were assessed before and after              vs. 0.21, p=0.876).
                 treatment and at 6 months.
Dohle et al.     36 patients with severe hemiparesis         There were no significant differences in the
2009             due to first-ever ischemic stroke in the    mean FM subscores of any of the FM sub
Germany          territory of the middle cerebral artery     scores at the end of treatment. In the
7 (RCT)          were enrolled, no more than 8 weeks         subgroup of 25 patients with distal plegia at
                 after the stroke. They completed a          the beginning of the therapy, MT patients
                 protocol of 6 weeks of additional           regained more distal function than CT
                 therapy (30 minutes a day, 5 days a         patients. Furthermore, across all patients, MT
                 week), with random assignment to            improved recovery of surface sensibility.
                 either mirror therapy (MT) or an
                 equivalent control therapy (CT). The
                 primary outcome measures were the
                 Fugl-Meyer subscores for the upper
                 extremity, (arm, hand and finger
                 function) were evaluated before and
                 after treatment.
                                                            rehabilitation. In the 2 RCTs
Mirror therapy is a treatment for                           examined, there was an improvement
which there is a limited body of                            in motor function reported in one trial,
evidence in its application to stroke                       but not the other.

 Conclusions Regarding Mirror Therapy
                                                             function following stroke and moderate
 There is conflicting (Level 4) evidence                     (Leve 1b) evidence that it does not
 that mirror therapy improves motor                          reduce spasticity.

10.2.11 Feedback

As with athletic performance,                               pressure, and/or proprioception.
feedback can be used as a means to                          Extrinsic feedback can augment the
improve motor learning following                            effect of intrinsic and refers to
stroke. There are two types of                              feedback provided from the
feedback, intrinsic and extrinsic.                          environment. Extrinsic feedback can
Intrinsic feedback refers to the use of                     be both verbal and non-verbal.
a person’s own sensory-perceptual                           Comments from a therapist would be
information to enhance their                                an example of extrinsic verbal
performance during a given task. It                         feedback. Extrinsic feedback can be
may take the form of touch, sound,                          further classified as either knowledge

10. Upper Extremity Interventions                                                            pg. 64 of 150
of results (KR) or knowledge of                               about weight distribution that can
performance (KP). KR is often given at                        improve balance performance and
the end of a task and is feedback                             auditory feedback can improve sit-to-
related to the outcome of the                                 stand performace.
performance of that task. A patient’s
time performing a timed-walk test is
an example of KR. KP is information                           Subramanian et al. (2010) included
about the movement characteristics                            the results from 9 studies and
that led to the performance outcome.                          reported that there was evidence that
                                                              external feedback, particularly KP, in
Two reviews have been published on                            the forms of verbal, virtual
the topic of feedback. Van Vliet & Wulf                       environments, videotape, robotics,
(2006) concluded that visual feeback                          audition or vision, improved motor
can be used to provide information                            learning of the more affected limb.

Table 10.24 Studies Evaluating Feedback
 Author, Year                     Methods                                         Results
 PEDro Score
Winstein et al.   40 stroke subjects approximately 2          Both stroke and control groups demonstrated
1999              years post onset) and 40 age-matched        significant improvement in accuracy and
USA               controls practiced a rapid, spatially and   consistency over practice with relative
No Score          temporally constrained programmed           persistence of these changes during
                  action under one of two augmented           retention. There were no differences between
                  feedback practice conditions.               groups (stroke vs control) in performance
                  Participants in the stroke group used       patterns across trials for acquisition,
                  the upper limb ipsilateral to the lesion.   retention, or reacquisition phases. In
                  After an extended practice period (198      addition, there were no differential effects of
                  trials), acquisition, retention, and        the two augmented feedback conditions on
                  reacquisition performance was               performance and no interactions of feedback
                  assessed for accuracy and consistency       condition with group. However, independent
                  and compared over trials, between           of feedback condition, the stroke group
                  groups and feedback conditions.             performed with more error than did the
                                                              control group during all experimental phases
                                                              (i.e., acquisition, retention, reacquisition)
Cirstea et al.    37 chronic stroke patients, no longer       Patients in the KP group made faster, less
2006              receiving active therapy were               segmented and less variable movements,
Canada            randomized to one of 3 groups: 1)           which were maintained at follow-up.
6 (RCT)           Knowledge of Results (KR) (n=14)
                  practiced a reaching task involving 75
                  repetitions per day, 5 days per week
                  for 2 weeks, with 20% KR about
                  movement precision; 2) Knowledge of
                  Performance (KP) (n=14) trained on
                  the same task and schedule as KR but
                  with faded KP about joint motions; and
                  (3) control (C) (n=9) practiced a
                  nonreaching task. Outcomes were
                  assessed before and after treatment
                  and at 1-month and included motor,
                  gognitive, neuropshycological and

10. Upper Extremity Interventions                                                              pg. 65 of 150
               kinematic measures.
Cristea &      28 chronic stroke survivors were            Patients in the KP group increased the range
Levin          randomly assigned to 2 groups that          of shoulder movements and improved elbow
2007           practiced 10 sessions of 75 pointing        and shoulder temporal interjoint co-ordination
6 (RCT)        movements. During practice, groups          to a greater degree than patients in the KR
               received either 20% Knowledge of            group, immediately following treatment.
               Results (KR) about movement
               precision or faded (26.6% average)
               Knowledge of Performance (KP) about
               arm joint movements. A nondisabled
               control group (n = 5) practiced the
               same task with KR. Outcomes were
               assessed before and after treatment
               and at 1-month and included Fugl-
               Meyer (FM) scale, the Composite
               Spasticity Index (CSI) and the the
               TEMPA and kinematic measures.
Gilmore &      10 patients undergoing inpatient            There was no significant difference between
Spaulding      rehabilitation, who were unable to don      the two groups at either the start, or the end
2007           their own socks and shoes                   of treatment. Patients in both groups
Canada         independently, received a maximum of        improved. However, the group that received
5 (RCT)        10 training sessions in addition to         videotape feedback thought they performed
               routine therapy. Patients were              better and were more satisfied with their
               randomized to one of two groups. In         ability to don shoes.
               the experimental group, participants
               were videotaped and received both
               verbal feedback and viewed the tapes
               from the training sessions. Patients in
               the control group received verbal
               feedback only. The main outcome,
               assessed before and after each
               treatment, was the socks and shoes
               subtests of the Klein-Bell Activities of
               Daily Living Scale. The Canadian
               Occupational Performance Measure
               was assessed before and after
Piron et al.   50 chronic stroke subjects with upper       In the per protocaol analysis, there were no
2010           arm paresis were randomized to              significant differences between groups on any
Italy          receive a program of either reinforced      of the outcomes assessed. In an intention-to-
8 (RCT)        feedback in a virtual envirnoment or        treat analysis with imputed data for 3 missing
               conventional therapy, based on Bobath       subjects, from the control group, assuming
               principles for 1 hour of therapy daily (5   the best possible outcome, there was a
               days/week) x 4 weeks. The outcomes          significant difference in FM scores following
               were assessed before and after              treatment, favouring the treatment group.
               treatment and included FIM Fugl Meyer
               –UE (FM) and kinematic analysis of
                                                            There is strong (Level 1a) evidence
 Conclusions Regarding Extrinsic                           that extrinsic feedback helps to
 Feedback                                                  improve motor learning following

10. Upper Extremity Interventions                                                          pg. 66 of 150
10.3 Robotic Devices For                     the most effective technique in
Movement Therapy                             patients with mild to moderate
                                             impairments, active- assisted
Robotic devices can be used to assist        movement (with robotic devices) may
the patient in a number of                   be beneficial in more severely
circumstances. First of all, the robot       impaired patients…especially during
can aid with passive range of motion         the acute and subacute phases when
to help maintain range and flexibility,      patients are experiencing spontaneous
to temporarily reduce hypertonia or          recovery.” Krebs et al. (2002) noted
resistance to passive movement. The          that robotic devices rely on the
robot can also assist when the patient       repetition of specific movements to
has active movements, however,               improve functional outcomes. While
cannot complete a movement                   the majority of robotic devices focus
independently. Robotics may be most          on retraining of the upper extremity,
appropriate for patients with dense          specifically shoulder, elbow and wrist
hemiplegia, although robotics can be         movements, researchers have recently
used with higher-level patients who          begun to investigate the potential use
wish to increase strength by providing       of robotic devices for the fingers and
resistance during the movement.              legs (Krebs et al. 2002, Lum et al.
According to Lum et al. (2002), “even        2002).
though unassisted movement may be
A recent systematic review of robot-         (Hesse et al. 2005) was removed in
aided therapy on recovery of the             sensitivity analysis, there was a
hemiparetic arm on recovery of the           significant treatment effect. In the 5
hemiparetic arm was conducted                studies that evaluated improvement in
(Prange et al. 2006). The authors            ADL, no significant beneficial
included the results from 8 studies          treatment effect was found.
evaluating the MIT-Manus, MIME and
ARM Guide and concluded that robotic         A recent Cochrane review, authored
devices improved short and long term         by Mehrolz et al. (2008) included the
motor function of the paretic shoulder       results from 11 trials (328 subjects)
and elbow beyond that which could be         evaluating electromechanical and
achieved through therapy alone.              robot-assisted arm training devices.
                                             Compared with routine therapy,
Kwakkel et al. (2008) conducted a            usually conventional physical therapy,
systematic review of RCTs that               the authors reported no significantly
evaluated robotic devices in the             greater improvement in activities of
management of upper extremity                daily living, but did find greater gains
hemiplegia following stroke. The             in motor function (SMD=0.68; 95% CI
results from 10 studies involving 218        0.24 to 1.11, p=0.002) and motor
subjects were identified. Pooling the        strength (SMD=1.03; 95% CI 0.29 to
results from 7 trials assessing              1.78, p=0.007).
improvement in motor function
revealed a nonsignificant benefit of         The results of studies that have
robotic treatment. The summary effect        assessed a variety of these devices
size was 0.65 (95% CI -0.02 to 1.33,         are presented in tables 10.25 to
p=0.06). When one of the studies             10.32.

10. Upper Extremity Interventions                                        pg. 67 of 150
                                                          patient’s hand in a horizontal plane,
10.3.1 MIT-Manus                                          while visual, auditory and tactile
                                                          feedback is provided during goal-
MIT-Manus was one of the first robotic                    directed movements. A commercially
devices to be developed. It features a                    available unit (InMotion2) of this
2-degree-of-freedom robot                                 device is also available.
manipulator that assists in shoulder
and elbow movement by guiding the

Table 10.25 Studies Evaluating MIT-Manus Robotic Device
 Author, Year                  Methods                                      Results
 PEDro Score
Volpe et al.    20 patients admitted to rehabilitation  The robot-trained group showed significant
1999            were randomized to either robot or      improvement on the MSS (motor status
USA             sham treatment. All patients received   scores) for shoulder/elbow at discharge and 3
6 (RCT)         similar standard physical and           year follow up and from admission to
                occupational therapy. For 1 hr/day, 5   discharge for the Motor Power score. Both
                days/wk, patients in the robot group    groups showed comparable changes in the FM
                moved the handle of a robot, which      for shoulder/elbow and for wrist/hand, and
                provided visual feedback of the         MSS for the wrist/hand over both intervals.
                movement. The robot provided            These data show that the advantages
                assistance if the patient did not       conferred by robot training were specific to
                produce movement. Nearly 3 years        the muscle groups trained and were
                after discharge and without further     persistent. The motor scores for both groups
                robot training, 12 of the patients were improved 3 years after stroke, which showed
                re-examined.                            that for some patients, improvements may
                                                        continue long after discharge.
Volpe et al.    56 patients with stroke and             At the end of treatment, the robot-trained
2000            hemiparesis or hemiplegia received      group demonstrated improvement in motor
USA             standard poststroke multidisciplinary   outcome for the trained shoulder and elbow
6 (RCT)         rehabilitation, and were randomly       (Motor Power score, p < 0.001; Motor Status
                assigned either to receive robotic      score, p < 0.01) that did not generalize to the
                training (at least 25 hours) or         untrained wrist and hand. The robot-treated
                exposure to the robotic device without group also demonstrated significantly
                training. Patients were assessed before improved functional outcome (Functional
                treatment began and at the end of       Independence Measurement–Motor, p <
                treatment, with the upper extremity     0.01).
                component of the Fugl-Meyer Motor
                Assessment, the Motor Status score,
                the Motor Power score, and Functional
                Independence Measurement.
Fasoli et al.   Pre-post test design. 20 patients       Significant gains were found on the Fugl-
2003            diagnosed with a single, unilateral     Meyer test, Motor Status Scale (MSS) scores,
USA             stroke within the past 1-5 yrs, with    and motor power scores from admission to
No score        persistent hemiparesis. Robotic         discharge. The progressive-resistive group
                therapy was provided 3x /wk for 6       experienced non-specific improvements on
                wks. Subjects able to reach robot       wrist and hand MSS scores that were not
                targets at baseline and after 3 weeks   observed in the sensorimotor group.
                of intervention were randomly
                assigned to sensorimotor or
                progressive –resistive robotic group
                therapy. Robotic therapy consisted of

10. Upper Extremity Interventions                                                        pg. 68 of 150
                 goal-directed, planar reaching tasks to
                 exercise the hemiparetic shoulder.
                 Clinical evaluations occurred before
                 treatment, and after 3 and 6 weeks of
Ferraro et al.   34 chronic stroke patients participated    Both moderately and severely impaired
2003             in 18 sessions (one-hour x 3x/week x       subjects significantly improved their FM-
USA              6 weeks) involving a robotic device.       (shoulder/elbow coordination) scores and MP
No Score         Baseline/post intervention and 3           scores. Subjects with moderate impairment
                 month follow-up assessments included       improved FIM scores associated with upper
                 Fugl-Meyer (FM) Motor Status Scale,        extremity function.
                 Motor Power (MP) scale, Modified
                 Ashworth Scale and FIM. Subjects
                 were stratified as moderate or severe
Fasoli et al.    A follow-up study to Fasoli et al 2003,    Statistically significant gains from admission
2004             using a larger sample. 42 chronic          to discharge and from admission to follow-up
USA              stroke patients with hemiparesis           were found on the FMA, MSS score for
No Score         received robotic therapy for the paretic   shoulder and elbow, and motor power score
                 upper limb consisted of either             at four months.
                 sensorimotor active-assistive exercise,
                 or progressive-resistive training during
                 repetitive, planar reaching tasks, 3
                 times a week for 6 weeks. Outcome
                 measures included Modified Ashworth
                 Scale, Fugl-Meyer Assessment (FMA),
                 Motor Status Scale (MSS) score, and
                 Medical Research Council motor power
Stein et al.     Stroke patients (n=46) capable of          The incorporation of robot-aided progressive
2004             doing resistance training were             resistance exercises into a program of robot-
USA              randomized to receive either robot-        aided exercise did not favourably or
5 (RCT)          aided progressive resistance training      negatively affect motor control or strength.
                 or active-assisted robot-aided
                 exercises. There was no control for
                 robot assistance.
Day et al.       12 moderate to severe chronic stroke       There was a significant reduction in mean
2005             survivors were randomized to one of        AMAT scores between the beginning and end
USA              two treatments: robotics and motor         of treatment for subjects in the ROB-ML
5 (RCT)          learning (ROB-ML) or functional            group but not for those in the control group
                 neuromuscular stimulation and motor        (-475 vs. -242).
                 learning (FNS-ML). Treatment was 5
                 h/d, 5 d/wk for 12 wk. ROB-ML group
                 had 1.5 h per session devoted to
                 robotics shoulder and elbow (S/E)
                 training. FNS-ML had 1.5 h per session
                 devoted to functional neuromuscular
                 stimulation (surface electrodes) for
                 wrist and hand (W/H)
                 flexors/extensors. The primary
                 outcome measure was the functional
                 measure Arm Motor Ability Test

10. Upper Extremity Interventions                                                           pg. 69 of 150
Finley et al.   15 stroke patients (<6 months post         From baseline to 3 months post treatment
2005            stroke) with severe (Fugl-Meyer <15)       there were statistically significant
USA             upper limb (UL) paresis underwent          improvement seen for FM and MPA. Also,
No Score        robot-assisted UL rehabilitation.          improvement was seen with robot-assisted
                Therapy included 2 1-hour sessions, 3      rehabilitation for reduced aiming error,
                days/week for 3 weeks and included         movement duration, mean speed and mean
                goal-orientated reaching tasks.            peak speed ratio (p<0.01).
                Outcomes measures were the Wolf
                Motor Function Test (WMFT), the Motor
                Power Assessment (MPA), the Fugl-
                Meyer (FM) UL Assessment, Stroke
                Impact Scale, 5 robot-derived
                measures (aiming error, mean speed,
                peak speed, mean: peak speed ratio,
                movement duration)
MacClellan et   30 chronic stroke patients with            From baseline to end of treatment, there
al. 2005        moderate to sever impairment               were significant improvements in total FM and
USA             received 18 sessions over 3 weeks of       MPA scores. From Post treatment to follow-up
No Score        robot-assisted task-specific therapy       there were no significant changes in any of
                using a robot device (InMotion-2),         the outcome measures assessed.
                which focused on a reaching task
                involving the elbow and shoulder.
                Primary outcomes included the upper-
                limb Motor Status Score, the Wolf
                Motor Function test, Motor Power
                Assessment (MPA) and the Fugl Meyer
                (FM), assessed at baseline, end of
                treatment and 3-month follow-up.
Daly et al.     13 chronic stroke subjects living in the   The results of between group comparisons
2005            community were randomized to               are not reported. Subjects in the robotic
USA             receive a 12-week program (5 hrs/day       group achieved significant improvements on
5 (RCT)         x 5 days/week) of robotics (InMotion2)     the AMAT (Total) and AMAT (S/E subsection),
                + motor learning or motor learning +       FM scores and target accuracy. Subjects in
                functional neuromuscular stimulation.      control group improved significantly on AMAT
                Outcome measures, assessed before          (W/H) and FM scores only.
                and after treatment, included the Arm
                Motor Ability Test (AMAT) +
                shoulder/elbow (SE) and wrist hand
                (WH) subsections, the Fugl-Meyer (FM)
                scale and motor control measures of
                target accuracy.
Volpe et al.    21 chronic stroke patients were            Patients in both groups demonstrated
2008            randomized to receive a course of          improvement over time, which was
USA             intensive upper-extremity treatment        maintained at 3 months; however, there were
5 (RCT)         that was provided by either a therapist    no significant between group differences on
                or a robotic device (InMotion2).           either the primary or secondary outcomes.
                Treatment consisted of 1 hr sessions,
                3x/week for 6 weeks. Primary outcome
                was the Fugl-Meyer (FM) score for
                shoulder/elbow. Secondary outcomes
                were the FM wrist/hand and the Motor
                Power Scale for Shoulder/elbow.
                Assessments were conducted monthly
                for 3 months.

10. Upper Extremity Interventions                                                         pg. 70 of 150
Rabadi et al.    30 acute stroke patients (< 5 weeks)       After adjusting for age, stroke type and
2008             admitted for inpatient rehabilitation      outcome measures at baseline, a similar
USA              were randomized to 1 of 3 groups: 1)       degree of improvement in the discharge
5 (RCT)          Occupational therapy (OT) group            scores was found in all of the primary
                 (control) (n = 10), 2) arm ergometer       outcome measures. Overall, the OT group
                 (n = 10) or 3) robotic (n = 10)            experienced the greatest gains.
                 therapy group. All patients received
                 standard, inpatient, post-stroke
                 rehabilitation training for 3 hours a
                 day, plus 12 additional 40-minute
                 sessions of the activity-based therapy.
                 The primary outcome measures,
                 assessed before and after treatment,
                 were discharge scores in the Fugl-
                 Meyer (FM) Assessment Scale for
                 upper limb impairment, Motor Status
                 Scale and FIM.
Lo et al. 2010   127 patients with moderate-to-severe       At 12 weeks, subjects in the robot assisted
USA              upper-limb impairment 6 months or          group had gained more FM points, compared
7 (RCT)          more after a stroke, were randomly         to subjects in the usual care group (1.11 vs. -
                 assigned to receive intensive robot-       1.06, p=0.08). Subjects in the intensive
                 assisted therapy (n=49), intensive         therapy group gained more FM points
                 comparison therapy (n=50), or to           compared with subjects in the robot-assist
                 usual care (n=28). Therapy consisted       group (4.01 vs. 3.87, p=0.92). No other
                 of 36, 1-hour sessions over a period of    treatment comparisons were significant at 12
                 12 weeks. The primary outcome was          weeks. No serious adverse events were
                 the Fugl-Meyer Assessment (FM) at 12       reported.
                 weeks. Secondary outcomes were
                 scores on the Wolf Motor Function Test
                 and the Stroke Impact Scale.
                 Secondary analyses assessed the
                 treatment effect at 36 weeks.

                                                           bilateral shoulder or elbow movement.
10.3.2 Mirror-Image Motion Enabler                         The unit applies force to the more
Robots (MIME)                                              affected forearm during goal-directed
MIME robotic devices were developed
to enable unrestricted unilateral or

Table 10.26 Studies Evaluating MIME Robotic Device
 Author, Year                   Methods                                        Results
 PEDro Score
Burgar et al.    21 subjects were randomized to a          There were no significant differences between
2000             course of treatment using either a        the groups on either of the ADL assessments
USA              robotic device (n=11) or to a control     while subjects in the robotic group exhibited a
5 (RCT)          group (n-10) that received a physical     trend towards greater improvement in FM
                 therapy program of stretching, weight     scores. These differences achieved statistical
                 bearing, facilitation games and           significance if only the shoulder and elbow
                 activities. 24, 1 hr sessions were        portions of the FM test were considered.
                 provided over 2 months. Motor
                 function was assessed using the Fugl-

10. Upper Extremity Interventions                                                           pg. 71 of 150
                Meyer (FM) Assessment and ADLs
                were assessed using FIM and BI.
Lum et al.      27 patients with chronic hemiparesis   After the first and second months of
2002            (> 6 months post-stroke) were          treatment, the robot group had significantly
USA             randomly allocated to receive either   larger improvements in the proximal
6 (RCT)         robot assisted movement training or    movement portion of the Fugl-Meyer test. The
                conventional therapy. The robot        robot group also had larger gains in strength
                group practiced shoulder and elbow     and larger increases in reach extent after 2
                movements assisted by a robot          months of treatment. At 6 months, no
                manipulator while the control group    significant differences were seen between the
                received NDT and 5 minutes of          two groups on the Fugl-Meyer test, however,
                exposure to the robot in each session. the robot group did have significantly larger
                                                       improvements on the FIM.
Lum et al.      30 subacute stroke patients were       Significantly greater gains were attained at
2006            randomized to receive 1 of 4           post treatment for the robot-combined group
USA             treatments: i.) robot-unilateral group in proximal FM and MSS synergy scale
4 (RCT)         (n=9), ii.) robot-bilateral group      compared to the control group. However,
                (n=5), iii.) robot-combined group –    these gains were not maintained at 6 months
                unilateral + bilateral (n=10), or iv.) follow-up. Also, a significantly greater
                conventional therapy (n=6). Over 4     improvement was seen for the robot-unilateral
                weeks patients in each group received group compared to the robot-combined group
                15 one-hour treatment sessions. Main for distal FM (P<0.05).
                measures included Fugl-Meyer
                Assessment (FM), Motor Status Score
                (MSS), Functional Independence
                Measure (FIM), modified Ashworth
                scale and were assessed pre and post
                intervention and at 6 months.
Patel et al.    19 stroke patients with hemiplegia     There were no significant changes seen in the
2006            received a standardized passive        Hmax, Mmax or Hmax/Mmax ratio from pre-
USA             exercise program using a robotic arm to post-exercise responses.
No Score        targeting the upper extremity.
                Outcome Measures included
                Hoffmann relax (Hmax), motor
                response (Mmax), and Hmax/Mmax
                were attained before and after the
                exercise program.

                                                         to move the user’s hand along a linear
10.3.3 Assisted Rehabilitation and                       rail, which assists reaching in a
Measurement (ARM) Guide                                  straight-line trajectory.

This unit uses a motor and chain drive

Table 10.27 Studies Evaluating ARM Guide
 Author, Year                  Methods                                      Results
 PEDro Score
Kahn et al.     19 Chronic (>1 year post stroke)         Significant improvements were attained with
2006            patients were randomly assigned to       training for functional ability movement,
USA             receive 24 sessions of either active-    velocity and range of motion of supported
4 (RCT)         assistive reaching exercise using a      reaching and straightness of unsupported

10. Upper Extremity Interventions                                                       pg. 72 of 150
                robotic device (n=10) or a task-           reaching. No significant differences existed
                matched amount of reaching without         between groups. The group without assistance
                assistance (n=9). Both groups              with reaching exercises showed a greater
                completed an 8-week therapy                improvement in smoothness of arm movement
                program involving a total of 24, 45-       compared to the robot-assisted group.
                minute sessions. Main outcomes
                measures include the Rango Los
                Amigos Functional Test of Upper
                Extremity Function, range,
                smoothness and straightness of
                unsupported arm movement, and
                speed and range of supported
                reaching, assessed at baseline, post
                intervention and 6 months.

10.3.4 Bi-Manu-Track
This arm-training device enables
bilateral and passive and active
practice of forearm and wrist

Table 10.28 Studies Evaluating the Bi-Manu-Track Device
 Author, Year                  Methods                                         Results
 PEDro Score
Hesse et al.    12 stroke patients with chronic            MAS score of the wrist and fingers joints
2003            hemiparesis received daily additional      decreased significantly (P<.0125) from a
Germany         15 minute exercise program with a          median of 3 and 3 to 2 and 2.5. The RMA
No Score        robotic bilateral arm trainer, enabling    score minimally increased in 5 cases without
                passive and active practice of forearm     improvement in functional tasks. The median
                pronation and supination and wrist         RMA score before therapy was 2.0 and 2.0
                dorsiflexion and volarflexion.             after therapy. The effect of treatment was lost
                Outcome assessed included Modified         by 3 month follow-up.
                Ashworth scale scores, Rivermead
                Motor Assessment.
Hesse et al.    44 subacute stroke patients with           At the end of 3 months Fugl-Meyer scores
2005            severe arm paresis were randomized         among patients in the AT group improved to a
Germany         to computerized arm training (AT)          greater degree than those in the ES group.
8 (RCT)         enabling repetitive practice of passive    Upper limb motor power scores also improved
                and active bilateral forearm and wrist     more among patients in the AT group
                movement cycle (yielding 800               compared to the ES group.
                repetitions) or electromyography-
                initiated electrical stimulation (ES) of
                the paretic wrist extensor. 60-80
                wrist extensions were achieved with
                each ES session. The therapy was
                conducted for 20 minutes/5 days a
                week for 6 weeks.
Hesse et al.    54 patients enrolled in a                  Patients in both groups improved over time
2008            comprehensive inpatient rehabilitation     but there was no significant difference in FM
Germany         program, within 4-8 wks from stroke        scores between groups. FM scores improved
8 (RCT)         onset were randomized to practice          from a mean of 8.8 at baseline to 28.9 at
                with an arm trainer (AT) or to receive     follow up (AT group) and from 8.6 to 18.4 (ES

10. Upper Extremity Interventions                                                           pg. 73 of 150
                 electrical stimulation (ES) (75 Hz, 0.5   group). No patient could transport a block
                 msec, 0-80 mA). The arm trainer           initially, but at completion significantly more
                 device (Reha-Slide) consists of 2         arm trainer patients were able to transport at
                 handles spaced .75 m apart and            least three blocks (five vs. zero, P = 0.023).
                 connected by a rod and mounted on 2
                 parallel tracks. The patient can move
                 the handles forwards and backwards,
                 as well as sideways. The handles are
                 yoked. Both groups received
                 treatments for 20-30 mins, 5x/week x
                 6 wks (30 sessions). Primary outcome
                 was the Fugl-Meyer (FM) assessment.
                 Secondary outcomes were the Box
                 and Block test, the Medical Research
                 Council and the modified Ashworth
                 scale, assessed at enrollment, after 6
                 wks, and at 3-mos follow-up.

                                                           sensorimotor stimulation. The device
10.3.5 Neuro-Rehabilitation-Robot                          can perform spatial movements of the
(NeReBot)                                                  shoulder and elbow, is portable and
                                                           can be used when the patient is either
The NeReBot device was developed in                        prone or sitting.
Italy designed to produce

Table 10.29 Studies Evaluating the NeReBot Device
  Author, Year                  Methods                                        Results
  PEDro Score
Masiero et al.   20 acute stroke patients with             At the end of treatment patients in the
2006             hemiplegia or hemiparesis all received    experimental group achieved significantly
Italy            traditional multidisciplinary             better scores on the shoulder/elbow section of
5 (RCT)          rehabilitation following stroke and       the FM (shoulder elbow coordination) (p<0.05)
                 they were randomized to receive           and the motor FIM (p<0.02) compared to
                 either additional sensory motor           patients in the control group. These
                 training 4 hours per week for 4 wks       improvements were still evident at 3-month
                 or exposure to a robotic device           follow-up, as well as significantly greater
                 without training. Assessments             improvement in MI scores (p<0.04)
                 conducted before/after treatment and
                 at 3 months included the Fugl-Meyer
                 scale (FM), upper-Motricity Index
                 (MI), motor FIM and Medical Research
                 Council (MRC) scale.
Masiero et al.   35 acute stroke patients were             There were significant gains made for the
2007             randomly assigned to either an            treatment group in functional recovery and
Italy            experimental group (n=17) who             motor impairment of the upper extremity
5 (RCT)          received an additional 5 weeks of         following robot-assisted training seen in MRC
                 early sensorimotor robotic training for   deltoid and biceps, FMA for the upper proximal
                 4 hours/ wk or a control group            extremity, FIM motor score and FIM
                 (n=18) who performed exercises with       instrument. These gains were maintained at 3
                 their unaffected upper limb 30 min a      and 8 months follow-up. No significant
                 week for 2 weeks using the robotic        differences were found for the MAS and TCT.
                 device. Main outcome measures

10. Upper Extremity Interventions                                                            pg. 74 of 150
                included: Fugl-Meyer Assessment
                (FMA), the Medical Research Council
                (MRC) score (specifically MRC deltoid,
                MRC biceps and MRC wrist flexors),
                FIM instrument, Trunk Control Test
                (TCT) and Modified Ashworth Scale
                (MAS) assessed at the end of
                treatment and at 3 and 8 months
                                                          be effective in the prevention of
10.3.6 Continuous Passive Motion                          contractures. A single study evaluated
(CPM)                                                     the effectiveness of the device in
                                                          maintaining gleno-humeral joint
Passive range of motion is a standard                     mobility.
therapeutic technique, considered to

Table 10.30 A Study Evaluating a CPM Device
 Author, Year                  Methods                                      Results
 PEDro Score
Volpe et al.    2 studies reported. Study 1: Patients     Study 1: no significant differences in
2004            with acute stroke (within 3 weeks)        improvement between CPM and Control group.
USA             and hemiparesis randomly assigned
4 (RCT)         to one of 2 groups: i) Continuous         Study 2: Within group comparisons-both
                Passive Motion (CPM) group: 25            groups showed significant improvement in
                min/day of CPM training + standard        motor function (p=.01) and power in the
                post-stroke therapy (minimum 3.5          trained shoulder and elbow (p=.0001).
                hrs/day of physical, occupational, and
                speech therapy) or ii) Control group:
                standard post-stroke therapy + extra
                25 min/day of occupational therapy
                Study 2: Chronic stroke patients
                received interactive robotic arm
                training – lasted 1 hr/day, 3 days/wk
                for 6 wks. Patients assigned to two
                groups: i) moderate stroke severity,
                ii) severe stroke severity. Patients in
                both groups performed over 1000
                flexion extension movements of the
                paralyzed arm with gravity eliminated
                to move the end of a robotic arm in
                the direction represented by eight
                points of a compass. Evaluations
                included Fugl-Meyer for
                shoulder/elbow and coordination;
                Motor Power; Motor Status Scale of
                shoulder and elbow; Joint stability;
                Ashworth score; Fugl-Meyer for pain;
                Functional Independence Measure.
                Evaluations-3 baseline 2 months prior
                to start of training, a midpoint
                evaluation, and a discharge

10. Upper Extremity Interventions                                                        pg. 75 of 150
Hu et al.       27 hemiplegic subjects with chronic      Subjects in the Interactive group
Hong Kong       stroke were randomly assigned to         demonstrated statistically greater
2009            receive 20-sessions of wrist training    improvement in the following outcomes: FMA:
5 (RCT)         with a continuous electromyography       shoulder/elbow, MAS elbow and MAS wrist,
                (EMG)-driven robot (interactive          compared with subjects in the passive group.
                group, n = 15) and a passive motion
                device (passive group, n = 12),
                completed within 7 consecutive
                weeks. Training effects were
                evaluated with clinical scores by
                pretraining and posttraining tests
                (Fugl-Meyer Assessment [FMA] and
                Modified Ashworth Score [MAS]).

                                                         de-weighted through a free moving
10.3.8 GENTLE/s                                          elbow splint attached to the overhead
                                                         frame. The subject is connected to the
This is a three-degree of freedom                        device by a wrist splint. Exercises such
haptic interface arm with a wrist                        as hand-to-mouth and reaching
attachment mechanism, two                                movements can then be practised,
embedded computers, a monitor and                        while feedback is provided.
speakers and an overhead arm
support system. The affected arm is

Table 10.31 A Study Evaluating the GENTLE/s Device
 Author, Year                  Methods                                       Results
 PEDro Score
Coote et al.    Following a baseline period, 20           Each subject had a varied response to the
2008            subacute and chronic stroke patients     measurement and intervention phases. The
UK              were crossed over to receive robot-      rate of recovery was greater during the robot-
6 (RCT)         mediated therapy or a sling              mediated therapy phase than in the baseline
                suspension phase which acted as the      phase for the majority of subjects. The rate of
                control condition. In robot-mediated     recovery during the robot-mediated therapy
                therapy, they practiced three            phase was also greater than that during the
                functional exercises with haptic and     sling suspension phase for most subjects.
                visual feedback from the system. In
                sling suspension, they practiced three
                single-plane exercises. Each
                treatment phase was three weeks
                long. Main measures included range
                of active shoulder flexion, the Fugl-
                Meyer (FM) motor assessment and
                the Motor Assessment Scale (MAS)
                were measured at each visit.

10. Upper Extremity Interventions                                                         pg. 76 of 150
10.3.9 Other Devices

Table 10.32 Studies Evaluating Other Robotic Devices
 Author, Year                  Methods                                        Results
 PEDro Score
Kutner et al.   17 subjects 3 to 9 months poststroke      The combined therapy group had a
2010            were randomized to receive 60 hours       significantly greater increase in rating of mood
USA             of therapist-supervised repetitive task   from preintervention to postintervention
7 (RCT)         practice (RTP) or 30 hours of RTP +       compared with the RPT group while the RTP
                30 hours of robotic-assisted therapy      group had a greater increase in rating of social
                over 3 weeks. The primary outcome         participation from preintervention to follow-up
                measure was the Stroke Impact Scale       compared with the combined group.
                (SIS), assessed at baseline,
                postintervention and 2 months
                                                          therapy. Summarizing the results from
Discussion                                                the above studies can be challenging
Robotic therapies show promise for                        as a variety of devices were assessed
helping to provide safe and intensive                     using patients in the acute, sub acute
rehabilitation to patients who have                       and chronic stages of stroke.
mild to severe motor impairment.
Robotic devices can be used to                            A summary of the studies evaluating
provide rehabilitation that is of high-                   any form of robotic training is
intensity, repetitive and task-specific                   presented in table 10.33.
in a manner that is similar to physical

Table 10.33 Summary of Results From Studies Evaluating Sensorimotor Training:
Robotic Devices
      Author/          n                 Intervention                        Main Outcome(s)
    PEDro Score                                                                   Result
Kutner et al. 2010    17               Hand Mentor                    Stroke Impact Scale (Mood +)
Rabadi et al. 2009    30        Robot-unilateral group vs.             Fugl-Meyer Assessment (-)
5 (RCT)                      ergometer (bilateral) group vs.
                                   conventional therapy
Coote et al. 2008     20      Robot-aided therapy vs. sham          Fugl-Meyer Assessment (+)
8 (RCT)                             robot-aided therapy
Volpe et al. 2008     21        Robot assisted movement             Fugl-Meyer Assessment (-)
5 (RCT)                     training vs. conventional therapy
Lum et al. 2006       30    Robot-unilateral group vs. robot-         Fugl-Meyer Assessment
4 (RCT)                         bilateral group vs. robot-      (+ robot-groups vs. control at post-
                            combined group vs. conventional                 treatment)
                                          therapy             (+ robot-unilateral vs. robot-combined)
                                                                        Motor Status Score
                                                                (+ robot-groups vs. control at post-
                                                                              FIM (-)
                                                                    Modified Ashworth Scale (-)
Masiero et al. 2007   35       Additional Robotic Training 4          Fugl-Meyer Assessment
5 (RCT)                      hrs/wk x 5 wks vs. exposure to            (+ upper extremity)

10. Upper Extremity Interventions                                                           pg. 77 of 150
                            robotic device 30 min/wk x 2                        (- wrist)
                                         wks                         Medical Research Council
                                                                       (+ deltoid and biceps)
                                                                                (- wrist)
                                                                                FIM (+)
                                                                       Trunk Control Test (-)
                                                                    Modified Ashworth Scale (-)
                                                              (All outcomes are taken at the end of
Masiero et al. 2006   20   Additional sensorimotor robotic                Fugl-Meyer scale
5 (RCT)                    training or exposure to robotic            (+ shoulder and elbow)
                               device with no training               upper-Motricity Index (+)
                                                                Functional Independence Measure
                                                                       (+ motor component)
                                                                Medical Research Council scale (-)
Kahn et al. 2006      19      Robot-assisted training vs.           Biomechanical Assessment
4 (RCT)                          reaching unassisted                           (- range)
                                                                               (- speed)
                                                                           (- straightness)
                                                                       (+ smoothness at d/c)
                                                              Rangos Los Amigos Functional Test (-)
Hesse et al. 2005     44     Computerized arm training                Fugl-Meyer Assessment
8                           enabling repetitive practice v.                        (+)
                                 electrical stimulation
Fasoli et al. 2004    56      Robot assisted movement                  Fugl-Meyer Assessment
6                            training vs. robot exposure             (+ from adm-hospital d/c)
                                                               Motor Status score for shoulder/elbow
Lum et al. 2002       27       Robot assisted movement                 Fugl-Meyer Assessment
6 (RCT)                    training vs. conventional therapy            (+ at 1 and 2 months)
                                                                           ( - at 6 months)
                                                                     Upper Extremity Strength
                                                                           (+ at 2 months)
                                                                       Upper Extremity Reach
                                                                           (+ at 2 months)
                                                                         FIM (+ at 6 months)
Volpe et al. 2000     56   Robotic training (at least 25 hrs)             Motor Power score
6 (RCT)                       vs. exposure to the robotic              (+ shoulder and elbow)
                                 device without training                  (- wrist and hand)
                                                                          Motor Status score
                                                                       (+ shoulder and elbow)
                                                                          (- wrist and hand)
                                                                            FIM–Motor (+)
                                                               (All outcomes are taken at the end of
Volpe et al. 1999     20       Robot vs. sham treatment                  Motor status Scores
6 (RCT)                                                       (+ for shoulder/elbow at discharge and
                                                                            3 yr follow up)
                                                                ( - for wrist/hand at discharge and
                                                                          Motor Power Score
                                                              (+ for shoulder and elbow at discharge)

10. Upper Extremity Interventions                                                      pg. 78 of 150
                                                              (- for shoulder/elbow and wrist/hand at
                                                                      discharge and follow-up)
Stein et al. 2004     49     Robot-aided vs. robot assisted               Motor control (-)
5 (RCT)                                program                              Strength (-)

Volpe et al. 2004     32       Continuous Passive Motion                Fugl-Meyer pain (-)
4 (RCT)                            Device vs. control          Motor Status scores (- elbow/shoulder)
                                                                         Ashworth scale (-)

                      36       Interactive Robotic Therapy           Fugl-Meyer elbow/shoulder (+)
                                                                            Motor Power (+)
- Indicates non-statistically significant differences between treatment groups
+ Indicates statistically significant differences between treatment groups

 Conclusions Regarding Robotics in the                 According to Merians et al. (2002), a
 Rehabilitation of the Upper Extremity                 computerized virtual environment has
                                                       opened the doors to an “…exercise
 There is strong (Level 1a) evidence that              environment where the intensity of
 sensorimotor training with robotic                    practice and positive feedback can be
 devices improves upper extremity                      consistently and systematically
 functional outcomes, and motor                        manipulated and enhanced to create
 outcomes of the shoulder and elbow.
                                                       the most appropriate, individualized
                                                       motor learning approach. Adding
 There is strong (Level 1a) evidence that
 robotic devices do not improve motor                  computerized VR to computerized
 outcomes of the wrist and hand.                       motor learning activities provides a
                                                       three-dimensional spatial
                                                       correspondence between the amount
  Sensorimotor training with robotic
                                                       of movement in the real world and the
  devices improves functional and
                                                       amount of movement seen on the
  motor outcomes of the shoulder and
  elbow; however, it does not improve                  computer screen. This exact
  functional and motor outcomes of the                 representation allows for visual
  wrist and hand.                                      feedback and guidance for the

10.4 Virtual Reality Technology                        Henderson et al. (2007) conducted a
Virtual reality, also known as virtual                 systematic review that included 6
environment, is a technology that                      studies evaluating immersive and
allows individuals to experience and                   nonimmersive VR technology in the
interact with three-dimensional                        rehabilitation of the upper extremity.
environments. The most common                          The authors concluded that immersive
forms of virtual environments                          VR might be more effective compared
simulators are head-mounted displays                   no therapy, while the results from
(immersion) or with conventional                       studies examining nonimmersive VR
computer monitors or projector                         were conflicting.
screens (nonimmersion) (Sisto et al.

10. Upper Extremity Interventions                                                       pg. 79 of 150
Table 10.34 Studies Evaluating Virtual Reality Technology
 Author, Year                      Methods                                        Results
 PEDro score
Merians et al.   3 chronic stroke patients, 3 to 6 years        Each patient showed improvement in a
2002             post stroke, participated in a 2-week          subset of variables, with transfer of this
USA              training program (3 1/2 hours a day)           improvement to function on the Jebsen
No Score         including dexterity tasks performed in a       Test of Hand Function, on which they were
                 computerized VR environment, alternating       not trained. Two of the 3 patients showed
                 with hand activities performed in a real-      improvement in the use of their hand in
                 world environment. The VR simulations          several functional ADL tasks as well as on
                 were targeted for range of motion,             the Jebsen Test of Hand Function after
                 movement speed, fractionation, and force       this intervention, and 1 of the patients did
                 production.                                    not transfer this improvement to
                                                                functional activities.
Holden et al.    2 stroke patients were trained on a            After training, reaching errors during real-
1999             computer generated virtual environment         world performance were reduced by about
USA              (VE) to train upper extremity reach in the     50%. Both subjects improved in the
No Score         impaired limb. During VE subjects              trained task, and reaching also improved
                 practiced a virtual task similar to the real   in untrained tasks. Motor recovery (Fugl-
                 task, involving the patient mailing an         Meyer) and functional ability scores
                 envelope into a mail box slot that was         (SAILS) showed little to no change, but
                 simulated on the computer screen.              one subject acquired the ability to perform
                                                                several functional tasks that were not
                                                                measured on the formal test (SAILS).
Jack et al.      3 chronic stroke patients completed 9          Improvements were seen in finger motion,
2001             rehabilitation sessions in 2 weeks that        movement speed, and fractionation
USA              lasted about 1.5 hrs/day with the              (isolation of individual finger movements).
No Score         remaining time spent on traditional rehab      Patients also improved in functional scores
                 for up to 3.5 hours.                           and increased ability for force
Jang et al.      Controlled trial evaluating virtual reality    Following treatment VR patients scored
2005             (VR) training for 60 min/day x 5               significantly higher, compare to controls
Korea            days/week x 4 weeks vs. no VR                  on BBT, FMA and MFT scores. Cortical
5 (RCT)          intervention. 10 chronic stroke patients       activation by the affected movements
                 participated. Outcome measures included        were reorganized from contralateral
                 the box and block test (BBT), Fugl-Meyer       (before VR) to ipsilateral (after VR), in the
                 (FM) and the Manual function test (MFT).       laterality index.
                 Qualitative information on the amount of
                 use and the quality of movement was also
                 collected. Functional MRI was also
                 conducted. VR was designed to provide a
                 virtual rehabilitation scene where the
                 intensity of practice and sensory feedback
                 could be systematically manipulated to
                 provide the most appropriate,
                 individualized motor retraining program.
Piron et al.     50 mild to moderate stroke patients            Patients improved significantly from
2005             suffering from arm impairments received        baseline to post-treatment for mean
Italy            virtual-environment therapy for an hour        scores on the FIM and FM scales
No Score         every day for 1 month. Outcome                 (P<0.05). Also, there were significant
                 measures included Fugl-Meyer scale for         improvements for mean velocity of

10. Upper Extremity Interventions                                                             pg. 80 of 150
                 upper extremity (FM), Functional              movement and mean duration of
                 Independence Measure (FIM), kinematic         movement from baseline to post-
                 measures (velocity, duration and              treatment (P<0.05).
                 morphology of reaching movements.
                 Assessments were made before and after
You et al.       10 chronic stroke patients were assigned      There was a statistically significant in the
2005             to a virtual reality treatment designed to    median FAC and MMAS scores between
USA              simulate real life situations (n=5) or to a   the groups at the end of treatment.
4 (RCT)          no treatment control group (n=5). VR
                 treatment was provided for 60 min/day,
                 5x/wk x 4 wks. Motor function was
                 assessed by Functional Ambulation
                 Categories (FAC) and the walking item of
                 the modified motor assessment scale
Lam et al.       58 stroke patients were randomly              Over a 4-wk interval significant
2006             assigned to 1 of 3 groups: i.) 2DVR           improvements were seen for subjects in
Israel           computer based training programme for         both treatments groups in MTR
4 (RCT)          training in mass transit railway (MTR)        knowledge, skills and self-efficacy
                 skills, ii.) video modelling-based            (p<0.01). However, the control group
                 psychoeducational programme of similar        failed to improve and remained stable in
                 structure and content, or iii.) control.      skills and self-efficacy in using MTR.
                 Assessments were made using a
                 questionnaire on the demographic
                 characteristics of the patients, a
                 behavioural rating scale on using MTR
                 skills and an MTR self-efficacy rating
Merians et al.   8 chronic stroke patients underwent 3         Subjects as a group improved in
2006             weeks of training which required them to      fractionation of the fingers, thumb and
USA              use their hemiparetic hand to play a series   finger range of motion, and thumb and
No Score         of interactive computer games. The            finger speed, retaining those gains at the
                 training was 13 days in total and subjects    1-week retention test. Transfer of these
                 trained 2-2.5 hours daily. Outcome            improvements was demonstrated through
                 measure included fractionation, the           changes in the Jebsen Test of Hand
                 Jebson Test of Hand Function, variations      Function and a decrease after the therapy
                 in computerized measures of finger and        in the overall time from hand peak
                 thumb velocity, strength of finger and        velocity to the moment when an object
                 thumb, and a Kinematic reach to grasp         was lifted from the table.
Holden et al.    11 patients (>6 months following stroke)      Significant differences were seen from
2007             participated in a series of virtual-          pre- to post-test for the FM (P<0.0001),
USA              environment based telerehabilitation          WMT (P=0.0097), ShS (P=0.0010) and
No Score         treatment sessions that were conducted at     hand grip strength (P=0.0253). Almost all
                 their homes. Patients underwent 1 hr          improvements were maintained at 4-
                 sessions, 5x a week in 2 3-wk blocks.         months follow-up.
                 Outcome measures included the Fugl-
                 Meyer test of motor recovery (FM), Wolf
                 motor test (WMT), shoulder strength
                 (ShS) and hand grip strength.
Fischer et al.   15 chronic stroke patients with upper         Participants demonstrated a significant
2007             extremity hemiparesis were randomly           decrease in time to perform functional

10. Upper Extremity Interventions                                                            pg. 81 of 150
USA              assigned to 1 of 3 groups: i.) digit            tasks for the WMFT (p = .02), an increase
4 (RCT)          extension assisted by a novel cable             in the number of blocks successfully
                 orthosis (n=5), ii.) digit extension assisted   grasped and released during the BB (p =
                 by a novel pneumatic orthosis (n=5), iii.)      .09), and an increase for the FM score (p
                 or no assistance provided. The training         = .08). There were no statistically
                 consisted of 6 weeks of therapy (1 hour         significant changes in time to complete
                 sessions held 3 x/week). Each session           tasks on the RLA or any of the
                 required patients to grasp and release          biomechanical measures. Assistance of
                 virtual and actual objects. Outcome             extension did not have a significant effect.
                 measures included Wolf Motor Function
                 Test (WMFT), Rancho Los Amigos
                 Functional test of the Hemiparetic Upper
                 Extremity (RLA), Box and Blocks Test
                 (B&B), Upper Extremity Test Fugl-Meyer
                 Test (FM), and Biomechanical
                 Assessments. Assessments were
                 conducted before/after and at one and 6
                 months post intervention.
Broeren et al.   22 subjects with chronic stroke living in       All the participants in the VR group
2008             the community in their own homes were           reported that they enjoyed using the
Sweden           randomized to receive extra rehabilitation      system. There were no significant
3 (RCT)          by training on a computer 3 times a week        differences between the control and the
                 during a 4-week period or to continue           VR group on tests of manual abaility or
                 their previous rehabilitation (no extra         executive funcrion. There were significant
                 computer training). The VR training             improvements in some of the kinematic
                 consisted of challenging games which            measuresments associated with reaching.
                 provided a range of difficulty levels. An
                 additional group of 11 right-handed, aged
                 matched individuals without history of
                 neurological or psychiatric illnesses served
                 as reference subjects. Outcomes were
                 assessed before and after treatment and
                 included a semi-structured interview, Box
                 & Block test (BBT), ABILIHAND, Trail
                 Making Test (to assess executive function
                 and attention) and kinematic analyses.
Yavuzer et al.   20 hemiparetic inpatients, all within 12        There was a significantly greater
2008             months post stroke, received 30 minutes         improvement in both mean hand and UE
Turkey           of treatment with ''Playstation EyeToy          components of the Brunnstrom scale and
6 (RCT)          Games'' per day, consisting of flexion and      FIM scores following treatment among
                 extension of the paretic shoulder, elbow        subjects in the EyeToy group. The
                 and wrist as well as abduction of the           improvement in FIM self-care scores
                 paretic shoulder or placebo therapy             remained significantly greater at follow-up
                 (watching the games for the same                among subjects in the EyeToy group (5.5
                 duration without physical involvement into      vs .1.8, p=0.018).
                 the games) in addition to conventional
                 program, 5 days a week, 2-5 hours/day
                 for 4 weeks. Brunnstrom's staging and
                 self-care sub-items of the functional
                 independence measure (FIM) were
                 performed at baseline, 4 weeks (post-
                 treatment), and 3 months follow-up.
Housman et       34 chronic stroke subjects were                 At the end of 6 months, subjects in the T-
al. 2009         randomized to receive 24 treatment              WREX group achieved and maintained a

10. Upper Extremity Interventions                                                             pg. 82 of 150
USA            sessions using either an arm orthosis         higher FMA score compared with control
6 (RCT)        using the Therapy Wilmington Robotic          (3.6 vs. 1.5 points, p<0.045). There were
               Exoskeleton (T-WREX), which supports          no other significant between group
               the arm against gravity and measures          differences.
               arm movement and trace hand grasp as
               users interact with comuter games (n=14)
               or to a control group (n=14) whereby
               subjects participated in conventional
               exercises (self range of motion stretches
               and active range of motion strengthening
               exercises). Outcomes assessed before and
               after treatment and at 6 month follow-up
               included the Fugl-meyer Assessment, ADL
               assessment, Motor Activity Log and grip
Piron et al.   36 patients with mild arm motor               Both rehabilitative therapies significantly
2009           impairments due to ischaemic stroke in        improved all outcome scores after
Italy          the region of the middle cerebral artery      treatment. Subjects in the Telerehab
7 (RCT)        were randomized to one of 2, 4-week           group had a higher FM score at the end of
               outpatient treatment programs.                treatment compared with control (53.6 vs.
               Treatment lasted for 1 hr/day x 5 days/       49.5, p<0.05), although the difference
               week. The experimental treatment was a        was no longer significant at 3 months
               virtual reality-based system delivered via    (53.1 vs. 48.8). A similar pattern was
               the Internet, which provided motor tasks      observed for the ABIHAND, whereby the
               to the patients from a remote                 Telerehab group had significantly better
               rehabilitation facility (Telerehab). The      scores after the first 2 assessments but
               control group underwent traditional           not at follow-up. There were no
               physical therapy for the upper limb.          differences in Ashworth Scores between
               Outcomes were assessed one month prior        groups at any point.
               to therapy, at the beginning and end of
               therapies and one month post-therapy,
               with the Fugl-Meyer (FM) Upper
               Extremity, the ABILHAND and the
               Ashworth scales.
Saposnik et    22 patients within 2 months of stroke         The interventions were successfully
al. 2010       receiving standard rehabilitation were        delivered in 9 of 10 participants in the
Canada         randomized to receive either 8, 60            VRWii and 8 of 10 in the recreational
7 (RCT)        minutes sessions with either the Nintendo     therapy arm. Participants in the VRWii
               Wii gaming system (VRWii) or                  arm had a significant improvement in
               recreational therapy (playing cards, bingo,   mean motor function of 7 seconds (Wolf
               or "Jenga"). The primary feasibility          Motor Function Test, 7.4 seconds. There
               outcome was the total time receiving the      were no differences on any of the other
               intervention. Efficacy was evaluated with     outcomes.
               the Wolf Motor Function Test, Box and
               Block Test, and Stroke Impact Scale at 4
               weeks after intervention.

Discussion                                            RCTs have been conducted. One of the
                                                      studies included in this review used
Virtual reality training is an innovative             virtual reality technology as a more
new treatment approach, which may                     efficient method to test the efficacy of
enhance cortical reorganization                       a device used to improve finger
following stroke. To date only a few                  extension (Fisher et al. 2007). Two

10. Upper Extremity Interventions                                                         pg. 83 of 150
trials used popular gaming systems--                    attributed to the avoidance of learned
the Playstation EyeToy and the                          nonuse behaviour or by repeated
Ninetendo Wii gaming system, both                       practice of functional tasks. The
inexpensive video game device to                        results from the remaining RCTs
improve upper-extremiy function                         indicated that virtual reality treatment
following stroke, with equivocal                        was of benefit for chronic stroke
results. The authors hypothesized that                  patients in the improvement of motor
the beneficial effect could be                          function.

Table 10.35 Summary of RCTs Evaluating Virtual Reality Technology
    Author/      N                  Intervention                           Main Outcome(s)
 PEDro Score                                                                    Result
Saposnik et     20        Nintendo Wii gaming system                Wolf Motor Function Test (+)
al. 2010                (VRWii) vs. recreational therapy                 Box & Block test (-)
7 (RCT)                                                                Stroke Impact Scale (-)
Yavuzer et      20       Playstation EyeToy games vs.                   Brunnstrom score (-)
al. 2008                      conventional therapy                        FIM self-care (+)
6 (RCT)
Jang et al.     10        Virtual reality training for 60              Box and block test (+)
2005                    min/day x 5x/wk x 4 wks vs. no                   Fugl-Meyer test (+)
Korea                        Virtual reality training.                Manual function test (+)
5 (RCT)
You et al.      10       Virtual reality treatment for 60       Functional Ambulation Categories (+)
2005                     min/day, 5x/wk x 4 wks vs. no            Modified Motor Assessment Scale
USA                                  treatment                            (+ walking item)
4 (RCT)
Lam et al.      58       2DVR computer based training                   Mass Transit Railway
2006                   programme vs. video modelling-                          (- skills)
Israel                      based psychoeducational                        (- self-efficacy)
4 (RCT)                      programme vs. control
Broeren et      22 Semi-immersive workbench with                         Box & Block Test (-)
al. 2008                haptic and stereoscopic glasses                    ABILIHAND (-)
Sweden                         vs. no VR treatment                      Trail Making Test (-)
3 (RCT)                                                                    Kinematics (+)
- Indicates non-statistically significant differences between treatment groups
+ Indicates statistically significant differences between treatment groups

 Conclusions Regarding Virtual Reality
 Technology in the Treatment of Stroke                  10.5 Treatment for Spasticity or
 There is strong (Level 1a) evidence that               Contracture in the Upper
 virtual reality treatment can improve                  Extremity
 motor function in the chronic stages of
                                                        Stroke survivors often display a
                                                        constellation of signs and symptoms
  Virtual reality therapy may improve                   that together constitute the upper
  motor outcomes post stroke.                           motor neuron syndrome. The
                                                        syndrome consists of negative signs
                                                        including: weakness, loss of dexterity,

10. Upper Extremity Interventions                                                         pg. 84 of 150
fatigue and positive signs including         There are a number of interventions
increased muscle stretch reflexes,           used for limb spasticity. These include
abnormal cutaneous reflexes and              oral antispasticity agents, injection of
spasticity. Spasticity is classically        phenol to motor nerves or alcohol to
defined as a velocity dependent              muscle bellies, and physical modalities
increase of tonic stretch reflexes           such as stretching, orthoses, casting,
(muscle tone) with exaggerated               cold application and surgery. The
tendon jerks. Spasticity can be              mainstay of treatment for spasticity
painful, interfere with functional           has been physical therapy. Traditional
recovery in the upper extremity and          pharmacotherapies for spasticity
hinder rehabilitation efforts. However,      include centrally acting depressants
Gallichio (2004) cautioned that a            (baclofen, benzodiazepines, clonidine,
reduction in spasticity do not               and tizanidine) and muscle relaxants
necessarily lead to improvements in          (dantrolene). There is evidence from
function. Van Kuijk et al. (2002) noted      RCTs published in the 60’s and 70’s
that for most stroke patients,               that these treatments are only
“…spasticity is a variable phenomenon        partially effective in treating spasticity
in time and apparent in only certain         and have negative side effects of
muscle groups, and therefore, low            weakness and sedation. More recently,
threshold and “reversible” focal             Tizanidine hydrochloride was used to
treatment techniques seem to be the          successfully treat spasticity among 47
preferable first option.”                    chronic stroke patients, although, due
                                             to side effects, only a small
A recent study by Watkins et al.             percentage of patients reached the
(2002) reported that 39% of patients         maximum daily dose (Gelber et al.
with a first-ever stroke were spastic        2001). Motor point or nerve blocks
12 months after their stroke. More           with phenol or alcohol have been used
recently, a study by Sommerfeld et al.       but are often associated with variable
(2004) reported that of 95 patients          success rates, and high rates of
assessed initially (mean 5.4 days)           neuropathic pain. Botulinum toxin
after an acute stroke, 77 (81%) were         type A, a potent neurotoxin that
hemiplegic and 20 (21%) were                 prevents the release of acetylcholine
spastic. Overall, upper extremity            from the preseynaptic axon, has more
spasticity alone (n=13) was more             recently been studied as a potentially
common than lower extremity                  useful treatment for stroke related
spasticity alone (n=1) or spasticity in      spasticity. Intrathecal drug therapy
both upper and lower extremities             refers to the injection of a drug into
(n=6). At three months post-stroke,          the subarachnoid space of the central
64 patients (67%) were still                 nervous system and requires the
hemiparetic, and 18 (19%) were still         implantation of a programmable
spastic. At that point, there were           device into the subcutaneous tissue
more patients with spasticity in both        surrounding the abdominal wall.
extremities (n=10) than in the upper         Intrathecal baclofen, the most
extremity alone (n=7) or in the lower        commonly used intrathecal drug for
extremity alone (n=1). The authors           relieving spasticity associated with
also reported that severe disabilities       stroke has not been well studied,
were found in almost the same                particularly for spasticity of the upper
number of nonspastic patients as             extremity.
spastic patients.

10. Upper Extremity Interventions                                         pg. 85 of 150
10.5.1 Splinting                                          splinting post stroke look at outcomes
                                                          such as tone, four studies looked
While the majority of studies                             specifically at spasticity post
examining the effectiveness of hand                       intervention.

Table 10.36 Splinting the Upper Extremity
    Author/                       Methods                                          Outcomes
 Pedro Score
Langlois et al. 9 patients within 12 months of acute           No significant differences were found in
1991            stroke were randomly allocated to one of       spasticity reduction between groups, however,
3 (RCT)         3 groups: wearing a finger spreader splint     all 3 groups demonstrated a reduction in
                for 6 hrs, 12 hrs or 22 hrs per day for a      spasticity. No significant differences were found
                period of 2 weeks.                             between groups on measures of expectation or
                                                               satisfaction and the reduction of spasticity
                                                               (Expectation and Satisfaction Questionnaire that
                                                               was developed specifically for this study), or on
                                                               reported compliance and prescribed wearing
                                                               schedule at 2 weeks.
Pizzi et al.    40 consecutive patients with hemiplegia        There was a significant improvement of wrist
2005            and upper-limb spasticity after stroke that PROM with greater changes in extension than in
Italy           had occurred at least 4 months before          flexion, and a reduction of elbow spasticity, wrist
No Score        wore an immobilizing hand splint custom- pain and spasms. There was significant
                fitted in the functional position for at least improvement in the mean MAS scores (elbow).
                90 minutes daily for 3 months. Outcome
                measures included Modified Ashworth
                Scale (elbow and wrist), passive range of
                motion (PROM) at the wrist and elbow;
                pain at the shoulder, elbow, and wrist
                using a visual analog scale and spasms.
Sheehan et al. 12 stroke patients were randomly                Effect sizes were small and failed to reach the
2006            assigned to either Group 1 or Group 2.         suggested smallest clinically worthwhile effect
Australia       Following one week of baseline when            size for amount and rate of change in resistance
5 (RCT)         neither group wore a splint, Group 1           in the short term. However, the average
                continued without a splint for week 2 and estimated size of the effect for rate of change
                then wore a splint in week 3. Group 2          with longer term splinting exceeded the smallest
                wore a splint during week 2 and 3. Both        clinically worthwhile effect, suggesting that
                groups wore a splint in weeks 4 to 7.          further research is warranted.
                Subjects in either group did not receive
                additional upper extremity therapy.
                Amount and Rate of change in resistance
                at the wrist was measured with a
                computerized torque apparatus.
Lai et al. 2009 36 chronic stroke patients with hypertonia 30 patients completed the study. Although no
USA             in elbow flexion were included. All            statistical tests were performed, the percentage
5 (RCT)         subjects received the current standard of of change in active range of motion in elbow
                care: botulinum toxin type-A injections        extension was greater for the experimental than
                and manual therapy. Experimental               for control subjects (33.5% vs. 18.7%). The
                patients were randomly assigned to also        Modified Ashworth Scale (extension) scores
                receive adjunct treatment with an Elbow        showed comparable changes of a mean 9.3%
                Extension Dynasplint to be worn for 6-8        improvement for experimental versus 8.6% for
                hours during sleep. Assessments were           the control subjects.
                conducted before and after treatment (14

10. Upper Extremity Interventions                                                            pg. 86 of 150
               weeks) and included active range of
               motion in elbow extension and the
               Modified Ashworth Scale (extension) for

Three RCTs were identified examining                      It remains uncertain if hand splinting
the benefit of splinting. The focus of                    decreases spasticity.
each of these studies was different
(finger, wrist and elbow). One of the                   10.5.2 Stretching Programs to
trials (Lai et al. 2009) assessed a                     Prevent Contracture
dynamic splint, which progressively
increases torque to reduce contracture                  Spastic contracture following stroke is
and maintain the joint at its end                       the expression of hypertonicity or
range. The remaining trials assessed                    increased active tension of the
resting or static splints. All trials were              muscle. Contracture may also occur as
of fair or poor quality. Of the 4 trials                a result of atrophic changes in the
reviewed, 2 reported a benefit of                       mechanical properties of muscles.
treatment, while 2 did not.                             Since surgery is the only treatment
                                                        option once a contracture has
 Conclusions Regarding Treatment of                     developed, prevention is encouraged.
 Spasticity: Hand Splinting
                                                        Stretching may help to prevent
                                                        contracture formation and, although
 There is conflicting (Level 4) evidence
 that hand splinting decreases                          well-accepted as a treatment strategy,
 spasticity.                                            has not been well-studied.

Table 10.37 Stretching Programs to Prevent Contracture
    Author/                     Methods                                           Outcomes
 Pedro Score
Turton &       In addition to usual care, 13 subjects on a    There were no significant effects of treatment.
Britton 2005   stroke rehabilitation unit, admitted within    By eight weeks post stroke the mean range of
UK             4 weeks post stroke and with no hand           wrist extension and shoulder external rotation
6 (RCT)        function, received two 30-min stretches        lost on the affected side in both groups was
               for wrist and finger flexors and two 30-       approximately 30 degrees. Compliance was
               min stretches targeting shoulder               variable. Only 6 patients completed the full
               adductors and internal rotators, perday        treatment. Patients declined to participate or
               for up to 12 weeks post stroke. Therapists     were unavailable and staff was also non-
               and nursing staff carried out stretches. 12    compliant.
               patients in the control group received
               standard care.
Tseng et al.   59 bedridden older stroke survivors in         Both intervention groups had statistically
2007           residential care were randomly assigned        significant improvement in mean joint angles,
Taiwan         to usual care or one of two intervention       activity function, perception of pain and
7 (RCT)        groups. The 4-week, twice-per-day, 6           depressive symptoms compared with the usual
               days-per-week range-of-motion exercise         care group (P < 0.05). Post hoc comparison
               protocols were similar in both intervention    revealed that the joint angles in intervention
               groups, and consisted of full range-of-        group II were statistically significantly wider than
               motion exercises of the upper and lower        in both the other groups (P < 0.01).
               extremities. To test the effect of different

10. Upper Extremity Interventions                                                           pg. 87 of 150
              degrees of staff involvement, in
              intervention group I, a Registered Nurse
              was present to supervise participants
              performing the exercises, while
              intervention group II involved a
              Registered Nurse physically assisting
              participants to achieve maximum range-
              of-motion within or beyond their present
              abilities. Assessments conducted before
              and after the intervention included joint
              angles of the shoulder, elbow and wrist
              and FIM and pain score.
                                                     upper extremity. Unlike chemical
Only two RCTs have been published                    neurolysis with either phenol or
examining the benefit of stretching                  alcohol, botulinum toxin is not
regimes to help prevent contracture                  associated with skin sensory loss or
formation. One was conducted during                  dysesthesia (Suputtitada & Sunanwela
the acute stage of stroke (Turton &                  2005). Dynamic EMG studies can be
Britton 2005) and the other, during                  helpful in determining which muscles
the chronic stage (Tseng et al. 2007).               should be injected (Bell and Williams
The development of contracture                       2003).
formation was not reported, therefore
the outcomes assessed focused on                     In a review by van Kuijk et al. (2002)
joint angles, ADL performance and                    on the benefit of botulinum toxin in
pain.                                                upper extremity spasticity in stroke
                                                     patients by focal neuronal or
 Conclusions Regarding Stretching                    neuromuscular blockade, 10 studies (4
 Programs to Prevent Contracture                     RCTs and 6 uncontrolled observational
 Formation                                           studies). The authors found that
                                                     there is evidence of the effectiveness
 There is moderate (Level 1a) evidence               of botulinum toxin treatment on
 that a nurse-led stretching program can             reducing muscle tone (modified
 help to increase range of motion in the             Ashworth Scale) and improving
 upper extremity and reduce pain in the
                                                     passive range of motion at all arm-
 chronic stage of stroke.
                                                     hand levels in chronic patients for
10.5.2 Botulinum Toxin Injections                    approximately 3 to 4 months.
                                                     However, the authors concluded that
                                                     while overall, the effectiveness on
Botulinum works by weakening spastic
                                                     improving functional abilities was not
muscles through selectively blocking
                                                     justified, specific stroke groups may
the release of acetylcholine at the
                                                     benefit from botulinum injections in
neuromuscular junction. The benefits
                                                     the upper extremity.
of botulinum injections are generally
dose-dependent and last
                                                     While many controlled studies have
approximately 2 to 4 months
                                                     demonstrated a reduction in spasticity
(Simpson et al. 1996, Bakheit et al.
                                                     following treatment with botulinum
2001, Smith et al. 2000, Francisco et
                                                     toxin, usually BTX-A (Botox or
al. 2002, Brashear et al. 2002). One
                                                     Dysport), measured using the Modified
of the advantages of botulinum is that
                                                     Ashworth Scale or range of motion, it
it is safe to use on small, localized
                                                     is less clear whether treatment is
areas or muscles, such as those in the

10. Upper Extremity Interventions                                              pg. 88 of 150
associated with an improvement in                         for upper limb spasticity following
upper extremity function. Fancis et al.                   stroke. They included five RCTs
(2004) suggested several reasons why                      (Simpson et al. 1996; Smith et al.
this might be so. These authors                           2000; Bakheit et al. 2000; Bakheit et
suggested that underlying muscle                          al. 2001; Brashear et al. 2002) and
weakness and not spasticity contribute                    reported that there was a significantly
to the limitation in function. However,                   greater reduction in spasticity for
they speculated that the most likely                      patients who underwent BTX-A
reasons were insufficiently sensitive                     treatment compared to patients
outcome measures chosen and under-                        receiving the placebo treatment, as
powered studies. A meta-analysis                          measured by the modified Ashworth
authored by the same authors                              Scale and the Global Assessment
included the results from two RCTs                        Scale. They concluded that BTX-A
(Bakheit et al. 2000, 2001), suggested                    reduces spasticity and that the
that there was a benefit, albeit modest                   treatment was tolerated well, although
of BTX-A on improved function. The                        the effects of long-term use of BTX-A
authors of this review pooled the data                    are unknown. Levy et al. (2007)
and assessed the effect on the arm                        reported additional benefits when a
section of the Barthel Index (dressing,                   course of constraint-induced
grooming and eating) and reported a                       movement therapy followed treatment
modest improvement in upper arm                           with BTX-A. Unfortunately the gains in
function following botulinum toxin.                       motor function were lost at the end of
Pooling was only possible for two RCTs                    24 weeks when spasticity returned.
due to heterogeneity of interventions
and outcomes.                                             The results from controlled and
                                                          uncontrolled studies, which evaluated
Cardoso et al. (2005) conducted a                         the effect of botulinum toxin on
meta-analysis investigating botulinum                     spasticity, were reviewed. The results
toxin type A (BTX-A) as a treatment                       are presented in Table 10.38.

Table 10.38 Botulinum Toxin Injection and Spasticity in Upper Extremity Post Stroke
   Author/                      Methods                                       Outcomes
 PEDro Score
Bhakta et al.   17 patients received a single course of     Significant improvement in grading of biceps
1996            intramuscular botulinum toxin to biceps     spasticity and forearm finger flexor spasticity
UK              brachii, flexor digitorum profundus,        were observed. Significant mean
No Score        flexor digitorum superficialis and flexor   improvement of 170 for passive shoulder
                carpi ulnaris.                              abduction or adduction, 170 for passive
                                                            shoulder flexion and extension and 160 for
                                                            passive elbow flexion and extension were
                                                            noted. Significant mean improvement of 310
                                                            for passive wrist dorsiflexion and palmar
                                                            flexion were observed. Hand hygiene
                                                            improved in 14 of 17 patients. Shoulder pain
                                                            improved in 6 or 9 patients and wrist pain
                                                            improved in 5 of 6 patients.
Simpson et al. A double blind, placebo controlled trial     Treatment with 300-unit BTX-A dose resulted
1996           of 37 patients randomized to receive         in clinical significant mean decrease in wrist
USA            either a single treatment of either 75       flexor tone at 2, 4 and 6 weeks post-injection.
8 (RCT)        units, 150 units or 300 units of total       BTX-A groups reported significant

10. Upper Extremity Interventions                                                          pg. 89 of 150
                 doses of BTX-A or placebo into the          improvement on physician and patient Global
                 biceps, flexor carpi radialis and flexor    Assessment of Response to Treatment at
                 carpi ulnaris muscles.                      weeks 4 and 6 post-injection.
Smith et al.     Double blind placebo trial of 25 patients   Combining data from active treatment,
2000             randomized to receive 500 Mu, 1000          botulinum toxin showed significantly greater
UK               Mu or 1500 Mu of botulinum toxin by         improvement in modified Ashworth scale at
7 (RCT)          intra-muscular injection, or placebo of     fingers, passive range of movement at the
                 an equal volume of sterile saline.          wrist, and finger curl distance at rest. Only
                                                             significant difference between dose groups in
                                                             favour of 1500 Mu for improved movement at
                                                             the elbow.
Bakheit et al.   International, multi-center, randomized,    All 3 groups receiving Dysport showed
2000             double-blind placebo-controlled trial of    significant reduction in MAS (Modified
UK               82 patients randomized to one of four       Ashworth Scale) scores in any joint at week 4
8 (RCT)          groups (500 U of Dysport, 1000 U of         compared with placebo. At 16 weeks, the MAS
                 Dysport, 1500 U of Dysport or placebo).     scores were significantly reduced in the
                 Injections were made to the biceps          hemiparetic arm for all doses in the elbow and
                 brachii, flexor digitorum profundus and     wrist and also in the fingers in the 1000 U
                 flexor digitorum superficialis, flexor      Dysport group. No significant differences were
                 carpi ulnaris and flexor carpi radialis     found between groups on the Rivermead
                 muscles.                                    Motor Assessment, pain scores, or Barthel
                                                             Index scores.
Bhakta et al.    40 patients with stroke and spasticity in   There was a significant reduction in disability
2000             a functionally useless arm (median          at 2 and 6 weeks, but not at 12 weeks post
UK               duration 3.1 years) were randomised to      treatment. Caregiver burden was significantly
7 RCT            receive intramuscular botulinum toxin       reduced throughout the study period. MAS
                 type A (BT-A; Dysport) (n=20) or            scores were reduced significantly at 2 weeks
                 placebo (n=20) in a total dose of 1000      (finger and elbow), 6 weeks and 12 weeks
                 MU divided between elbow, wrist, and        (finger only). There was no significant
                 finger flexors. Spasticity was assessed     improvement in grip strength or reduction in
                 using the modified Ashworth scale           pain associated with BT-A treatment.
                 (MAS). Muscle power, joint movement,
                 and pain, disability and carer burden
                 were also assessed. Two baseline and
                 three post-treatment assessments were
                 conducted at weeks 2, 6, and 12 weeks.
Lagalla et al.   34 stroke patients suffering from        Of the 28 patients who completed the trial, all
2000             moderate to severe chronic spasticity    received at least 6 injections. There was a
Italy            received BTX-A injections every 3 to 5   significant decline in Ashworth scores after
No Score         months for 28 months and were            the first injection. However, there were no
                 observed for 3 years. Mean dose per      subsequent changes in scores. Passive and
                 session was 128 U. Patients also         active ROM also improved significantly after
                 received physical therapy twice weekly.  the first treatment. Frenchay arm test scores
                                                          improved in only 8 subjects.
Bakheit et al.   International, multi-center, randomized, The group who received Dysport had a
2001             double-blind placebo-controlled trial of significant reduction in the summed Modified
UK               59 patients who received either placebo Ashworth Scale score at week 4 compared
8 (RCT)          injections or a total of 1000 IU of BtxA with the placebo group. The magnitude of
                 (Dysport) into 5 muscles of the affected benefit over the 16 week follow-up period was
                 arm.                                     significantly reduced for the BTX-A group in
                                                          the wrist and finger joints compared with the
                                                          placebo group. No significant difference was
                                                          noted between the groups in the joint ROM,

10. Upper Extremity Interventions                                                           pg. 90 of 150
                                                         muscle pain, goal-attainment or the Barthel
                                                         Index scores at week 4 of the study. At week
                                                         16, the BTX-A group showed significantly
                                                         greater improvement in elbow PROM.
Brashear et al. 126 stroke patients were randomized to 122 patients completed the study. The
2002            receive a single injection of BTX-A      primary outcome was improvement in the 4-
USA             (n=64) or placebo (n=64) (50 units       point Disability Assessment scores at 6 weeks
7 (RCT)         injected in each of 4 wrist and finger   (hygiene, dressing, pain and limb position).
                muscles).                                Six weeks after injection with BTX-A 83% of
                                                         subjects reported at least a one-point
                                                         improvement of DAS score compared to 53%
                                                         of patients who were treated with placebo
Francisco et    13 patients (10 strokes) with Modified   Assessments were completed at 4, 8 and 12
al. 2002        Ashworth Scores (MAS) of 3 or 4 were weeks post injection. MAS scores of both
USA             randomized to receive either high        wrist and finger flexors were assessed. While
7 (RCT)         volume BTX-A (50 units/1 mL saline:1.2 MAS scores decreased significantly in both
                mL delivered per 4 muscles) or low       treatment groups, there were no differences
                volume BTX-A (100 units/1 mL saline      between the low and high volume BTX-A
                delivered per 4 muscles). On average,    regimens.
                patients in the high volume group
                received 417 units BTX-A compared to
                patients in the low volume group (432
Pandyan et al. 14 stroke patients with elbow flexor      There was a significant improvement in upper
2002            spasticity received 3 injections of      limb function (P<0.05) as measured by the
UK              botulinum toxin Type A (BTX-A) into the ARAT.
No Score        m.biceps branchii, m.branchioradialis
                and m.flexor digitorum longus (mean
                dose 70U, 56.5 U and 83.3 U
                respectively). Assessments were made
                before and 4 weeks after treatment.
                They included the modified Ashworth
                Scale (MAS), grip strength and upper
                limb function (Action Research Arm
                TEST (ARAT)), and strength at the
                elbow (isometric).
Bakheit et al. An open label study in which 51           41 subjects completed all 3 treatment cycles.
2004            patients with established post stroke    Improvement from the cycle one baseline was
UK              upper limb spasticity received 1000      observed in all the outcome measures. 100%
No Score        units of BtxA (Dysport) into five        of subjects achieved at least a 1- point
                muscles of the affected arm. Treatment decrease on MAS scores in at least 1 joint. By
                was repeated every 12, 16, or 20         the end of the 3rd cycle, 98% had achieved a
                weeks as clinically indicated. Each      1-point reduction. 90% of subjects who
                patient received a total of three        completed the 3 cycles reported that the
                treatment cycles. Efficacy of treatment treatment had been beneficial. Mild to
                was assessed using the Modified          moderately severe treatment related adverse
                Ashworth Scale. Patients were assessed events were reported in 24% of cases. No
                on study entry and on week 4 and 12 of BtxA antibodies were detected
                each treatment cycle for all safety and
                efficacy parameters. Blood samples for
                BtxA antibody assay were taken at
                baseline and on completion of the trial.

10. Upper Extremity Interventions                                                      pg. 91 of 150
Brashear et al. 15 stroke patients were randomized to       There was no significant decrease in muscle
2004            receive a single Botox type B injection     tone in the elbow, wrist, or finger. A
USA             (10,000U) in the elbow, wrist, finger       decrease in Ashworth scale scores was
7 (RCT)         and thumb (n=10) or placebo (n=5).          observed at the wrist at week 2 in the
                Measures were recorded at 2, 4, 8, 12       treatment group. Improvement was also
                and 16 weeks.                               observed at week 4 for the elbow (p=.039),
                                                            wrist (p=.002), finger (p=.001) and thumb
                                                            (p=.002) in the treatment gr. Improvements
                                                            were not sustained.
Gordon et al.     Additional component of study by          Compared to baseline values from the double-
2004              Brashear et al. 2002.                     blind portion of the study, there were
USA                                                         significant improvements in each of the four
No Score          111 patients who completed the study      domains of the Disability Assessment Scale.
                  entered into an open label study of       There were also improvements in Modified
                  BTX-A and received up to four             Ashworth Scores.
                  treatments. The mean dose was 220U.
                  The longest interval between cycles was
                  24 weeks.

Childers et al.   91 patients were randomized to 4          A dose-dependent response in muscle tone
2004              groups: (1) 90U Botox type A; (2) 180U    was generally observed in tone reduction in
USA               Botox; (3) 360U Botox; (4) placebo.       the wrist (p<.03), elbow (p<.04, and finger
7 (RCT)           Efficacy outcome measures were            (p<.04), but not in pain, FIM scores, or SF-36
                  completed for the 4 groups as follows:    scores.
                  (1) n=16; (2) n=15; (3) n=18; (4)
Suputtitada &     Patients received either a placebo       All doses of Dysport studied showed a
Suwanwela         (n=15) or one of three does of Dysport   significant reduction from baseline of muscle
2005              (350 U n=15, 500 U n=15, 1000 U          tone and pain compared to placebo. However,
Thailand          n=15) into five muscles of affected arm  the effect of functional disability was best at a
6 (RCT)           by anatomical and electromyography       dose of 500 U and the peak improvement was
                  guidance. Efficacy was assessed          at week 8 after injection. A dose of 1000 U
                  throughout the 6-month study period      Dysport produced such an excess degree of
                  by the Modified Ashworth Scale (MAS),    muscle weakening that the number of
                  the Action Research Arm Test (ARA),      randomized patients was reduced to five. BI
                  the Barthel Index (BI) and the Visual    and ARA of all patients were decrease after
                  Analogue Pain Scale (VAS).               injection. No other adverse event was
                                                           considered related to the study medication.
Slawek et al. Open-label study of 21 stroke patients There were statistically significant
2005            with onset of symptoms from 3 months improvements in baselines scores to week 16
Poland          to less than 3 years. Patients received    for MAS (elbow and wrist), Bhahkta finger
No Score        an average dose of botulinum toxin-A of scale in passive movements and muscle tone
                255 U, based on individual spasticity.     analyses. The only significant result for active
                Outcome assessed included Modified         movement analysis was MAS (arm). Pain was
                Ashworth scores, finger flexion scale,     present only in 11 patients and did not
                nine-hole peg test, Motor Assessment       significantly improve following treatment.
                Scale, assessed up to week 16.             Individualised BTX-A injection regimens may
                                                           be an effective, reversible and safe new
                                                           treatment option for patients with spasticity.
                                                           Nevertheless, functional improvement may be
                                                           reached only in selected patients.
Jahangir et al. 27 patients, at least 3 months following At the end of 3 months there were significant
2007            stroke, with focal spasticity of the wrist improvements favouring the Botox group in

10. Upper Extremity Interventions                                                           pg. 92 of 150
Malaysia       and fingers were randomized to receive     terms of MAS score of both the wrist and
6 (RCT)        a single injection of 40 U of botulinum    finger, but no significant differences on any of
               toxin (Botox) or placebo. 20 U were        the other outcomes assessed. No serious
               injected into the wrist and finger         Botox related adverse effects were reported.
               flexors. All subjects received physical
               therapy for 1 hour, twice a week for 3
               months. Assessments were performed
               at baseline and 1 and 3 months after
               injection and included the Modified
               Ashworth Scale (MAS), Barthel Index
               (BI) and EQ-5D and EQ VAS for quality
               of life.
Bhakta et al. Additional results from 2000 study          Peak associated reactions force was reduced
2008           evaluating the impact of associated        at week 6 with botulinum toxin A compared
UK             reactions on activities of daily living.   with placebo (mean group difference 19.0 N;
9 (RCT)        Associated reactions were measured         95% confidence interval (CI): 7.2, 30.9; p <
               using hand dynamometry. The effort         0.01) and week 2 (p = 0.005), with the effect
               used was measured using maximum            wearing off by week 12 (p = 0.09). 31
               voluntary grip in the unaffected arm.      patients noted associated reactions on a
               Measurements were recorded at 2 pre-       regular basis and 24 said that these
               treatment and 3 post-intervention          movements interfered with daily activities.
               times. Activities that patients felt       Ten of 12 patients receiving botulinum toxin A
               caused associated reactions and            and 2 of 12 receiving placebo reported
               activities that were affected by           reduction in interference with daily activities
               associated reactions were recorded.        (p = 0.02)
Caty et al.    20 chronic stroke patients were            There was significant improvement in
2009           injected with 400-1,000 U of               SIAS scores, MAS scores and MRC
Belgium        Dysport into several upper limb            grading (elbow flexor and thumb flexor).
No Score       spastic muscles. An ICF-based              There was no significant improvement on
               assessment set was used to                 measures of activity or participation.
               measure improvement before and 2
               months after treatment: Body
               function and structure- Stroke
               impairment assessment set (SIAS),
               Modified Ashworth Scale (MAS),
               Medical Research Council muscle
               testing, grip and pinch strength,
               Box and Block test, Purdue
               Pegboard test, Activity-
               (ABILHAND), Participation (SATIS-
               Stroke), and quality of life (SF-36).
Simpson et al. 60 subjects with upper-limb spasticity     At 6 weeks the mean changes from baseline
2009           due to stroke or traumatic brain injury    in wrist flexor tone scores were: BoNT: -1.32,
USA            (TBI) were randomised to 1 of 3            TZD: -0.22, placebo: -0.68. BoNT produced
8 (RCT)        groups: (1) intramuscular BoNT plus        significantly greater reductions compared with
               oral placebo; (2) oral tizanidine (TZD)    either placebo or TZD. There were more
               plus intramuscular placebo; (3)            adverse events associated with TDZ compared
               intramuscular placebo plus oral placebo.   with BoNT
               Wrist flexors were systematically
               injected, while other upper limb
               muscles were injected as per
               investigator judgement (total maximum
               dose: 500 U). The study duration was
               22-24 weeks. The primary outcome was

10. Upper Extremity Interventions                                                         pg. 93 of 150
               the difference in change in wrist flexor
               modified Ashworth score (MAS) at 6
               weeks. Other outcome measures
               included adverse events (AE).
McCrory et al. 96 patients an average of 5.9 years        There were no significant between group
2009           post stroke were randomized to receive     differences in AQoL change scores, pain,
Australia      either 500-1,000U botulinum toxin type     mood, disability or carer burden. However,
9 (RCT)        A or placebo into the affected distal      patients treated with botulinum toxin type A
               upper limb muscles on 2 occasions, 12      had significantly greater reduction in
               weeks apart. Assessment was                spasticity (MAS) (p < 0.001), higher GAS
               undertaken at baseline, 8, 12, 20 and      scores (p < 0.01) and greater global benefit
               24 weeks. The primary outcome              (p < 0.01).
               measure was the Assessment of Quality
               of Life scale (AQoL) assessed at week
               20. Secondary outcome assessments
               included Goal Attainment Scaling
               (GAS), pain, mood, global benefit,
               Modified Ashworth Scale (MAS),
               disability and carer burden.
Meythaler et   21 subjects with stroke onset more         The use of BTX-A combined with therapy as
al. 2009       than 6 months with tone greater than 3     compared with therapy only improved the
USA            on the Ashworth Scale for 2 joints in      functional status of the subjects on the MAL
6 (RCT)        the involved upper extremity were          Quality of Movement subscale (P=.0180).
               randomized in a crossover trial to         There was a trend toward significance in the
               receive either 100 U BTX-A combined        Amount of Use subscale (P=.0605). After
               with a defined therapy program or          each 12-week period, tone had largely
               placebo injection combined with a          returned to baseline (P>.05).
               therapy program in two 12-week
               sessions. The primary functional
               outcome measure was the Motor
               Activity Log (MAL), assessed at baseline
               and at the end of treatment. Subjects
               were also assessed on physiologic
               measures including tone (Ashworth
               Scale), range of motion, and motor
Turner-Stokes Additional analysis from McCrory et al.   Goal-attainment scaling outcome were highly
et al. 2010    2009. Main outcome measures were:        correlated with reduction in spasticity
UK             individualized goal attainment and its   (rho=0.36, p=0.001) and global benefit
9 (RCT)        relationship with spasticity and other   (rho=0.45, p<0.001), but not with other
               person-centred measures - pain, mood,    outcome measures. Goals related to passive
               quality of life and global benefit.      tasks were more often achieved than those
                                                        reflecting active function.
Sun et al.     32 patients (≥1 year after stroke) with The combination group showed significantly
2010           ability to actively extend >10 degrees   greater improvements in elbow, wrist, and
Taiwan         at metacarpophalangeal and               finger spasticity (P = .019, P = .019, and P <
6 (RCT)        interphalangeal joints and 20 degrees    .001, respectively), MAL and ARAT scores
               at wrist of the affected upper limb were than the control group at 6-month
               randomized to receive BtxA + modified postinjection. Patients reported considerable
               constraint-induced movement therapy      satisfaction and no serious adverse events
               (CIMT) (combination group) (n=16) or occurred.
               BtxA + conventional rehabilitation
               (control group) (n=16) for 2 hours/day,

10. Upper Extremity Interventions                                                        pg. 94 of 150
                 3 days/week for 3 months. The primary
                 outcome assessed spasticity on the
                 Modified Ashworth Scale (MAS) at 6
                 months. Secondary outcomes assessed
                 were Motor Activity Log (MAL), Action
                 Research Arm Test (ARAT), and
                 patients' global satisfaction.
Bensmail et      15 outpatients with spastic hemiparesis      Significant differences were found between
al. 2010         and 9 healthy controls were included in      hemiparetic patients and healthy participants
France           this single-site, open-labeled study of      for all kinematic parameters. All parameters
No Score         Botox with dosages specific to each          tended to improve after Botox. This effect was
                 participant. The trajectories of reaching    significant for velocity and smoothness.
                 movements were recorded, and                 Functional scores also tended to improve.
                 kinematic variables were computed. A         Improvements were greater at M4 than at
                 clinical evaluation included the Motor       M1, although the differences were not
                 Activity Log, the Action Research Arm        significant. There were no significant
                 Test (ARAT), and the Box and Block           improvements from baseline to follow-up on
                 Test (BBT). Patients were assessed           any of the functional assessments.
                 before (M0), 1 month after a first (M1),
                 and 1 month after a second BTI (M4, at
                 4 months) in proximal and distal
Kaji et al.      109 subjects with upper limb spasticity      Higher dose BoNTA was associated with
2010             following chronic stroke were                greater reductions in MAS and DAS scores
Japan            randomized to receive a single               associated with limb positioning compared
9 (RCT)          treatment with lower-dose (120–150           with a lower dose, or placebo at 12 weeks.
                 U)(n=21) or placebo (n=11) or higher         There were no differences in the number of
                 dose (200–240 U) (n=51) BoNTA or             adverse events reported among the treatment
                 placebo (n=26). The tone of the wrist        or placebo groups.
                 flexor was assessed at baseline and at
                 weeks 0, 1, 4, 6, 8 and 12 using the
                 Modified Ashworth Scale (MAS) for
                 wrist, finger, thumb and disability in
                 activities of daily living (ADL) was rated
                 using the 4-point Disability Assessment
                 Scale (DAS).
Cousins et al.   30 subjects following stroke within the      Arm function, assessed using the ARAT
2010             previous 3 weeks with impaired               improved in all three groups between baseline
UK               grasping ability were randomized to          and week 20, but there were no significant
5 (RCT)          receive a single injection of either one-    differences among groups. In the subgroup
                 quarter, or half standard dose               analysis restricted to subjects without arm
                 botulinum toxin, or placebo (saline) to      function at baseline, there were no significant
                 prevent the development of spasticity.       differences among groups except for the
                 Arm function, active and passive             active range of elbow flexion, which was
                 movement, and spasticity at elbow and        greater in the ¼ dose groyup compared with
                 wrist were recorded at baseline, and at      the other 2 groups.
                 4, 8, 12 and 20 weeks post
                 intervention. A pre-planned subgroup
                 analysis included only subjects with no
                 arm function at baseline (Action
                 Research Arm Test score = 0).
Kanovsky et      148 chronic stroke patients from 23          A significantly higher proportion of patients
al. 2009         sites in 3 European countries with wrist     treated with NT 201 had improved by at least

10. Upper Extremity Interventions                                                            pg. 95 of 150
Germany       and finger flexor spasticity and with at   1 point on the Ashworth Scale score on the
8 (RCT)       least moderate disability in their         wrist flexors, compared with the placebo
              principal therapeutic target of the        group at the end of 20 weeks (39.7% vs.
              Disability Assessment Scale (DAS) were     19.2%, p=0.007). The proportion of
              treated either with NT 201                 responders for all other muscle groups
              (Xeomin)(median, 320 U) or placebo         (clenched fist, thumb-in-palm, flexed elbow
              and followed up for up to 20 weeks.        and pronated forearm) was significant greater
              Outcomes assessed included the             for patients in the active drug group, at 4
              Ashworth Scale, DAS, Carer Burden          weeks. On all of the other outcomes
              Scale and the global sssessment of         assessed, patients in the active drug group
              benefit, as indenitifed by the             fared better than patients in the control group
              investigators, carers and patients.        on at one assessment point.

A summary of the results from the                     Table 10.39.
“good” quality RCTs is presented in

Table 10.39 Summary of Botulinum Toxin Injection and Spasticity in Upper Extremity Post
  Author/      n              Intervention                                Main Outcome(s)
PEDro Score                                                                     Result
Kaji et al.   109     120 U Botox vs. placebo &                       Modified Asworth Scale
2010                   200 U Botox vs. placebo           (+ hi dose Botox vs. placebo, - low dose Botox vs.
9                                                                              placebo)
                                                                    Disability Assessment Scale
                                                                          (+ both groups)
McCrory et    96      500-1,000U of Dysport vs.                The Assessment of Quality of Life scale
al. 2009                placebo x 2 occasions                              (- at week 20)
Kanovsky et   148     Median of 320 U Zeomin vs.                   Ashworth Scale scores (-/+)
al. 2009                        placebo
8 (RCT)
Bakheit et    82    500 U of Dysport vs.1000 U of                    Modified Ashworth Scale
al. 2000            Dysport vs. 1500 U of Dysport     (+ for all three groups at wk 4 and week 16 in the
8                             vs. placebo           elbow and wrist and in the fingers in the 1000U group
                                                                   compared to placebo group)
                                                     Rivermead Motor Assessment (- at 4 and 16 weeks)
Bakheit et    59       Total of 1000 IU of BtxA             Summed Modified Ashworth Scale score
al. 2001            (Dysport) into 5 muscles of the                       (+ at week 4)
8                      affected arm vs. placebo          Magnitude of benefit in wrist and finger joints
                               injections           (+ over 16 wk follow-up period) Joint ROM (- at wk 4)
                                                                      Muscle pain (- at wk 4)
                                                                   Goal-attainment (- at wk 4)
                                                                     Barthel Index (- at wk 4)
                                                                    Elbow PROM (+ at 16 wks)
Simpson et    37     Single treatment of 75 units                 Decrease in wrist flexor tone
al. 1996             vs. 150 units vs. 300 units of     (+ in 300 BTX-A group at 2,4 and 6 wks post-
8                          BTX-A or placebo                                  injection)
                                                         Global Assessment of Response to Treatment
                                                        (+ with all BTX-A groups at 4 and 6 wks post-
Simpson et    60        Up to 500 U of BT-X vs                    Decrease in wrist flexor tone
al. 2009                tinzanidine vs. placebo                  (+ at 6 weeks-favouring BT-X)

10. Upper Extremity Interventions                                                       pg. 96 of 150
Bhakta et al.    40        Total of 1000 IU Dysport               Disability (+ at 2 & 6 weeks)
2000, 2008               (n=20) vs. placebo (n=20)           Caregiver burden (+ at 2, 6 & 12 weeks)
7                       divided between elbow, wrist,           MAS (finger) (+ at 2,6 &12 weeks)
                              and finger flexors.                  MAS (elbow) (+ at 2 weeks)
                                                                               Pain (-)
                                                                     Associated reactions (+)
Brashear et      126    Injection of botulinum toxin A             Disability Assessment scores
al. 2002                    (50 units) vs. placebo                        (+ at 6 weeks)
Smith et al.     25      500 units vs. 1000 units vs.            Modified Ashworth Scale at fingers
2000                    1500 units of botulinum toxin               (+ for all botulinum groups)
7                                or placebo                     Passive range of movement at wrist
                                                                    (+ for all botulinum groups)
                                                                     Finger curl distance at rest
                                                                    (+ for all botulinum groups)
                                                       Only significant difference between dose groups was
                                                      seen in improved movement at the elbow (+ 1500 Mu
Francisco et      13   High volume BTX-A (50 units/1                  Modified Ashworth Scale
al. 2004         (10 mL saline:1.2 mL delivered per            (-at 4, 8 and 12 weeks post injection)
7               stroke    4 muscles) vs. low volume
                   )    BTX-A (100 units/1 mL saline
Brasher et        15    10000 U of BTX-B or placebo                    Modified Ashworth scale
al. 2004                                                    (+ at week 2, - at weeks 4, 8, 12, and 16)
7                                                                   Global Assessment of Change
Childers et       91   Up to 2 treatments of placebo,                        Muscle tone
al. 2004                 or 90, 180, or 360U of BTX.                      (+ at weeks 1-6)
7                                                                               FIM (-)
                                                                               SF-36 (-)
Meythayler        21     100 U Botox + therapy vs.            Motor Activity Log (Quality of Use) (+)
et al. 2009                    saline + therapy               Motor Activity Log (Amount of Use) (-)
6                                                                         Asworth Scale (-)
Sun et al.        32   1,000 U Dysport + mCIMT vs.                             MAS (+)
2010                          1,000 U Dysport +               Motor Activity Log (amount of use) (+)
6                             conventional rehab
Jahangir et       27        50 U Botox vs. placebo                    Modified Ashworth Scale
al. 2007                                                                   (+ at 3 months)
6                                                                         Barthel Index (-)
                                                                              EQ-5D (-)

Suputtitada     60       Single dose of either placebo                  Modified Ashworth scale
&                         or one of 3 doses of BTX-A                 (+ in 500 and 1,000 U groups)
Suwanwela                    (350, 500 or 1,000U)                  ARAT (+ at 8 and 24 weeks 500 U)
2005                                                                BI (+ at 8 and 24 weeks 500 U)
- Indicates non-statistically significant differences between treatment groups
+ Indicates statistically significant differences between treatment groups

Discussion                                               Assessing the effectiveness of
                                                         botulinum toxin in the treatment of

10. Upper Extremity Interventions                                                       pg. 97 of 150
upper limb spasticity was difficult                 the effects with the placebo. This
owing to the broad range of doses and               approach presented difficulties when
types of agents administered. Among                 attempting to determine if escalating
the RCTs reviewed, many assessed                    doses were associated with greater
the between a single dose,                          reductions in spasticity. Generally all
administered to several sites, of                   doses of BT resulted in reduction in
botulinum toxin A as either Dysport ®,              muscle tone; however, increasingly
Botox® or Xeomin® vs. placebo. A                    higher doses were associated with
single trial assessed the benefit of BT-            muscle weakening.
type B (Brashear et al. 2004-10,000 U
BT-B). The dose equivalent is                         Conclusion Regarding Botulinum Toxin
approximately 300-500 Units of                        Injection in the Upper Extremity
Dysport equal 100 units of Botox
(O’Brian 2002). Among these trials,                   There is strong (Level Ia) that treatment
the results were ambiguous. The                       with BTX alone or in combination with
greatest benefit appeared to be                       therapy significantly decreases
realized in the patients who received                 spasticity in the upper extremity in
                                                      stroke survivors. However, it is not
Botox (Brashear et al. 2002) who had
                                                      clear that the improvements are
reductions in tone and also
                                                      sustained, nor is there strong evidence
experienced improvement in                            that they are associated with improved
functional outcome. Patients treated                  upper limb function or quality of life.
with BT-B (MyoBloc) appeared to have
the poorest response to treatment
                                                       Botulinum Toxin decreases spasticity
(Brashear et al. 2004). One trial, the
                                                       and increases range of motion;
most methodolically rigorous (McCrory
                                                       however, these improvements do not
et al. 2009) evaluated the                             necessarily result in better upper
effectiveness of 2 doses of Botox,                     extremity function.
given 12 weeks apart, compared with

Several trials assessed the effect of               10.5.3 Electrical Stimulation
several doses of botulinum toxin                    Combined with Botulinum Toxin
compared with placebo (Bakheit et al.               Injection
2000, Simpson et al. 1996, Smith et
al. 2000, Childers et al. 2004,
                                                    A single study evaluated the efficacy
Suputtitada & Suwanwela 2005). Due
                                                    of botulinum toxin injection combined
to the small sample sizes, many of the
                                                    with electrical stimulation.
authors of these studies grouped the
treatments together and compared

Table 10.40 Electrical Stimulation Combined with Botulinum Toxin (BTX) Injection in
the Upper Extremity
    Author/                     Methods                                    Outcomes
 Pedro Score
Hesse et al.   A placebo controlled trial of 24 patients    Significant muscle tone reduction of the
1998           randomized to one of four groups: 1000unit   elbow joint was most prominent for Group
Germany        BTX-A + electrical stimulation (Group A);    A. Group A experienced fewer difficulties
7 (RCT)        1000 units of BTX-A (Group B); Placebo +     while cleaning the palm of the hand, when
               electrical stimulation (Group C); and        compared to Group B and Group D.

10. Upper Extremity Interventions                                                     pg. 98 of 150
                Placebo (Group D). Intra-muscular               Patients in the BTX-A groups experienced
                injection of either BTX-A or placebo into six   fewer difficulties when putting the
                upper limb flexors. Electrical stimulation of   involved arm through a sleeve, compared
                the injected muscles with surface               to patients in groups C & D.
                electrodes, was conducted three times, ½
                hr each day for three days (Group A and C).

Table 10.41 Summary of Combined Therapy with Botulinum Toxin Injection in the
Upper Extremity
   Author/       N               Intervention                              Main Outcome(s)
 PEDro Score                                                                     Result
Hesse et al.    24         1000unit Btx A + electrical                  Muscle Tone Reduction
1998                    stimulation (Group A) vs. 1000                (- elbow joint for group A)
7 (RCT)                   units of Btx A (Group B) vs.       Reduction in difficulties while cleaning palm
                        Placebo + electrical stimulation      (+ group A compared to group B and D)
                      (Group C) vs. and Placebo (Group        Difficulties putting arm through a sleeve
                                        D).                 (+ reduction between botulinum groups and
- Indicates non-statistically significant differences between treatment groups
+ Indicates statistically significant differences between treatment groups

 Conclusions Regarding Treatment of                      One method of decreasing spasticity is
 Spasticity: Botulinum Toxin Injections                  by injecting alcohol or phenol into a
                                                         specific nerve (ie. the
 There is moderate (Level 1b) evidence                   musculocutaneous nerve) thus
 that electrical stimulation combined                    decreasing spasticity of the innervated
 with Botulinum Toxin injection is                       muscles. One of the side effects of this
 associated with reductions in muscle                    treatment is a loss of sensation;
                                                         therefore, this form of treatment is
                                                         not widely used in clinical practice. A
  Botulinum Toxin in combination with                    commonly reported side effect is
  electrical stimulation improves tone in                temporary pain (Kong and Chau
  the upper extremity.                                   1999).

10.5.4 Nerve Block and Spasticity

Table 10.42 Nerve Block and Spasticity in Upper Extremity
    Author/                   Methods                                       Outcomes
 Pedro Score
Kong and        20 patients received musculocut-      There was statistically significant improvement in
Chau 1999       aneous nerve block of hemiplegic      MAS and PROM. The mean baseline MAS score was
Singapore       upper extremity with 50% ethyl        3.7, which improved to 1.7, 2.0 and 2.1 at 4
No Score        alcohol. Outcome measures             weeks, 3 months, and 6 months respectively. The
                included spasticity severitiy as      elbow PROM was 87.3 degrees, 104.3, 103.8 and
                measured by the Modified              101.6 degrees, respectively. Four subjects had
                Ashworth Scale (MAS) and the          concomitant improvement of finger flexor

10. Upper Extremity Interventions                                                           pg. 99 of 150
                elbow passive range of motion          spasticity and another four had relief of shoulder
                (PROM) of the elbow.                   pain
Kong and        30 patients with mean onset since      20 patients had significant swelling over their
Chua 2002       stroke of 8.3 months and with          forearm associated with treatment. MAS scores
Singapore       complications secondary to flexion     improved significant from baseline 4.0 to 2.0 (4
No Score        spasticity of the wrist and fingers    wks), and then fell slightly to 2.2 (3 months) and
                were given intramuscular               2.6 (6 months). Baseline PROM values were not
                neurolysis or motor point blocks of    reported.
                the finger flexors of the hemiplegic
                upper extremity with 50% ethyl
                alcohol. Asssessments included the
                Modified Ashworth Scale (MAS),
                passive range of motion (PROM) of
                the interphalangeal joints of the
                second and fifth digits.
                Assessments were conducted
                before treatment and at 4 weeks,
                3 and 6 months.

 Conclusions Regarding Treatment of                      10.5.5 Physical Therapy in the
 Spasticity: Nerve Block                                 Treatment of Spasticity

 There is limited (Level 2) evidence that                As previously mentioned, physical
 treatment with ethyl alcohol improves                   therapy is a mainstay in the treatment
 elbow and finger PROM and can                           of spasticity. Common physical
 decrease spasticity in the upper                        modalities used in the treatment of
 extremity in stroke survivors.                          spasticity include stretching, orthoses,
                                                         casting, and cold application.
   More research is needed to determine
   whether nerve blocks decrease
   spasticity in the upper extremity.

Table 10.43 Physical Therapy and Spasticity in Upper Extremity
    Author/                        Methods                                       Outcomes
 Pedro Score
Carey 1990      24 patients were randomly assigned to             The JMTT and FTT performances of both
USA             either a no treatment control group or to a       control and experimental groups were
4 (RCT)         treatment group in which manual stretch           significantly subnormal on the pretest.
                was applied to the extrinsic finger flexor        The JMTT improvement on posttest was
                muscles for 5 minutes between the pretest         significantly greater (p < 0 .05) in
                and post-test. A joint-movement tracking          subjects than in controls. However, the
                test (JMTT) quantified control of active finger   change in FTT performance was not
                extension movement at the                         significantly different between the two
                metacarpophalangeal joint of the index            groups
                finger within the available range of active
                movement, and a force tracking test (FTT)
                quantified control of isometric finger
                extension force at the same joint within the
                subject's available force range.
Miller et al.   9 patients performed 10 quick and forceful        Although no differences between tasks

10. Upper Extremity Interventions                                                         pg. 100 of 150
1997             isometric contractions of the biceps with the    were noted, there was a trend in favour
USA              sequence of pre-test, graded resistive           of the GRE suggesting that it is not
No score         exercise (GRE) and post-test applied to both     detrimental to post-stroke spastic
                 paretic and on-paretic arm. Testing was          muscles.
                 randomized to which arm was started first.
                 Treatment and no-treatment occurred on
                 two consecutive days in counterbalanced
                 order and at the same time of day.
Horsley et al.   40 patients admitted to a rehabilitation         There were no significant differences
2007             service (19 with stroke) received routine        between groups on any of the outcomes
Australia        upper-limb retraining five days a week. In       assessed either immediately following
8 (RCT)          addition, the experimental group (n=20)          the treatment or at follow-up.
                 received 30 minutes daily stretch of the wrist
                 and finger flexors five days a week for four
                 weeks. The primary outcome was
                 contracture, measured as torque-controlled
                 passive wrist extension with the fingers
                 extended. Secondary outcomes were pain at
                 rest measured on a 10-cm visual analogue
                 scale, and upper-limb activity measured
                 using the Motor Assessment Scale.
                 Outcomes were assessed at baseline, post-
                 intervention, and one and five weeks after
                 cessation of intervention.

 Conclusions Regarding Treatment of
 Spasticity: Physical Therapy                            10.5.6 Shock Wave Treatment

 There is strong (Level 1a) evidence that                Shock wave therapy has been
 physical therapy does not reduce                        demonstrated to effectively treat a
 spasticity in the upper extremity.                      variety of bone and tendon diseases
                                                         by reducing hypertonia and may be an
  Physiotherapy does not decrease                        attractive treatment option for stroke
  spasticity in the upper extremity.                     patients instead of botulinum toxin. A
                                                         single study, which investigated the
                                                         effects of shock wave therapy, was

Table 10.44 Shock Wave Therapy and Spasticity in Upper Extremity
     Author/                       Methods                                        Outcomes
  Pedro Score
Manganotti et    20 patients with severe hypertonia               Ashworth scores for wrist flexors and
al. 2005         associated with chronic stroke received one      finger flexors decreased significantly
 Italy           sham treatment and one active treatment          after active treatment The positive
(No Score)       with extracorporeal shock wave therapy,          effects persisted for finger flexors at 12
                 given one week apart. The National               weeks, but not for wrist flexors. Passive
                 Institutes of Health Stroke Scale (NIHSS)        range of motion increase following active
                 and Ashworth Scale were assessed at 1, 4         treatment. The benefit was maintained
                 and 12 weeks following active treatment.         for 4 weeks. There were no changes in
                                                                  the NIHSS scores following treatment.

10. Upper Extremity Interventions                                                          pg. 101 of 150
A single treatment of shock wave                     10.5.7 Centrally Acting Muscle
therapy among a small group of                       Relaxants (tolperisone)
patients with spasticity in the hand
was effectively reduced for a period of              Tolperisone is a centrally acting
more than 12 weeks, with no adverse                  muscle relaxant, similar in action to
effects. This may be a promising new                 lidocaine, which acts by reducing
treatment.                                           sodium influx through nerve
                                                     membranes. It may be superior to
 Conclusions Regarding Treatment of                  other muscle relaxants in that it does
 Spasticity: Shock Wave Therapy                      not cause sedation or muscle
                                                     weakness, nor does it impair
 Based on the results from a single non-             attention-related brain functions.
 RCT, there is limited (Level 2) evidence            Tolperisone and its analogue
 that shock wave therapy can reduce                  epersione have been used successfully
 tone in the upper extremity.
                                                     in patients with spinal cord injuries.

Table 10.45 Tolperisone in the Treatment of and Spasticity in Upper Extremity
      Author/                         Methods                                    Outcomes
    Pedro Score
Stamenova et al.    120 acute stroke patients (having suffered    Patients treated with tolperisone had a
2005                from a stroke within the previous 2 months)   greater decrease in mean Ashworth
Bulgaria/Germany    with spasticity were randomized to receive    Scores (1.03 +/- 0.71 vs. 0.47 +/-
8 (RCT)             either a daily dose of 300-900 mg of          0.54). 78.3% of the patients on
                    tolperisone or placebo for 12 weeks.          tolperisone vs. 45% of the placebo
                    Outcomes assessed at the end of treatment     patients experienced a reduction by at
                    included Ashworh Scale scores.                least 1 point on the Ashworth Scale.
                                                                  Functional and overall assessments of
                                                                  efficacy confirmed superior efficacy of
                                                                  tolperisone. Adverse events occurred
                                                                  more frequently in the placebo group
                                                                  compared to treatment (26 vs. 19) and
                                                                  were mostly of mild-to-moderate
Tariq et al. 2005   26 stroke patients with hemiparesis were      17 patients completed the study; 8/13 in
Pakistan            alternatively assigned to receive either a 3- the epersione group and 9/13 who
No Score            week treatment of Eperisone (t.i.d) (n=13)    received physiotherapy. No inferential
                    or one-hour of physical therapy daily         statistics were conducted or reported.
                    (n=13). Tone was assessed at the end of the Tone was improved in 6/8 patients who
                    treatment period.                             received eperisone and in 4/9 patients
                                                                  who received phsyio. Although tone was
                                                                  assessed in both the upper and lower
                                                                  extremities it is unclear what the
                                                                  reduction in tone was in the upper

 Conclusions Regarding Centrally                       There is moderate (Level 1b) evidence
 Acting Muscle Relaxants                               that tolperisone can reduce spasticity
                                                       following stroke.

10. Upper Extremity Interventions                                                    pg. 102 of 150
10.6 EMG/Biofeedback                                    research overview and meta-analysis
                                                        on the efficacy of electromyographic
EMG biofeedback uses instrumentation
                                                        biofeedback compared with
applied to the patient’s muscle(s) with                 conventional physical therapy for
external electrodes to capture motor                    upper-extremity function in stroke
unit electrical potentials. As the                      patients. They concluded that neither
instrumentation converts the                            therapy was superior to the other.
potentials into visual or audio
information, the patient is able to                     Eleven RCTs evaluating
have a visual picture or auditory                       EMG/biofeedback therapy were
indication of how much they are                         identified. The results are presented
activating the muscle. In 1994,                         in Tables 10.46 and 10.47.
Moreland and Thomson published their

Table 10.46 EMG/Biofeedback Studies
    Author/                        Methods                                      Outcomes
  Pedro Score
Lee et al. 1976   18 patients with reduced deltoid muscle       There were no significant differences
USA               strength were randomly assigned to 1 of 6     between the 3 treatment conditions.
4 (RCT)           possible treatment orders of 3
                  interventions conditions, over 3
                  consecutive days in a crossover designed
                  i) True myofeedback
                  ii) Placebo myofeedback
                  iii) No myofeedback with conventional
                  Each daily training section consisted of 20
                  consecutive contractions of the deltoid
                  muscle for 5 sec with a rest interval of 10
                  sec. Peak amplitude of the averaged
                  myolectric signal maintained for 1 sec
                  were collected.
Mroczek et al.    5 chronic stroke patients received 4          ROM improved following treatment with
1978              weeks of biofeedback therapy followed by      both physical therapy and EMG therapy.
USA               4 weeks of physical therapy while 4           When the results were pooled there no
5 (RCT)           patients received training in the opposite    significant differences between treatment
                  order. Active range of motion (ROM) was       conditions.
                  assessed at baseline and weeks 4, 7, 10
                  and 12.
Smith 1979        12 stroke patients with movement              Greater improvements reported for
Australia         disorder were randomly assigned to            patients receiving biofeedback therapy in
4 (RCT)           receive a biofeedback therapy or              the areas of sensation, muscle tone,
                  physiotherapy for 6 weeks.                    voluntary isolated movement, synergistic
                                                                movement pattern, functional activities,
                                                                gross motor abilities and coordination.
                                                                However, no inferential statistics were
Greenberg and     20 stroke patients were randomized to         There were no differences in active elbow
Fowler 1980       receive either conventional OT (n=10) or      extension measured on three occasions by

10. Upper Extremity Interventions                                                          pg. 103 of 150
USA              audiovisual kinesthetic biofeedback           a conventional goniometer.
5 (RCT)          associated with active elbow extension
                 (n=10). All subjects were treated 30 min
                 x 2/week x 4weeks.
Hurd et al.      24 hemiplegic inpatients were randomized      There were no statistically significant
1980             to receive 10 sessions over a two-week        differences in either active range of
6 (RCT)          period of either actual or simulated          motion or muscle activity between the two
                 myofeedback                                   groups.
Prevo et al.     28 patients were assigned to receive          EMG feedback therapy had no specific
1982             either EMG biofeedback (n=9), confined        effect on proximal and distal agonists of
Netherlands      to augmenting the reduced muscle              the hemiplegic arm compared to
3 (RCT)          activity in one proximal and one distal       conventional therapy.
                 agonist, and to decreasing the excessive
                 activity in one proximal agonist and in two
                 distal muscle groups or to conventional
                 physical therapy (n=9) for 2.5 months.
Basmajian et     37 hemiplegic stroke patients were            There were no differences between the
al. 1982         randomized to receive physical therapy +      groups on any of the outcome measures
Canada           EMG biofeedback or physical therapy           (Upper Extremity Function Test,
6 (RCT)          using a general neurophysiological            Minnesota rate of manipulation test, 9-
                 approach, for 40 min x 3x/week for 5          hole peg test, grip and pinch test).
Wolf et al. 1983 22 chronic stroke patients who each           Those patients receiving feedback training
USA              received 60 EMG feedback training             showed significant improvements in
No Score         sessions. EMG data were compared with         numerous neuromuscular measures but
                 changes measured from a Control Group         not in functional measures.
                 of 9 (no treatment) patients.
Inglis et al.    30 patients with stroke onset of at least 6   Patients in the experimental condition
1984             months were randomized to receive 20          demonstrated significant improvements in
5 (RCT)          sessions of EMG biofeedback treatments        all measured parameters compared to
                 + routine physiotherapy or to routine         patients in the control condition.
                 physiotherapy. Patients in the control
                 group were then crossed over to receive
                 the experimental therapy. Assessments of
                 muscle strength, active range of motion,
                 picture goniometry and Brunnstrom’s
                 staging were conducted at baseline and at
                 the end of treatment.
Turczynski et    12 chronic stroke patients with               The standarized tests of motor skills could
al. 1984         hemiparesis received electromyographic        only be performed properly in 6 patients.
Germany          feedback exercises applied to upper           The test scores of those 6 patients were
No Score         extremity muscles. Outcomes measures          not significant from pre- to post-
                 included objective and standardized tests     treatment.
                 of motor skills.
Basmajian et     29 hemiparetic stroke patients were           There were no differences between the
al. 1987         randomized to receive either integrated       groups on any of the outcome measures
Canada           behavioural and physical therapy              (Upper Extremity Function Test, finger
6 (RCT)          (including EMG) (n=13) or physical            oscillation tests).
                 therapy based on neurofacilitatory
                 techniques (n=16), of 45 min x 3
                 days/week x 5 weeks.
Crow et al.      40 stroke patients were randomized to         After 6 weeks of treatment, the difference
1989             receive routine physical therapy + EMG        between the groups was statistically

10. Upper Extremity Interventions                                                        pg. 104 of 150
UK               biofeedback (n=20) or sham treatment +      significant for both outcome measures
8 (RCT)          routine physical therapy, for 12 weeks.     (ARA-Action Research Arm test and BFM-
                                                             Brunnstrom-Fugl Meyer test). However,
                                                             following an additional 6 weeks without
                                                             the treatment, there were no significant
                                                             differences between the groups.
Bate et al. 1992 16 stroke patients recruited from day       Both groups tracked the target more
Australia        hospitals or rehabilitation hospitals were  accurately following training. Transfer
4 (RCT)          randomized to receive EMG biofeedback       tests failed to demonstrate effects of
                 from the spastic elbow flexor muscles       feedback on accuracy of tracking or on
                 during movement practice or to a control electromyographic activity during
                 condition, which did not receive EMG in a performance of the practiced task without
                 single training session. Patients practiced feedback. Moreover, the group that was
                 a pursuit tracking task by following a      trained with electromyographic feedback
                 moving target with elbow flexion and        exhibited negative transfer on variants of
                 extension, simultaneously attempting to     the practiced task: tracking faster or less
                 reduce the activity of elbow flexors.       predictable targets
Armagan et al. 27 patients with hemiparesis resulting        No significant difference between the
2003             from stroke 3-6 months previously were      groups in Brunnstrom stages of hand
Turkey           randomized to i) an exercise program        recovery or scale for judging the
7 (RCT)          (Brunnstrom approach) + EMG                 performance of the movement complex of
                 biofeedback or to the same intervention     drinking from a glass. Significant between
                 but with placebo EMG therapy (machine       group differences for active range of
                 turned on, but to feedback to patient).     motion scores and changes in EMG surface
                 Both treatments were applied 5x/week for potentials.
                 20 days. Evaluations included:
                 Goniometric measurements for wrist
                 extension (range of motion), scale for
                 judging the performance of drinking from
                 a glass, Brunnstrom’s stages of recovery
                 for hand, and surface EMG potentials,
                 assessed before and after treatment.
Hemmen &         27 patients with stroke onset of greater    There were no significant differences
Seelen 2007      than 3 weeks were randomized to receive between groups on either of the outcomes
Netherlands      conventional electrostimulation (n=14) or assessed; however, patients in both
7 (RCT)          to a group that received arm-hand           groups experienced significant
                 function training based on EMG-             improvement from baseline to one-year
                 biofeedback combined with movement          follow-up.
                 imagery. In both groups training was
                 carried out for 3 months (5 days/wk, 30
                 min) in addition to usual therapy.
                 Outcomes were assessed at baseline, 12
                 months and 12 months and included Fugl-
                 Meyer (FM) scores and the Action
                 Research Arm (ARA) test.

10. Upper Extremity Interventions                                                      pg. 105 of 150
Table 10.47 Summary Table of RCTs for EMG/Biofeedback Therapy and the
Hemiparetic Upper Extremity
    Author         n                 Intervention                           Main Outcome(s)
 PEDro Score                                                                      Result
Crow et al.       40       EMG/Biofeedback Therapy vs.                 Action Research Arm test
1989                         Sham EMG/biofeedback                          post-treatment (+)
8 (RCT)                                                                   6 weeks follow-up (-)
Hemmen &          27       EMG biofeedback + movement                        Fugl-Meyer (-)
Seelen 2007                  imagery vs. conventional                 Action Research Arm test (-)
Armagan et        27       EMG/Biofeedback Therapy vs.                Active range of motion (+)
al. 2003                      Sham EMG/biofeedback               Changes in EMG surface potentials (+)
7 (RCT)                                                                 Brunnstrom stages (-)
                                                                        Complex movement (-)
Basmajian et      29       EMG/Biofeedback Therapy vs.             Upper extremity function test (-)
al. 1987                   Physical Therapy using neuro-               Finger Oscillation test (-)
6 (RCT)                              facilitatory
Hurd et al.       24     Actual myofeedback vs. simulated              Active range of motion (-)
1980                                myofeedback                            Muscle activity (-)
6 (RCT)
Bamajian et       37      EMG/Biofeedback Therapy vs.              Upper extremity function test (-)
al. 1982                  Physical Therapy using neuro-            Minn rate of manipulation test (-)
6 (RCT)                      physiological approach                       9-hole peg test (-)
Inglis et al.     30     EMG/Biofeedback+ physiotherapy               Active range of motion (+)
1984                            vs. Physiotherapy                          Brunnstrom (+)
5 (RCT)                                                                   Muscle strength (+)
Bate et al.       16              EMG vs. no EMG                           Tracking task (-)
4 (RCT)
Greenberg &       20       EMG/Biofeedback Therapy vs.                 Active elbow extension (-)
Fowler 1980              Conventional Occupational Therapy
5 (RCT)
Smith 1979        12          Biofeedback therapy vs.            No inferential statistics were reported.
Australia                          Physiotherapy
4 (RCT)
Mrocek et al.      9        EMG biofeedback vs. Physical                  Range of Motion (-)
1978                                  therapy
5 (RCT)
Lee et al.        18      True myofeedback vs. Placebo                     Peak amplitude (-)
1976                     myofeedback vs. No myofeedback
4 (RCT)                     with conventional training.

Prevo et al.       28        EMG/Biofeedback Therapy vs.             Proximal and distal agonists (-)
1982                              Conventional Therapy
3 (RCT)
- Indicates non-statistically significant differences between treatment groups
+ Indicates statistically significant differences between treatment groups

10. Upper Extremity Interventions                                                         pg. 106 of 150
 Conclusions Regarding Efficacy of                      of the brain, which in turn is believed
 EMG/Biofeedback Therapy                                to be beneficial to rehabilitation
                                                        (Sonde et al. 1998). Robbins et al.
 There is strong (Level 1a) evidence that               (2006) described the TENS current
 EMG/Biofeedback therapy is not                         intensity to be beneath motor
 superior to other forms of treatment.                  threshold, although capable of
                                                        generating a “pins-and-needles
  EMG/Biofeedback therapy is not                        sensation.” Similar to acupuncture,
  superior to other forms of treatment                  transcutaneous electrical nerve
  in the treatment of the hemiparetic                   stimulation is one method of achieving
  upper extremity.                                      increased afferent stimulation (Sonde
                                                        et al. 1998). Several trials have
10.7 Transcutaneous Electrical                          examined the use of TENS treatment
Nerve Stimulation (TENS)                                in the restoration of motor function
                                                        following stroke. Many of the trials
The use of afferent stimulation in                      presented below assessed motor
stroke rehabilitation is not an                         function, pain, spasticity and a variety
uncommon practice. The rationale                        of other outcomes in both the upper
behind afferent stimulation is that                     and lower extremity.
increased inflow of signals from
sensory modalities enhances plasticity

Table 10.48 TENS in the Treatment of Upper Extremity
  Author,                    Methods                                     Outcomes
Potisk et al.
            20 stroke patients with hemiplegia (>3      18 of the 20 patients significantly decreased
1995        months post-stroke) had surface             in resistive torques at all frequencies of
Slovenia    electrodes placed over the affected         passive ankle movements after 20 min of
No Score    limbs’ sural nerve. The patients then       applying TENS. This significant reduction
            received 20 minutes of TENS with            persisted 15, 30 and 45 minutes following
            impulse frequency of 100 Hz.                TENS, but was not significant 60 minutes
            Evaluation measures included resistive      after TENS. The decrease in resistive
            torques, spasticity, and                    torques was frequently linked with a
            electromyography (EMG) stretch reflex       reduction in reflex electromyographic
            activity.                                   activity.
Sonde et    44 patients randomized to receive           Fugl-Meyer scores of patients in the
al. 1998    either physiotherapy or to receive in       treatment group increased significantly
Sweden      addition to physiotherapy of the upper      compared to patients in the control group.
5 (RCT)     extremity 2 x/ week, low-intensity, low-    No changes in spasticity or pain associated
            frequency (1.7Hz) transcutaneous            with the treatment were observed. Barthel
            electrical stimulation (TENS) for 60        scores did not change significantly.
            minutes, 5 days per week for 3 months,
            for treatment of a paretic arm.
Tekeoðlu et A double blind randomised controlled        At 8 weeks, patients in both groups had
al.         trial of 60 patients. Patients received     significantly improved their BI scores
1998        either basic neurophysiological             compared to baseline. Patients in the
Turkey      rehabilitation or received in addition to   treatment group experienced greater
 6 (RCT)    the basic neurophysiological rehab          improvement in BI scores compared to the

10. Upper Extremity Interventions                                                        pg. 107 of 150
              treatment, active TENS for 40 sessions       control group. Significant reduction in
              over 8 weeks with a frequency of 100Hz       Ashworth scores was observed in both
              at intensity that patients could tolerate.   groups.
Johansson     150 patients were randomized to              No significant differences were observed
et al. 2001   receive either acupuncture (including        between groups on any of the outcome
Sweden        electroacupuncture), high-intensity, low     measures (Rivermead Mobility Index,
8 (RCT)       frequency TENS or subliminal high-           Walking Ability, Barthel Index, Nottingham
              frequency, low intensity transcutaneous      Health Profile, Nine Hole Peg Test).
              electrostimulation (control).
Sonde et      Three-year follow-up of 28 from 1998   Motor function of the paretic arm had
al. 2000      study.                                 deteriorated in both groups. Increased
Sweden                                               spasticity was seen in both groups. ADL
5 (RCT)                                              scores remained at a similar level in the low
                                                     TENS group, whereas the control group had
                                                     deteriorated during the same time period.
Peurala et Cutaneous stimulation was delivered to Modified Motor Assessment Scale, 10-metre
al. 2002    59 patients with chronic stroke, twice   walking test, paretic hand function, upper
Finland     daily for 20 min, as part of their 3-    limb skin sensation and somatosensory
No score    week, yearly, inpatient rehabilitation   evoked potentials, normality classification
            program. 32 patients received            of paretic upper limb and paretic lower limb
            treatment of their affected hand, while improved significantly in the treatment
            8 received a no treatment control        group. When active hand treatment and
            (sham). 19 patients received treatment placebo hand treatment were compared, a
            in their affected foot.                  significant improvement in the sensory and
                                                     motor function was observed only in the
                                                     actively treated group.
Rorsman & 54 patients were randomized to 1 of 3      No significant differences were found
Johansson groups: 1) acupuncture +                   between the control group and the
2006        electroacupunture, 2) high intensity,    experimental groups for Activities of Daily
Sweden      low-frequency transcutaneous electrical Living (Barthel Index) or overall motor
8 (RCT)     nerve stimulation (TENS) group or 3)     function (Rivermead Mobility Index).
            low-intensity subliminal high-frequency Although significant differences were seen
            TENS (control). Treatment sessions       in patients emotional status for a variety of
            were 30 min, 2 days/week for 10 weeks tests.
            Outcome Measures included: cognition
            performance and emotional functioning.
Yozbatiran 36 acute stroke patients were assigned There were significant between group
et al. 2006 in a ranked order to either a TENS       differences in mean changes in the hand
Turkey      group of a control group. Patients in    function test, favouring the TENS group
No Score    both groups received an hour per day     (2.38 vs. 1.22, p=0.044).
            for 10 days of physical therapy
            according to the Bobath method. The
            treatment group received 1 hour of
            electrical stimulation of the finger and
            wrist extensors. Evaluations performed
            at the beginning and the end of
            treatment included kinaesthesia and
            position sense tests, and hand function
            and movement scales.

In five studies the effect of TENS was                     1998, Johansson et al. 2001 and
investigated in both upper and lower                       Peurala et al. 2002, Rorsman and
extremity functioning (Tekeoðlu et al.                     Johansson 2006, Potisk et al. 1995).

10. Upper Extremity Interventions                                                          pg. 108 of 150
Three RCTs examining the                                 studies, the effects of TENS were
effectiveness of TENS treatment were                     tested only on the upper extremity,
reviewed Tekeoðlu et al. (1998)                          while three evaluated the effect of the
hypothesized that repeated application                   treatment in the lower extremity only.
of TENS might decrease clinical                          The results from RCTs examining
spasticity and also improve motor                        TENS treatment are summarized in
function of the paretic extremity in the                 Table 10.49.
hemiparetic patient. In four other

Table 10.49 Summary of the RCTs Evaluating TENS in the Treatment of Upper
  Author/PEDro           N              Intervention                           Outcome
Johansson et al.             Acupuncture vs. TENS vs. control   Barthel Index (-)
2001                150                                          Nottingham Health Profile (-)
8                                                               Nine Hole Peg Test (-)
Tekeoðlu et al.     60       Rehabilitation + TENS vs.          Barthel Index improvement (+)
1998                         rehabilitation
Sonde et al. 1998   44       TENS + physiotherapy vs.           Fugl Meyer (+)
5                            physiotherapy                      Pain (-)
                                                                Barthel Index (-)

 Conclusions Regarding                                   neurons or reflex pathways. Three
 Transcutaneous Electrical Nerve                         forms of NMES are available: 1) cyclic
 Stimulation                                             NMES, which contracts paretic
                                                         muscles on a preset schedule and
 There is conflicting (Level 4) evidence                 does not require participation on the
 that treatment with TENS in the upper                   part of the patient; 2) EMG triggered
 extremity improves a variety of                         NMES, which may be used for patients
 outcomes, including motor recovery,
                                                         who are able to partially activate a
 spasticity and ADLs.
                                                         paretic muscle and may have a
                                                         greater theraputic effect; 3)
  It is uncertain whether TENS                           neuroprosthetic applications of NMES,
  improves outcomes post stroke.                         which can ultimately improve or
                                                         restore the grasp and manipulation
10.8 Functional Electrical                               functions required for typical ADLs
Stimulation (FES)                                        (Popovic et a. 2002).

Neuromuscular electrical stimulation                     Several reviewsand meta-analyses
(NMES) can be used to improve motor                      examining the benefit of NMES have
recovery, reduce pain and spasticity,                    been conducted. A meta-analysis of
strengthen muscles and increase                          four studies concluded that FES
range of motion following stroke.                        enhanced strength (Glanz et al.
Functional electrical stimulation (FES)                  1996). However, conclusions are
refers to the application of NMES to                     limited by the methodology of the
help achieve a functional task. FES is                   trials (small sample size, inadequate
a technique that uses bursts of short                    blinding) and it was difficult to link
electrical pulses to generate muscle                     improved strength with improved
contraction by stimulating motor                         function.

10. Upper Extremity Interventions                                                    pg. 109 of 150
A systematic review by de Kroon et al.                and two ADL measures. The review
(2002) assessed the effect of                         included four planned treatment
therapeutic electrical stimulation of                 contrasts:1) ES vs. no treatment; 2)
the affected upper extremity in                       ES vs. placebo stimulation; 3) ES vs.
improving motor control and                           conventional therapy and 4) One type
functional abilities after stroke. The                of ES vs. an alternative type of ES.
authors included 6 RCTs in their                      With respect to the assessment of
review. The authors concluded that                    treatments specific to the upper
there is a positive effect of electrical              extremity, five outcomes were
stimulation on motor control, but that                associated with a statistically
no conclusions could be drawn                         significant treatment effect. With one
regarding its effect on functional                    exception, all of the pooled analyses
abilities.                                            were based on the results from only
                                                      one study. The results from pooled
A Cochrane review (Pomeroy et al.                     analyses with positive results are
2006) examined the use of all forms                   presented in Table 10.50. The
of electrostimulation (ES) in the                     authors concluded that there was
recovery of functional ability following              insufficient evidence to guide practice
stroke. This review assessed the                      on the efficacy of ES.
efficacy of functional electrical
stimulation (both as a form of
neuromuscular retraining and as a                         Studies included in de Kroon et al.
form of neuroprosthesis/orthosis),                        (2002)
transculaneous electrical nerve
stimulation, EMG and                                      De Kroon et al. 2002
electroacupuncture. Twenty-four RCTs                      Chae et al. 1998
evaluating the efficacy of treatment on                   Francisco et al. 1998
both the upper and lower extremities                      Sonde et al. 1998
were included. Among the trials of                        Powell et al. 1999
                                                          Cauraugh et al. 2000
upper extremity interventions, the
                                                          Bowman et al. 1979
primary outcome included nine
measures of functional motor ability

Table 10.50 Pooled Analysis from 2006 Cochrane Review Assessing Efficacy of ES as a
Therapy for the Upper Extremity
Treatment Contrast                               Standardized Mean Difference (95% CI)
Outcome Assessed
ES vs. No treatment
Motor reaction time                       1.18 (0.00, 2.37)
Isometric torque                          1.02 (0.46, 1.59)
Box & Block test                          1.28 (0.00, 2.56) *
Upper Extremity Drawing test              -1.40 (-2.25, -0.56) (favours no treatment)
ES vs. Placebo
Jebsen Hand Function test feeding         1.36 (0.24, 2.48)
ES vs. Conventional Therapy               No outcomes were statistically significant

Comparison of Different Forms of ES       No comparisons conducted or reported

* All 3 studies included in the pooled analysis were authored by the same person (Cauraugh)

10. Upper Extremity Interventions                                                       pg. 110 of 150
Meilink et al. (2008) examined the                    reaction time and sustained
effectiveness of EMG-triggered NMES                   contraction). The authors speculated
applied to the extensor muscles of the                that one of the reasons for the null
forearm to improve hand function                      findings was that the majority of
following stroke. The review included                 studies had included subjects in either
the results of 8 studies (157 patients).              the sub acute or chronic stages of
Compared with usual care, there was                   stroke. They hypothesized that there
a non-statistically significant                       may be a critical 5 week time window
treatment effect for all outcomes                     following stroke during which dexterity
assessed (Fugl-Meyer scores, Box &                    is most likely to be regained.
Block test, Action Research Arm test,

Individual Studies

Table 10.51 FES Studies in the Upper Extremity
    Author/                     Methods                                     Outcomes
  Pedro Score
Bowman et al.   30 acute stroke patients were randomly   With the wrist positioned at 300 flexion, the
1979            assigned to receive conventional         average isometric extension torque had
USA             treatment of hand and wrist (n=15) or    increased by 280% compared to a 70%
3 (RCT)         conventional treatment + positional      increase in the control group (p<0.25). Study
                feedback stimulation therapy (n=15)      group patients achieved a 200% gain in
                for 30 min x 5 days/week x 4 weeks.      selective range of motion over baseline levels,
                                                         compared to a 50% gain in the control group
                                                         patients (p<0.05).
Kraft et al.    22 right-handed patients were assigned The aggregate Fugl-Meyer (FM) scores for the
1992            to one of 4: 1) EMG-electrical           groups receiving treatment was significantly
USA             stimulation (es)(n=6), 2) low intensity increased from baseline to post-treatment and
No Score        es (n=4), 3) proprioceptive nerve        the improvements were maintained at 9
                facilitation (PNF) (n=3) or 4) no        months follow-up (p<0.005). There was no
                treatment control (n=5). Patients        significant improvement in grip strength at 3 or
                received treatment for 3 months and      9 months. There were no significant
                were assessed at 3 and 9 months          improvements in either grip strength or FM
                following treatment.                     scores among the control patients.
Faghri et al.   26 acute stroke patients were            After treatment, the FES group showed
1994            randomized to receive either FES +       significant increased arm function, tone and
USA             conventional therapy, or only            EMG activity compared with the control group.
4 (RCT)         conventional therapy.
King 1996       21 chronic stroke subjects were          Subjects in the NMES group experienced
USA             randomized to receive a single 10 min    significantly greater reduction in tone, measure
4 (RCT)         session of either neuromuscular          by a torque meter (9.6 cm/kg vs. 4.6 cm/kg).
                electrical stimulation (NMES) or passive
                stretch as a means to reduce tone.
Faghri and      26 acute stroke patients were            Weekly evaluation of arm and shoulder muscle
Rodgers 1997    randomized to receive either             function showed significant improvement.
USA             conventional therapy + FES, daily for 6 Significantly improved function assessed by
4 (RCT)         weeks (n=13) or conventional therapy range of motion at 4/7 testing times, shoulder
                (n=13).                                  muscle tone at 5/7 testing times, and EMG

10. Upper Extremity Interventions                                                      pg. 111 of 150
                                                          activity of the posterior deltoid muscle at 3/7
                                                          testing times.
Heckermann et 28 stroke patients with severe              All patients improved from baseline to post
al. 1997      hemiparesis were randomized to              treatment. The only significant between group
Germany       receive EMG-triggered FES + inpatient       result was improvement in the range of motion
4 (RCT)       physical therapy (n=14) or physical         of the hand extensors.
              therapy alone based on the Bobath
              method (n=14). Each patient was
              treated by a physiotherapist for 45 min
              x 5 days/week. FES treatment was
              delivered to the target muscles (upper
              arm extensors, knee flexors and ankle
              extensors. Each group of muscles was
              stimulated 15 times per session, 5
              days/ week x 4 weeks. Outcomes
              included spasticity, range of motion and
              Barthel Index.
Hummelsheim 12 stroke patients were studied using a       Non-statistically significant improvement in
et al. 1997   multiple baseline design. Following a       Modified Ashworth and Rivermead Motor
Germany       baseline phase (phase A) that lasted        Assessment scores over the treatment period.
No Score      between one and three weeks, all            A statistically significant improvement was
              patients received electrical nerve          reported in functional motor capacity during
              stimulation for 20 min 2x/day x 2           Phase C (p<0.008).
              weeks (phase B), followed by two
              weeks of a standardized training
              programme (phase C) emphasizing
              repetitive motor training, in addition to
              routine therapy.
Pandyan and   11 stroke patients received 2 weeks of       Passive extension of the wrist had improved
Granat 1997   regular rehabilitation, followed by 2        significantly immediately following treatment,
UK            weeks with electrical stimulation +          but were lost both at 1 hour post treatment and
No Score      rehabilitation, followed by 2 weeks of       at 2 weeks. The measures of resting wrist angle
              rehabilitation only.                         showed the same trend, whereby there was an
                                                           improvement shown right after treatment, but
                                                           which quickly declined.
Chae et          46 acute stroke rehabilitation inpatients Treatment group had significantly greater gains
al.1998          were randomized to receive surface        in upper extremity Fugl-Meyer scores
USA              neuromuscular stimulation to produce      compared to controls immediately following
6 (RCT)          wrist and finger extension exercises in   treatment (13.1 vs. 6.5, p=0.05), at 4 weeks
                 addition to routine rehabilitation (n=14) (17.9 vs. 9.7, p=0.05), but not at 12 weeks
                 or to sham stimulation + routine          (20.6 vs. 11.2, p=0.06) following treatment.
                 rehabilitation (n=14), for 1 hr/day x 15
                 sessions. 18 subjects were excluded
                 after randomization.
Francisco et al. 9 acute stroke rehabilitation inpatients Treatment patients had significantly greater
1998             were randomized to receive EMG-           gains in upper extremity Fugl-Meyer (27 vs.
USA              electrical stimulation + standard         10.4, p=0.05) and upper extremity FIM scores
5 (RCT)          therapy (n=4), for 30 min/day x 5         (6.0 vs. 3.4, p=0.02), compared to controls.
                 days/week for the duration of hospital
                 stay or to standard therapy alone
Powell et al.    60 hemiparetic stroke patients, 2-4       Change in isometric strength of wrist extensors
1999             weeks post stroke were randomized to was significantly greater in the ES group, at 8

10. Upper Extremity Interventions                                                      pg. 112 of 150
UK              receive standard rehabilitation+            and 32 weeks (p=0.004 and p=0.014). Grasp
7 (RCT)         electrical stimulation (ES) of wrist        and grip scores on the Action Research Arm
                extensors for 30 min/day x 3x/week x 8      test had increased significantly in the ES group
                weeks (n=25) or to routine                  at 8 weeks (p=0.013 and p=0.02).
                rehabilitation (n=23).
Cauraugh et al. 11 patients, greater than one-year post     ES group moved significantly more blocks on
2000            stroke onset, with upper extremity          the Box and Block test (p<0.05). No
USA             impairment were randomized to receive       differences on the Motor Assessment Scale and
4 (RCT)         either passive range of motion and          the Fugl-Meyer (upper extremity) test.
                stretching exercises + electrical
                stimulation (ES), 30 min x 12 sessions
                (n=7) or to a control group without ES
Wang et al.     32 hemiplegic stroke patients were          Patients in the short duration group
2002            placed into either a short or long-         demonstrated significant improvement in Fugl-
China           duration group, depending on the            Meyer scores compared to patients in the
5 (RCT)         length of their hemiplegia and then         control group after the first 6 weeks of therapy,
                subjects in each group were randomly        during the following 6 weeks of no therapy and
                assigned to either control or               again following an additional 6 weeks of
                experimental condition. Subjects in         therapy. There was no significant improvement
                experimental sub groups were treated        for patients receiving FES in the long-duration
                in an A-B-A design that consisted of        group at any point, compared to the control
                FES training (A), routine therapy or        group.
                regular daily activity without FES (B)
                and another FES training (A). Each
                period lasted for 6 wks.
Cauraugh and 26 stroke patients with chronic                Both the 5 sec and 10 sec groups moved
Kim 2003        hemiparesis were randomly assigned to       significantly more blocks (Box and Block Test),
USA             one of three groups: 1) 0 sec               significantly decreased their reaction times
5 (RCT)         stimulation, 2) 5 sec stimulation, and 3)   across test sessions, and significantly improved
                10 sec stimulation. Stimulation was         muscle contraction, while the 0 sec group did
                applied to the back of the impaired         not show significant improvement on any of the
                forearm. All patients completed 4 days      outcome measures.
                (90 min/day) of rehabilitation training
                over a 2 wk period. Outcome measures
                included: Box and Block Test (manual
                dexterity), Reaction time, and sustained
                wrist/finger contraction.
Popovic et al.  28 patients with acute stroke were          Patients in both the HFG/FET and LFG/FET
2003            divided into lower functioning and          groups picked up significantly more objects in 2
Denmark         higher functioning groups (LFG, HFG)        min (UEFT) at all testing intervals, except
6 (RCT)         and randomized to receive functional        baseline, compared to controls. Patients in the
                electrical therapy (FET) + conventional     HFG/FET group had significantly higher DT
                therapy or regular inpatient therapy.       scores compared to control at all testing
                Time from stroke onset to                   intervals, except baseline, compared to
                randomization ranged from 4-11 weeks.       controls. DT differences between treatment and
                Patients in the FET group performed 30      control groups were only significantly different
                min exercise with the paretic arm and       at one testing point (week 13).
                hand, facilitated with a neural
                prosthesis that controlled the opening
                and closing of the hand regular therapy
                +.everyday for 3 weeks. Patients in
                the control group received similar

10. Upper Extremity Interventions                                                        pg. 113 of 150
                treatment, less electrical stimulation.
                Assessments at study start, 3,6,12, and
                26 weeks included the Upper Extremity
                Function test (UEFT), drawing test (DT),
                coordination of elbow and shoulder
                movements, spasticity and a structured
                interview (Reduced Upper Extremity
                Motor Activity Log)
Popovic et al.  41 acute stroke patients were                Patients who participated in the FES during the
2004            randomized to receive 3 weeks of daily       acute phase of hemiplegia reached functionality
Denmark         FES treatments lasting 30 min each,          of the paretic arm, on average, in less than 6
6 (RCT)         either immediately or following a delay      weeks, and maintained this near-normal use of
                of 52- 56 weeks. All subjects also           the arm and hand throughout the follow-up.
                participated in a consecutive 3-week         The gains in all outcome scores were
                task-oriented therapy training program       significantly larger in the early group after FES
                in addition to routine rehabilitation. The   and at all follow-ups compared with the scores
                outcomes assessed were the Upper             before the treatment. Subjects in the delayed
                Extremity Function Test (UEFT), the          group also experienced gains on all outcomes
                Drawing Test (DT), the Modified              assessed, although they were not statistically
                Ashworth Scale (MAS), range of               significant. The speed of recovery was larger
                movement, and the questionnaire              during the period of the FES compared with the
                estimating the patients' satisfaction        follow-up period. The gains in the immediate
                with the usage of the paretic arm.           FES group were significantly greater compared
                Assessments were conducted at                with the gains made by subjects in the delayed
                baseline, at 78 weeks and several other      group.
                times during the study period, which
                varied by outcome.
Kimberly et al. 16 chronic stroke patients were              There was statistically significant improvement
2004            randomized to receive intensive              in strength only from pre-test to post-test.
USA             neuromuscular electrical stimulation         Following the active treatment, patients
7 (RCT)         (NMES) at home (6 hrs/day for 10 days        improved significantly on the BB test,2
                over a 3-week period) or a similar           components of the MAL (amount of use score
                (sham) treatment. Patients were then         and how well score) and two components of the
                crossed over to the other treatment          JTHFT (page turning and feeding). Using fMRI
                arm. Time from stroke onset to               and a finger-tracking task, an index of cortical
                randomization was 7-58 months Pre-           intensity in the ipsilateral somatosensory cortex
                test, post-testing included Block & Box      increased significantly from pre-test to post-
                test (BB), Motor Activity Log (MAL),         test following treatment.
                Jebsen Taylor Hand Function Test
                (JTHFT), isometric strength of the index
                finger extension finger movement
                control and fMRI.
De Kroon et al. 30 chronic stroke patients with spastic      No significant differences between the groups.
2004            paresis were randomized to one of two
Netherlands     treatment programs i) Alternating
6 (RCT)         electrical stimulation of the extensor
                and flexor muscles of the hand or ii)
                Electrical stimulation of the extensors
                only. Treatment lasted for 6 wks, with
                stimulation time increased gradually
                from 20 min/session to 1 hr/session.
                Assessed immediately before start of
                treatment, at the end of 6-wk
                treatment period, and after a follow-up

10. Upper Extremity Interventions                                                         pg. 114 of 150
               period of 6 wks. Primary outcome was
               Action Research Arm test. Secondary:
               grip strength; Motricity Index;
               Ashworth Scale; and range of motion of
               the wrist.
Ring &         22 chronic stroke patients with             Patients in the FES group experienced
Rosenthal 2005 moderate to severe upper extremity          significant improvement in reduction of
Israel         hemiparesis were randomized to              spasticity and improved performance on the
6 (RCT)        receive home-based neurostimulation         Box and Block test and 2/3 items on the
               with a 5-electrode neuroprosthetic          Jebsen-Taylor assessment, relative to the
               device or to a no device control            control group. Among the few number of
               condition. Patients in the treatment        patients with edema and pain, only those in the
               group received FES treatments which         FES group reported improvement.
               peaked at 50 min, 3x per day for 6
               weeks. All patients participated in a day
               hospital outpatient rehabilitation
               program. Assessments included
               modified Ashworth Scale, Blocks and
               Box test and 3 items on the Jebsen-
               Taylor assessment.
Gabr et al.     In a crossover designed study, 12          No statistically-based between group
2005           chronic stroke patients with muscle         comparisons were reported. Patients who first
USA            contraction in their affected wrist, but    received ETMS (n=8) gained 7 points on the FM
4 (RCT)        with no movement were randomized to         scale following treatment (8 weeks), but the
               begin with a home-based                     effects were lost when patients were switched
               electromyographic-triggered                 to the home exercise program and reassessed
               neuromuscular stimulation, twice daily      at 16 weeks. They lost 9 points on FM from
               for 35 minutes for 8 weeks or to a          baseline. There were no changes in ARA scores.
               home exercise program. Outcomes             Patients who first received home exercise
               included the Action Research Arm test,      (n=4) gained less than a point on the FM scale
               Fugl-Meyer and goniometry.                  at 8 weeks. When crossed over to the active
                                                           therapy group and reassessed at 16 weeks,
                                                           there was again a less than 1 point gain in FM
                                                           points. There were no changes in ARA scores.
Hara et al.     14 stroke patients (>1 yr post stroke)     Patients in the hybrid FES group demonstrated
2006            with spastic upper-extremity               significant improved active ROM in finger and
Japan           impairments were randomized to             wrist extension. There was a trend towards
4 (RCT)         receive 4 mos of power-assisted FES        greater improvement in MAS among patients in
                (40 min sessions, once or twice a week     the FES group.
                for 4 mos) plus traditional therapy
                (n=8) or traditional therapy alone
                (n=6). Outcome was assessed before
                and after training included active range
                of motion (ROM), Modified Ashworth
                Scale and 2 clinical tests.
Alon et al.     15 acute ischemic stroke patients were     At 12 weeks the FES group moved significantly
2007            randomly assigned to receive 12 weeks      more blocks compared with the control group
USA             of either FES along with task-specific     (42 vs. 26, p=0.049). Patients in the FES group
5 (RCT)         upper extremity rehabilitation (n=7) or    completed the J-T task faster (6.7 vs. 11.8,
                task-specific rehabilitation alone         p=0.049) and the mFM scores were higher
                (control) (n=8). Outcome measures          among patients in the FES group (49 vs. 40,
                were recorded at baseline, 4, 8, and 12    p=0.042).
                weeks and they included Box and Block

10. Upper Extremity Interventions                                                      pg. 115 of 150
               (B&B), Jebsen-Taylor light object lift (J-
               T) and modified Fugl-Meyer (mFM).
Bhatt et al.   20 chronic stroke patients were              from pre-test to post-test on the Box and Block
2007           randomly assigned to receive electrical      test and the Jebsen Taylor Test. There were no
USA            stimulation (ES), tracking training (TR),    significant differences between groups.
3 (RCT)        combination (CM) group. All groups
               received 10 1-hr training sessions for
               over 2-3 weeks. Assessments included
               the Jebson Taylor tests of manual
               dexterity, Box and Block test and a
               finger-tracking test.
Sullivan and   10 chronic stroke patients with arm          6 of the 10 participants improved significantly
Hedman 2007    hemiparesis received an individualized       on the ARAT after treatment. 5 of the 10
USA            home programme of neuromuscular and          patients improved significantly on the Stroke
No Score       sensory amplitude electrical                 Rehabilitation Assessment of Movement and 4
               stimulation. All patients participated in    of the 10 subjects attained a ≥10% change
               stimulation-assisted task-specific           from pre- to post-test on the MAS.
               exercises for 15 min 2-3 times daily, 7
               days/wk for 8 wks. Patients with
               sensory deficits received 15 mins, twice
               daily of additional sensory stimulation.
               Assessments included the Action
               Research Arm Test (ARAT), the Stroke
               Rehabilitation Assessment of Movement
               and the Modified Ashworth Assessment
               (MAS) of Spasticity.
Kowalczewski   19 subacute stroke subjects with severe      Improvements in the WMFT and CKS were
et al. 2007    upper-limb dysfunction were                  significantly greater in the high-intensity group
Canada         randomized to receive a program of           (effect size, .95) than the low-intensity group
6 (RCT)        either high (n=10) or low (n=9)              (effect size, 1.3). The differences in MAL and
               intensity FES stimulation combined with      FMA were not statistically significant.
               an exercise workstation with
               instrumented objects were used to
               perform specific motor tasks with their
               affected upper extremity. Subjects in
               the high-intensity FES-ET group
               received treatment for 1 hour a day on
               15 to 20 consecutive workdays, while
               subjects in the low-intensity FES-ET
               group received 15 minutes of sensory
               electric stimulation 4 days a week and
               on the fifth day they received 1 hour of
               FES-ET. The primary outcome measure
               included the Wolf Motor Function Test
               (WMFT). Secondary outcome measures
               included the Motor Activity Log (MAL),
               the upper-extremity portion of the Fugl-
               Meyer Assessment (FMA), and the
               combined kinematic score (CKS)
               derived from workstation
               measurements. Evaluations were
               performed before and after treatment
               and at 3 and 6-month follow-up.

10. Upper Extremity Interventions                                                         pg. 116 of 150
Alon et al.     26 severely impaired stroke patients         At the end of 12 weeks, the mean FM scores
2008            were randomized an average of 20 days        were significantly higher among subjects in the
USA             following stroke to receive either cyclic    FES group (24 vs. 14.2, p=0.05). There were
3 (RCT)         FES + task-specific exercise program or      non significant differences between groups on
                to an exercise-only group for 12 weeks.      the B&B test, although subjects in the FES
                All participants trained with task-          group were able to move more blocks (10.5 vs.
                specific exercises, 30 min, twice daily.     2.5). The JT task time did not differ significantly
                The FES group practiced the exercises        between groups. Eight (FES) compared with
                combined with FES that enabled               three (control) patients regained the ability to
                opening and closing of the paretic hand      transfer five or more blocks (P = 0.051), and
                and continued with FES without               six (FES) compared with two (control)
                exercises for up to 90 mins of additional    completed the J-T task in 30 sec or less after
                time twice a day. Outcomes assessed          12 wks of training (P = 0.09).
                included modified Fugl-Meyer (FM), Box
                & Blocks test (B&B) and the Jebsen-
                Taylor light object lift (JT). Outcomes
                were assessed at baseline and at 4, 8,
                and 12 wks.
Hara et al.     20 chronic stroke patients were              The FES group displayed significantly greater
2008            randomized to a home-based 5-month           improvements in RMS, active ROM, MAS and
Japan           program of FES or physical therapy.          functional hand tests.
5 (RCT)         The FES group used a power-assisted
                FES device to induce greater muscle
                contraction by electrical stimulation in
                proportion to the integrated
                electromyography (EMG) signal picked
                up on surface electrodes. Target
                muscles were the extensor carpi radialis
                longus (ECRL) and extensor carpi
                radialis brevis (ECRB), extensor
                digitorum communis (EDC), extensor
                indicis proprius (EIP), and deltoid (Del).
                Patients underwent 30 approximately
                60 min FES sessions at home about 6
                days/week. Root mean square (RMS) of
                ECRL, EDC and Del maximum voluntary
                EMGs, active range of motion (ROM) of
                wrist and finger extension and shoulder
                flexion, modified Ashworth scale (MAS),
                and clinical tests were investigated
                before and after FES training.
De Kroon &      22 chronic stroke patients were              Both groups improved on the Action Research
Ijzerman 2008   randomly assigned to receive either          Arm test. The group receiving cyclic stimulation
Denmark         cyclic (n=11) or EMG-triggered               improved by 2.3 points, and the group
7 (RCT)         electrical stimulation (n=11) of the         receiving EMG-triggered stimulation improved
                wrist and finger extensor muscles for a      by 4.2 points. The difference in functional gain
                six-week period. The primary outcome         was not statistically significant. Differences in
                measure was the Action Research Arm          gain on the secondary outcome measures were
                test (ARAT). Grip strength, Fugl-Meyer       also not significant.
                Motor Assessment and Motricity Index
                were secondary outcome measures.
                Assessments were made at the start of
                the treatment and after 4, 6 and 12

10. Upper Extremity Interventions                                                          pg. 117 of 150
Chae et al.      26 chronic stroke survivors were            There were no significant differences between
2009             randomly assigned to receive                groups over the testing periods, although
USA              percutaneous intramuscular ES for hand      subjects in both groups improved on all
8 (RCT)          opening (n = 13) or percutaneous ES         measures over time.
                 for sensory stimulation only (n = 13).
                 The intramuscular ES group received
                 cyclic, electromyography (EMG)-
                 triggered or EMG-controlled ES
                 depending on baseline motor status. All
                 participants received 1 hour of
                 stimulation per day for 6 weeks. After
                 completion of ES, participants received
                 18 hours of task-specific functional
                 training. The primary outcome measure
                 was the Fugl-Meyer Motor Assessment.
                 Secondary measures included the Arm
                 Motor Ability Test and delay and
                 termination of EMG activity. Outcomes
                 were assessed at baseline, at the end of
                 ES, at the end of functional training,
                 and at 1, 3, and 6 months follow-up.
Chan et al.      20 subjects, 6 months post stroke were      After 15 training sessions, the FES group had
2009             randomized to receive 15 sessions of        achieved significantly greater improvement in
Hong Kong        either FES (using a self-triggering         FMA (+7.7 vs. +2.1 points, P = .039), FTHUE
7 (RCT)          mechanism) + bilateral tasks (20            (+1.3 vs. 0.3, P = .001), and active range of
                 minutes), + 10 min of stretching            motion of wrist extension (+17 vs. 3.5 degrees,
                 exercises and occupational therapy          P = .020), when compared with the control
                 treatment (60 minutes), or the same         group.
                 duration of stretching and occupational
                 therapy training + placebo FES during
                 the bilateral tasks. The outcome
                 measures, assessed before and after
                 intervention, included Functional Test
                 for the Hemiplegic Upper Extremity
                 (FTHUE), Fugl-Meyer Assessment
                 (FMA), grip power, forward reaching
                 distance, active range of motion of wrist
                 extension, FIM and Modified Ashworth
Mangold et al.   23 subjects with stroke onset of 2-18       There were no statistically significant
2009             weeks admitted for inpatient                differences between groups on any of the
Switzerland      rehabilitation were randomly assigned       outcome measures assessed. The EBI subscore
5 (RCT)          to a 4-week intervention program of         and CMSA arm score improved significantly in
                 either FES or conventional therapy.         both groups. The CMSA hand function improved
                 Subjects in both groups received 3-5        significantly in the FES group. Resistance to
                 occupational therapy sessions per           passive movement of finger and wrist flexors
                 week, each lasting 45 min each. FES         increased significantly in the FES group.
                 training replaced 12 conventional           Shoulder pain did not change significantly.
                 training sessions in the intervention       None of the outcome measures demonstrated
                 group. Outcomes assessed before and         significant differences in gains between the
                 after treatment included the ADL            groups.
                 subscore of the Extended Barthel Index
                 (EBI), the Chedoke McMaster Stroke
                 Assessment (CMSA) measured hand

10. Upper Extremity Interventions                                                        pg. 118 of 150
                and arm function and shoulder pain and
                the Modified Ashworth Scale (MAS)
Thraher et al.  21 subjects with onset of stroke 2-7       The FES group improved significantly more
2009            weeks admitted for inpatient               than the control group in terms of object
Canada          rehabilitation were randomized to          manipulation, palmar grip torque, pinch grip
5 (RCT)         receive a 12-16 week program of either     pulling force, Barthel Index, Upper Extremity
                FES plus conventional occupational and     Fugl-Meyer scores, and Upper Extremity
                physiotherapy (FES group) or only          Chedoke-McMaster Stages of Motor Recovery.
                conventional therapy (control group) for   There were no significant differences in FIM
                45 min x 5 days a week. FES was            scores.
                applied to proximal and then distal
                muscle groups during specific motor
                tasks. At baseline and at the end of
                treatment, grasping function was
                assessed using the Rehabilitation
                Engineering Laboratory Hand Function
                Test, Barthel Index, Fugl-Meyer (FM)
                scores, and Upper Extremity Chedoke-
                McMaster Stages of Motor Recovery.
Hsu et al. 2010 66 acute patients receiving inpatient      At 4 and 12 weeks, both NMES groups showed
Taiwan          rehabilitation were randomized to 3        significant improvement on FM Assessment and
6 (RCT)         groups: high NMES, low NMES, or            ARAT scales compared with the control group.
                control. The low-NMES group received       The results for the high and low NMES groups
                30 minutes of stimulation per day, and     were similar.
                the high-NMES group received 60
                minutes per day, for 4 weeks. He
                control group received no additional
                treatment. Outcomes included the Fugl-
                Meyer Motor Assessment Scale (FM),
                Action Research Arm Test (ARAT), and
                Motor Activity Log and were assessed at
                baseline, 4 weeks, and 12 weeks post
Weber et al.    23 chronic stroke patients received        There were no statistically significant
2010            botulinum toxin-A injections + a home      differences between groups on any of the
USA             exercise program (60 min/day x 12          outcomes assessed at any of the assessment
7(RCT)          weeks) consisting of task-specific         points.
                practice. Participants were then
                randomly assigned to receive additional
                treatment with FES during practice time
                or no FES. Outcomes assessed at
                baseline, 6 and 12 weeks included the
                otor Activity Log (MAL)-Observation,
                Action Research Arm Test (ARAT) and

RCTs evaluating FES were categorized                  occurred greater than 6 months prior
according to chronicity of stroke.                    to inclusion in the study. The results
Patients were considered to be acute if               are presented in Tables 10.52 and
they had suffered a stroke within 6                   10.53.
months and chronic if their stroke had

10. Upper Extremity Interventions                                                      pg. 119 of 150
Table 10.52 Summary Table for FES for the Hemiparetic Upper Extremity in Acute or
SubacuteStroke (< 6 months)
     Author           N              Intervention                         Main Outcome(s)
   PEDro Score                                                                  Result
Powell et al. 1999   60     Standard rehabilitation and        Grasp and grip scores of Action Research
7 (RCT)                       electrical stimulation vs.                    Arm test (+)
                               standard rehabilitation
Hsu et al. 2010      66    High vs. low intensity FES vs.        Fugl-Meyer (hi and low FES +)
6                                   no treatment          Action Research Arm test (hi and low FES +)
                                                                     Motor Activity Log (-)
Kowalczewski et      19      High vs. low intensity FES           Wolf Motor Function Test (+)
al. 2007                                                             Motor Activity Log (-)
6 (RCT)                                                                 Fugl-Meyer (-)
Popovic et al.       41         Early vs. delayed FES       Upper Extremity Function test (+ acute)
2004                                                                 Drawing test (+acute)
6 (RCT)
Popovic et al.       28     Functional electrical therapy         Upper Extremity Function test (+)
2003                          + therapy vs. standard                      Drawing test (+)
6 (RCT)                               therapy
Chae et al. 1998     46      Neuromuscular stimulation             Fugl-Meyer: post-treatment (+)
6 (RCT)                    and routine rehabilitation vs.              12 weeks follow-up (-)
                           sham stimulation and routine
Mangold et al.       23    FES vs. conventional therapy       ADL subscore of Extended Barthel Index (-)
2009                                                           Chedoke McMaster Stroke Assessment (-)
5 (RCT)
Thrasher et al.      21      FES+ conventional therapy        Rehabilitation Engineering Laboratory Hand
2009                          vs. conventional therapy                     Function Test (+)
5 (RCT)
Alon et al. 2007     15      FES + task specific training                Box and Block (+)
5 (RCT)                     vs. only task specific training       Jebsen-Taylor light object lift (+)
                                                              Modified Fugl-Meyer (+ only at 12 weeks)
Francisco et al.      9       Electrical stimulation and                   Fugl-Meyer (+)
1998                            standard therapy vs.               Upper extremity FIM scores (+)
5 (RCT)                         Conventional Therapy
Faghri & Rodgers     26      Conventional therapy and                   Range of motion (+)
1997                        FES vs. conventional therapy              Shoulder muscle tone (+)
4 (RCT)
Heckermann et al.    28        Functional electrical therapy             Range of motion (+)
1997                              + therapy vs. standard
4 (RCT)                                    therapy
Faghri et al. 1994     26        Conventional therapy and                   Arm tone (+)
4 (RCT)                       FES vs. Conventional Therapy                 EMG activity (+)
Bowman et al.          30         Conventional therapy +                 Range of motion (+)
1979                                positional feedback
3 (RCT)                           stimulation therapy vs.
                                   conventional Therapy
- Indicates non-statistically significant differences between treatment groups
+ Indicates statistically significant differences between treatment groups

10. Upper Extremity Interventions                                                        pg. 120 of 150
Among the studies evaluating ES in                       acutely following stroke experienced
the acute stage of stroke, most                          improved recovery compared with
assessed the same treatment                              those who received FES a year
comparison, physical therapy plus FES                    following stroke.
(or sham FES) vs. physical therapy
alone. The results most of the studies                    Conclusions Regarding the Efficacy of
indicated that FES was an associated                      FES Therapy in Acute Stroke
with improvements in motor function,
ADL and dexterity. One study,                             There is strong (Level 1a) evidence that
Popovic et al. (2004) examined early                      FES treatment improves upper
vs. delayed treatment with FES and                        extremity function in acute stroke.
found that subjects who received FES

Table 10.53 Summary Table for FES in the Hemiparetic Upper Extremity in Chronic
    Author      N              Intervention                             Main Outcome(s)
 PEDro Score                                                                 Result
Chae et al.    26        Percutaneous electrical                        Fugl-Meyer (-)
2009                 stimulation (motor vs. sensory)
8 (RCT)
Chan et al.    20    Bilateral arm training + FES vs.                    Fugl-Meyer (+)
2009                  bilateral arm training + sham         Functional test for the Hemilplegic Upper
7 (RCT)                            FES                                    Extremity (+)
Weber et al.   23       FES + BT-A + home based                       Motor Activity Log (-)
2010                  exercise program vs. BT-A +                 Action Research Arm Test (-)
7                    home-based exercise program
De Kroon &     22     EMG-triggered vs. cyclic FES                Action Research Arm test (-)
7 (RCT)
Kimberly et    16            NMES vs. sham                            Box & Block test (+)
al. 2004                                                             Motor Activity Log (+)
7 (RCT)                                                       Jebsen Taylor Hand Function test (+)
Ring &         22     Neuroprothetic FES vs. control             Modified Ashworth Scores (+)
Rosenthal                                                             Box & Block test (+)
6 (RCT)                                                       Jebsen Taylor Hand Function test (+)
De Kroon et    30       Two different forms of ES                  Arm Research Arm test (-)
al. 2004                                                               Motricity Index (-)
6 (RCT)                                                                Ashworth Scale (-)
Conforto et     8     Electrical stimulation vs. Sham              Pinch muscle strength (+)
al. 2002                         Stimulation
6 (RCT)
Wu et al.       9    3 electrical stimulation sessions        Jebsen Taylor Hand Function test (+)
2006                         vs. no treatment
6 (RCT)
Cauraugh       26       0 sec stimulation vs. 5 sec                   Box and Block Test
and Kim                    stimulation vs.10 sec                   (+ for both stim groups)
2003                            stimulation                             Reaction time
5 (RCT)                                                            (+ for both stim groups)
                                                               Sustained wrist/finger contraction
                                                                   (+ for both stim groups)

10. Upper Extremity Interventions                                                      pg. 121 of 150
Hara et al.      20     Power-assisted FES vs. control                         ROM (+)
2008                                                                 Modified Ashworth Scale (+)
5 (RCT)
Gabr et al.      12      Electromyography-triggered                        Fugl- Meyer (+)
2005                    stimulation vs. home exercise                Action Research Arm test (-)
4 (RCT)
Hara et al.      14     Power-assisted FES vs. control              Modified Ashworth Scale (-)
2006                                                                    Range of Motion (+)
4 (RCT)                                                                 Root mean square (-)
Cauraugh et      11      Passive range of motion and                   Box and Block test (+)
al. 2000                   stretching exercises and                  Motor Assessment scale (-)
4 (RCT)                electrical stimulation vs. passive          Fugl-Meyer upper extremity (-)
                        range of motion and stretching
King             21        NMES vs. passive stretch                      Tone reduction (+)
4 (RCT)
Bhatt et al.     20        Electrical stimulation vs.         Jebson Taylor tests (- of manual dexterity)
2007                         tracking training vs.                       Box & Block test (-)
3 (RCT)                       combination group                        Finger tracking test (-)

- Indicates non-statistically significant differences between treatment groups
+ Indicates statistically significant differences between treatment groups

The treatment comparisons among                              There is strong (Level 1a) evidence that
studies evaluating electrical                                FES treatment improves upper
stimulation in the chronic stage of                          extremity function in chronic stroke.
stroke were more heterogeneous.
However, the weight of evidence                              There is moderate (Level 1b) evidence
suggested that there was a benefit of                        that EMG-triggered FES is not superior
treatment.                                                   to cyclic FES

 Conclusions Regarding the Efficacy of                        Functional Electrical Stimulation
 FES Therapy in Chronic Stroke                                therapy improves hemiparetic upper
                                                              extremity function.

10.9 Medications Used in Motor
                                                            the upper extremity has also been
Medications used following stroke to                        identified. These drugs include
augment the rehabilitation process                          stimulants (amphetamines and,
have manily been examined for their                         methylphenidate), levodopa and anti-
potential benefit in terms of global                        depressants (citapopram and
recovery and depression. The results                        reboxetine).
from these trials have been published
in other chapters (Mobility,                                10.9.1 Stimulants
Depression, Aphasia). However, a
small group of studies that evaluated                       Two RCTs have examined the effects
the efficacy of drugs for its effect on                     of amphetamine and methylphenidate
                                                            on motor recovery in the upper

10. Upper Extremity Interventions                                                         pg. 122 of 150
Table 10.54 The Use of Stimulants in Motor Recovery
    Author, Year                    Methods                                    Outcomes
Platz et al. 2005   31 patients with mild arm paresis, approx   A significant effect of active drug was
Germany             5 weeks post stroke, were randomized to     observed immediately following
9 (RCT)             receive 10 mg d-amphetamine 2x per          treatment for total TEMPA scores, but
                    week x 3 weeks + arm training or placebo    was lost at the end of one-year. The
                    + arm training. Primary outcome was         study was terminated prematurely
                    TEMPA scores assessed post intervention     before reaching target recruitment of
                    and at one-year.                            60 partially because of lack of efficacy.
Tardy et al. 2006    8 male patients with stroke onset of 35    There was a significant treatment
France              days or less with pure motor hemiparesis    effect of methlyphenidate with respect
9 (RCT)             were randomized to receive a single dose    to the finger tapping test. Patients
                    of 20 mg of methylphenidate and placebo,    increased the number of taps
                    7 days later, in random order. The effect   completed in 10 s by almost 5,
                    of drug on motor performance was            compared with the untreated
                    measured using hand grip strength,          condition. There was no treatment
                    number of taps completed in a finger        effect for either hand grip strength or
                    tapping test and speed during a target      the target pursuit task.
                    pursuit task.

A single trial examined the use of                    examined the effect of
amphetamines on improving                             methylphenidate (Tardy et al. 2006),
arm/finger function following stroke                  the same class of drug as
(Platz et al. 2005). Amphetamines                     amphetamines, which has the
have shown promise in recovery                        advantage that it does not produce
following stroke as they have the                     the same side effect profile as
potential to accelerate motor recovery                amphetamines (insomnia, lack or
following motor cortex lesions in the                 appetite).
rat model (Feeney et al. 1982),
especially when combined with task-
specific training. A single RCT also

 Conclusions Regarding Stimulants                     barrier, is converted to dopamine
                                                      (dopamine cannot cross the blood-
 There is conflicting (Level 4) evidence              brain barrier). Levodopa is used as a
 that stimulants can improve upper                    prodrug to increase dopamine levels,
 extremity impairment following stroke.               most commonly in the treatment of
                                                      Parkinson’s disease. Levodopa may
                                                      also improve arousal and motor
10.9.2 Levodopa
                                                      initiation following stroke (Horowitz
Levodopa is a dopamine precursor
which, once it crosses the blood-brain

10. Upper Extremity Interventions                                                       pg. 123 of 150
Table 10.55 Levodopa in Motor Recovery
   Author, Year                      Methods                                     Outcomes
Restermeyer et al. 10 patients > 6 months stroke onset            There were no statistically significant
2007               participated. On two different occasions,      differences among the groups at any
Germany            patients were randomized to receive either     of the testing times.
9 (RCT)            100 mg levodopa or placebo. Immediately
                   afterwards, they participated in a 1-hour
                   PT session aimed at an improvement of
                   dexterity. Motor functions tests included:
                   the Nine-Hole-Peg Test, grip strength
                   (dynamometer) and Action Research Arm
                   Test (ARAT). Outcomes were assessed
                   before drug intake, 45 minutes after drug
                   ingestion and after the physiotherapy.
Rosser et al. 2008 18 patients with chronic motor dysfunction     Subjects in the levodopa condition
Germany            because of stroke were randomized to           performed significantly better on the
5 (RCT)            receive 3 doses of levodopa (100mg of          performance test compared with the
                   levodopa plus 25mg of carbidopa) and           control condition.
                   placebo before 1 session of procedural
                   motor learning in a crossover trial. The
                   main outcome measure was a keyboard
                   tapping test performed with the paretic

 Conclusions Regarding Levodopa                        recovery through changes in
                                                       neurotransmission. Two small RCTs
 There is conflicting (Level 4) evidence               have investigated the effect of 2 types
 that levodopa can improve upper                       of drugs-- selective serotonin-
 extremity motor function following                    reuptake inhibitors (SSRI) and
 stroke.                                               noradrenaline reuptake inhibitor
                                                       (NARI). Both of these trials were
                                                       conducted by the same author and
10.9.3 Antidepressants
                                                       examined the use of a single dose of
                                                       the drug over a window of several
Beyond their ability to improve mood                   hours for an off-label purpose in a
disturbances following stroke,                         small group of chronic stroke patients.
antidepressants can be used to
enhance upper extremity motor

Table 10.56 Antidepressants in Stroke Recovery
    Author, Year                     Methods                                   Outcome
    PEDro Score
Zittel et al. 2007   10 chronic hemiparetic subjects received    Compared with placebo, reboxetine
Germany              a single 6 mg dose of the noradrenaline     ingestion was followed by an
6 (RCT)              reuptake inhibitor reboxetine or placebo.   increase of tapping speed and grip
                     Then the patients participated in one       strength in the paretic but not in the
                     hour of physiotherapy focused on            unaffected hand.

10. Upper Extremity Interventions                                                         pg. 124 of 150
                     function of the paretic hand. Three
                     different motor assessments (tapping
                     speed, grip strength, dexterity
                     evaluation) were performed before drug
                     intake, 1.5 hours later and after the
                     physiotherapy session.
Zittel et al. 2008   8 chronic stroke patients (>6 months         Compared with placebo, citalopram
Germany              onset) participated in a single-dose         intake was associated with
8 (RCT)              crossover experiment. Subjects received      significant improvement in
                     either 40 mg oral citalopram or placebo      performance of the nine-hole peg
                     followed by a separation of at least 2       test for the paretic hand but not for
                     weeks. A single session of PT was given      the unaffected hand. Hand grip-
                     2.5 hrs after drug/placebo. Motor            strength remained unchanged.
                     function was assessed by nine-hole peg
                     test, and measurements of hand grip-
                     strength before drug intake, 2 hours
                     after drug intake, and after 1 hour of PT.

 Conclusions Regarding                                  stroke patients suffered from hand
 Antidepressants                                        edema not associated with shoulder-
                                                        hand syndrome. More recently, Post et
 There is strong (Level 1a) evidence that               al. (2003) reported that based on
 a single dose of either a SSRI or NARI                 volumetric assessments, 33% of 96
 can enhance short-term manual                          stroke patients had hand edema,
 dexterity in the affected hand following               compared to 50% of patients assessed
                                                        through clinical evaluation.
                                                        Volumetric assessments of the hand
                                                        appear to provide the best estimation;
10.10 Treatment of Hand Edema                           while the reliability of clinical
Hand edema following stroke with                        evaluation through visual inspection,
hemiparesis is associated with pain                     is poor. A change of 12 mL or more is
and stiffness, which can lead to a                      considered clinically significant (Post
decrease in active motion and disuse.                   et al. 2003).
Hand edema may be an isolated
problem or occur as a symptom of                        Using data from the same patient
shoulder-hand syndrome. The etiology                    group as Post et al. (2003),
of the development of hand edema is                     Boomkamp-Koppen et al. (2005)
unclear. The most widely accepted                       reported a significant correlation
explanation is of increased venous                      between the presence of hand edema
congestion related to prolonged                         and measures of hand function
dependency and loss of muscle                           (measured by the Frenchay arm test).
pumping function in the paretic limb.                   Patients without hand edema were
(Leibovitz et al. 2007).                                more likely to have good hand
                                                        function. Significant predictor of hand
Diagnosis is difficult and depends, in                  function following stroke included the
part, on the method of assessment.                      degree of motor impairment,
Estimates of the incidence of hand                      hypertonia, tactile inattention and
edema vary widely. Tepperman et al.                     edema.
(1984) reported that 83% of 85 acute

10. Upper Extremity Interventions                                                         pg. 125 of 150
Leibovitz et al. (2007) compared the                  Three different treatment approaches
circumference of the hand in three                    to aid in the reduction of hand edema
places (mid-finger, hand and wrist)                   following stroke have been studied,
among subjects post stroke (m=188)                    including passive motion exercises,
and non-paretic institutionalized                     neuromuscular stimulation and
controls (n=70). Hand edema was                       intermittent pneumatic compression.
detected in 37% of post stroke                        The results are presented in table
subjects compared with only 2% of                     10.57.
control subjects.

Table 10.57 Treatment of Hand Edema
  Author, Year                      Methods                                        Outcome
 PEDro Score
Giudice 1990     16 patients with hand edema of greater than 4     11 patients in the study had suffered
USA              months duration received two treatments on        from a stroke. The treatment effect
No Score         two consecutive days. The effect of a 30 minute   sizes for all 3 measures were large
                 treatment of continuous passive motion (CPM)      (>0.5) indicating that CPM was more
                 exercises plus limb elevation was compared to     effective than elevation alone.
                 limb elevation alone. Hand volume, finger
                 circumference and finger stiffness were
Faghri 1997      8 patients with visible hand edema following      No inferential statistics were reported.
USA              stroke received neuromuscular stimulation         The reduction in mean hand volume
No Score         (NMS)-induced contraction of the paralyzed        (mL) of NMS and limb elevation were:
                 muscles to produce an active muscle pump for      –13.4 and + 1.9, respectively. Although
                 removing excess fluid and compare its effect      NMS was more effective for reduction of
                 with elevation of the upper extremity. The        hand edema than limb elevation alone,
                 effects of 30 minutes of NMS (stimulation         hand edema returned to pre-treatment
                 frequency of 35 Hz) of the finger and wrist       levels within 24 hours.
                 flexors and extensors were compared with the
                 effects of 30 minutes of limb elevation alone.
                 Each patient received both treatments, one on
                 each of 2 consecutive days. Measures of hand
                 and arm volume and upper and lower arm girth
                 were taken before and after each treatment.
Roper et al.     37 patients with first ever stroke and edema in   There were no statistically significant
1999             their affected hand were randomized to receive    between group differences reported.
UK               intermittent pneumatic compression (INC)+
5 (RCT)          standard physiotherapy 2 hours a day for one
                 month or standard physiotherapy. Hand volume
                 (measured by water displacement) and
                 Motricity Index scores were assessed at the end
                 of the treatment period.
                                                        compression does not reduce hand
 Conclusion Regarding Intermittent                      edema following stroke.
 Pneumatic Compression for Hand
 Edema                                                  There is limited (Level 2) evidence that
                                                        both neuromuscular nerve stimulation
 There is moderate (Level 1b) evidence                  and continuous passive motion help to
 that intermittent pneumatic                            reduce hand edema compared to limb

10. Upper Extremity Interventions                                                        pg. 126 of 150
 Continuous passive motion and
 electrical stimulation might be
 effective treatments for hand edema,
 while intermittent pneumatic
 compression is not.

10. Upper Extremity Interventions                   pg. 127 of 150
10.11 Summary
1. There is consensus (Level 3) opinion        7. There is conflicting (Level 4) evidence
  that in severely impaired upper                that specialized programs improve
  extremities (less than stage 4) the            reaching.
  focus of treatment should be on
  palliation and compensation. For             8. There is strong (Level 1a) evidence
  those upper extremities with signs of          that electrical somatosensory
  some recovery (stage 4 or better) there        stimulation improves hand function.
  is consensus (Level 3) opinion that
  attempts to restore function through         9. There is strong (level 1a) evidence
  therapy should be made.                        that mental practice may improve
                                                 upper-extremity motor and ADL
2. There is strong (Level 1a) evidence           performance following stroke.
  that neurodevelopmental techniques
  such as Bobath are not superior to           10. There is strong (Level 1a) evidence
  other therapeutic approaches. There is         that hand splinting does not improve
  moderate (level 1b) evidence that              motor function or reduce contracture
  indicates compared to Bobath, motor            formation.
  relearning programs may result in
  improved short-term motor                    11. There is conflicting (Level 4)
  functioning and shorter lengths of             evidence of benefit of CIMT in
  hospital stay.                                 comparison to traditional therapies in
                                                 the acute stage of stroke. There is
3. There is conflicting (Level 4) evidence       strong (Level 1a) evidence of benefit
  that enhanced therapies improve                of CIMT and mCIMT in comparison to
  short-term upper extremity function.           traditional therapies in the chronic
  There is evidence that results may not         stage of stroke. Benefits appear to be
  be long-lasting. There is moderate             confined to stroke patients with some
  (Level 1b) evidence that a program of          active wrist and hand movements,
  daily stretch regimens does not                particularly those with sensory loss
  prevent the development of                     and neglect. There is moderate (Level
  contractures.                                  1b) evidence that any intensity of CIMT
                                                 will provide benefit.
4. There is conflicting (Level 4) evidence
  that repetitive task specific training       12. There is conflicting (Level 4)
  techniques improve measures of                 evidence that mirror therapy improves
  upper extremity function compared to           motor function following stroke and
  traditional techniques.                        moderate (Level 1b) evidence that it
                                                 does not reduce spasticity.
5. There is conflicting (Level 4) evidence
  that sensorimotor training delivered         13. There is strong (Level 1a) evidence
  by a therapist improves upper                  that extrinsic feedback helps to
  extremity function, compared to                improve motor learning following
  traditional techniques.                        stroke.

6. There is conflicting (Level 4) evidence     14. There is strong (Level 1a) evidence
  that bilateral arm training is superior        that sensorimotor training with robotic
  to unilateral training.                        devices improves upper extremity
                                                 functional outcomes, and motor
                                                 outcomes of the shoulder and elbow.

10. Upper Extremity Interventions                                          pg. 128 of 150
  There is strong (Level 1a) evidence
  that robotic devices do not improve         23. There is limited (Level 2) evidence
  motor outcomes of the wrist and hand.         that shock wave therapy can reduce
                                                tone in the upper extremity.
15. There is strong (Level 1a) evidence
  that virtual reality treatment can          24. There is moderate (Level 1b)
  improve locomotor function in the             evidence that tolperisone can reduce
  chronic stages of stroke.                     spasticity following stroke.

16. There is conflicting (Level 4)            25. There is strong (Level 1a) evidence
  evidence that hand splinting reduces          that EMG/Biofeedback therapy is not
  spasticity.                                   superior to other forms of treatment.

17. There is moderate (Level 1a)              26. There is conflicting (Level 4)
  evidence that a nurse-led stretching          evidence that treatment with TENS in
  program can help to increase range of         the upper extremity improves a variety
  motion in the upper extremity and             of outcomes, including motor
  reduce pain in the chronic stage of           recovery, spasticity and ADLs.
                                              27. There is strong (Level 1a) evidence
18. There is strong (Level Ia) that             that FES treatment improves upper
  treatment with BTX alone or in                extremity function in both acute and
  combination with therapy significantly        chronic stroke.
  decreases spasticity in the upper
  extremity in stroke survivors.              28. There is moderate (Level 1b)
  However, it is not clear that the             evidence that EMG-triggered FES is
  improvements are sustained, nor is            not superior to cyclic FES.
  there strong evidence that they are
  associated with improved function and       29. There is conflicting (Level 4)
  quality of life.                              evidence that stimulants can improve
                                                upper extremity impairment following
19. There is moderate (Level 1b)                stroke.
  evidence that electrical stimulation
  combined with botulinum toxin               30. There is conflicting (Level 4)
  injection is associated with reductions       evidence that levodopa can improve
  in muscle tone.                               upper extremity motor function
                                                following stroke.
20. There is strong (Level 1a) evidence
  that a combination of physiotherapy         31. There is strong (Level 1a) evidence
  and botulinum toxin injection is              that a single dose of either a SSRI or
  associated with improved upper                NARI can enhance short-term manual
  extremity function.                           dexterity in the affected hand
                                                following stroke.
21. There is limited (Level 2) evidence
  that treatment with ethyl alcohol           32. There is moderate (Level 1b)
  improves elbow and finger PROM and            evidence that intermittent pneumatic
  can decrease spasticity in the upper          compression does not reduce hand
  extremity in stroke survivors.                edema following stroke. There is
                                                limited (Level 2) evidence that both
22. There is strong (Level 1a) evidence         neuromuscular nerve stimulation and
  that physical therapy does not reduce         continuous passive motion help to
  spasticity in the upper extremity.

10. Upper Extremity Interventions                                         pg. 129 of 150
  reduce hand edema compared to limb

10. Upper Extremity Interventions                   pg. 130 of 150
Alberts JL, Butler AJ and Wolf SL.The effects of        in upper limb spasticity in patients with
   constraint-induced therapy on precision grip:        stroke. Eur J Neurol. 2001;8(6):559-65.
   A preliminary study. Neurorehabilitation and
   Neural Repair 2004;18:G250-258.                    Bakheit AM, Fedorova NV, Skoromets AA,
                                                        Timerbaeva SL, Bhakta BB, Coxon L. The
Alon G, Levitt AF, McCarthy PA. Functional              beneficial antispasticity effect of botulinum
   Electrical Stimulation Enhancement of Upper          toxin type A is maintained after repeated
   Extremity Functional Recovery During Stroke          treatment cycles. J Neurol Neurosurg
   Rehabilitation: A Pilot Study. Neurorehabil          Psychiatry 2004;75:1558-1561.
   Neural Repair 2007;21:207-215.
                                                      Barker RN, Brauer SG, Carson RG. Training of
Alon G, Levitt AF, McCarthy PA. Functional              reaching in stroke survivors with severe and
   electrical stimulation (FES) may modify the          chronic upper limb paresis using a novel
   poor prognosis of stroke survivors with              nonrobotic device: a randomized clinical trial.
   severe motor loss of the upper extremity: a          Stroke 2008;39:1800-1807
   preliminary study. Am J Phys Med Rehabil
   2008;87:627-636.                                   Barreca S. et al. Management of the Post
                                                        Stroke Hemiplegic Arm and Hand: Treatment
                                                        Recommendations of the 2001 Consensus
Altschuler EL, Wisdom SB, Stone L, Foster C,            Panel. Heart and Stroke Foundation of
  Galasko D, Llewellyn DM, Ramachandran VS.             Ontario, 2001.
  Rehabilitation of hemiparesis after stroke with
  a mirror. Lancet 1999;353:2035-2036.                Barreca S, Wolf S L, Fasoli S, Bohannon R.
                                                        Treatment interventions for the paretic upper
Armagan O, Tascioglu F, Oner C.                         limb of stroke survivors: a critical review.
  Electromyographic biofeedback in the                  Neurorehabil Neural Repair 2003,17(4):220-
  treatment of the hemiplegic hand: a placebo-          226.
  controlled study. Am J Phys Med Rehabil
  2003;82:856-861.                                    Basmajian JV, Gowland C, Brandstater ME,
                                                        Swanson L, Trotter J. EMG feedback
Atteya AA. Effects of modified constraint               treatment of upper limb in hemiplegic stroke
   induced therapy on upper limb function in            patients: a pilot study. Arch Phys Med
   subacute stroke patients. Neurosciences              Rehabil 1982;63(12):613-616.
                                                      Basmajian JV, Gowland CA, Finlayson MA, Hall
Azab M, Al-Jarrah M, Nazzal M, Maayah M,                AL, Swanson LR, Stratford PW, Trotter JE,
  Sammour MA, Jamous M. Effectiveness of                Brandstater ME. Stroke treatment:
  constraint-induced movement therapy                   comparison of integrated behavioral-physical
  (CIMT) as home-based therapy on Barthel               therapy vs traditional physical therapy
  Index in patients with chronic stroke. Top            programs. Arch Phys Med Rehabil 1987;68(5
  Stroke Rehabil 2009;16:207-211.                       Pt 1):267-272.

Bakheit AM, Thilmann AF, Ward AB, Poewe W,            Bate PJ, Matyas TA. Negative transfer of
  Wissel J, Muller J, Benecke R, Collin C, Muller       training following brief practice of elbow
  F, Ward CD, Neumann C. A randomized,                  tracking movements with electromyographic
  double-blind, placebo-controlled, dose-               feedback from spastic antagonists. Arch Phys
  ranging study to compare the efficacy and             Med Rehabil 1992;73:1050-1058.
  safety of three doses of botulinum toxin type
  A (Dysport) with placebo in upper limb              Bell KR, Williams F. Use of botulinum toxin type
  spasticity after stroke. Stroke. 2000                 A and type B for spasticity in upper and
  Oct;31(10):2402-6.                                    lower limbs. Phys Med Rehabil Clin N Am
Bakheit AM, Pittock S, Moore AP, Wurker M,
  Otto S, Erbguth F, Coxon L. A randomized,           Bensmail D, Robertson JV, Fermanian C, Roby-
  double-blind, placebo-controlled study of the         Brami A. Botulinum toxin to treat upper-limb
  efficacy and safety of botulinum toxin type A         spasticity in hemiparetic patients: analysis of
                                                        function and kinematics of reaching

10. Upper Extremity Interventions                                                      pg. 131 of 150
  movements. Neurorehabil Neural Repair               Broeren J, Classon L, Goude D, Rydmark M,
  2010;24:273-281.                                      Sunnerhagen K. Virtual reality in an activity
                                                        centre for community-dwelling persons with
Bhakta BB, Cozens JA, Bamford JM,                       stroke. Cerebrovasc Dis 2008;26:289-296.
  Chamberlain MA. Use of botulinum toxin in
  stroke patients with severe upper limb              Bonaiuti D, Rebasti L, Sioli P. The constraint
  spasticity. J Neurol Neurosurg Psychiatry             induced movement therapy: a systematic
  1996;61:30-35.                                        review of randomised controlled trials on the
                                                        adult stroke patients. Eura Medicophys
Bhakta BB, Cozens JA, Chamberlain MA et al.             2007;43:139-146
  Impact of botulinum toxin type A on
  disability and carer burden due to arm              Bonifer NM, Anderson KM, Arciniegas DB.
  spasticity after stroke: a randomised double          Constraint-induced movement therapy after
  blind placebo controlled trial. J Neurol              stroke: efficacy for patients with minimal
  Neurosurg Psychiatry. 2000;69:217-221.                upper-extremity motor ability. Arch Phys
                                                        Med Rehabil 2005;86:1867-1873.
Bakheit AM, Fedorova NV, Skoromets AA,
  Timerbaeva SL, Bhakta BB, Coxon L. The              Boomkamp-Koppen HG, Visser-Meily JM, Post
  beneficial antispasticity effect of botulinum         MW, Prevo AJ. Poststroke hand swelling and
  toxin type A is maintained after repeated             oedema: prevalence and relationship with
  treatment cycles. J Neurol Neurosurg                  impairment and disability. Clin Rehabil 2005;
  Psychiatry 2004;75:1558-1561.                         19(5):552-559.

Barzel A, Liepert J, Haevernick K, Eisele M,          Bourbonnais D, Bilodeau S, Lepage Y, Beaudoin
  Ketels G, Rijntjes M, van den BH.                     N, Gravel D, Forget R. Effect of force-
  Comparison of two types of Constraint-                feedback treatments in patients with chronic
  Induced Movement Therapy in chronic stroke            motor deficits after a stroke. Am J Phys Med
  patients: A pilot study. Restor Neurol                Rehabil 2002; 81(12):890-897.
  Neurosci 2009;27:673-680.
                                                      Bovend'Eerdt TJ, Dawes H, Sackley C, Izadi H,
Bhakta BB, O'Connor RJ, Cozens JA. Associated           Wade DT. An integrated motor imagery
  reactions after stroke: a randomized                  program to improve functional task
  controlled trial of the effect of botulinum           performance in neurorehabilitation: a single-
  toxin type A. J Rehabil Med 2008;40:36-41.            blind randomized controlled trial. Arch Phys
                                                        Med Rehabil 2010;91:939-946.
Bhatt E, Nagpal A, Greer KH, Grunewald TK,
  Steele JL, Wiemiller JW, Lewis SM, Carey JR.        Bowman BR, Baker LL, Waters RL. Positional
  Effect of finger tracking combined with               feedback and electrical stimulation: an
  electrical stimulation on brain reorganization        automated treatment for the hemiplegic
  and hand function in subjects with stroke.            wrist. Arch Phys Med Rehabil
  Exp Brain Res 2007; [Epub ahead of print]             1979;60(11):497-502.

Blennerhassett J, Dite W. Additional task-            Boyd LA, Vidoni ED, Wessel BD. Motor learning
   related practice improves mobility and upper         after stroke: is skill acquisition a prerequisite
   limb function early after stroke: a                  for contralesional neuroplastic change?
   randomised controlled trial. Aust J Physiother       Neurosci Lett 2010;482:21-25.
                                                      Brashear A, Gordon MF, Elovic E et al.
Boake C, Noser EA, Ro T, Baraniuk S, Gaber M,           Intramuscular injection of botulinum toxin for
  Johnson R, Salmeron ET, Tran TM, Lai JM,              the treatment of wrist and finger spasticity
  Taub E, Moye LA, Grotta JC, Levin HS.                 after a stroke. N Engl J Med 2002;
  Constraint-Induced Movement Therapy                   347(6):395-400.
  During Early Stroke Rehabilitation.
  Neurorehabil Neurol Repair 2007;21:14-24.           Brashear A, McAfee AL, Kuhn ER, Fyffe J.
                                                        Botulinum toxin type B in upper-limb
Bobath, B. (Ed.). Adult Hemiplegia: evaluation          poststroke spasticity: a double-blind,
  and treatment. London: Heinemann Medical              placebo-controlled study. Arch Phys Med
  Books; 1990.                                          Rehabil 2004;85:705-9.

10. Upper Extremity Interventions                                                        pg. 132 of 150
Brogårdh C, Sjölund BH. Constrain-induced             Byl NN, Pitsch EA, Abrams GM. Functional
  movment threrapy in patients with stroke: a           outcomes can vary by dose: learning-based
  pilot study on effects of small group training        sensorimotor training for patients stable
  and of extended mitt use. Clin Rehabil                poststroke. Neurorehabil Neural Repair
  2006;20:218-227.                                      2008;22:494-504.

Brogardh C, Flansbjer UB, Lexell J. What is the       Cambier DC, De Corte E, Danneels LA,
  long-term benefit of constraint-induced               Witvrouw EE. Treating sensory impairments
  movement therapy? A four-year follow-up.              in the post-stroke upper limb with
  Clin Rehabil 2009a;23:418-423.                        intermittent pneumatic compression. Results
                                                        of a preliminary trial. Clin Rehabil
Brogardh C, Vestling M, Sjolund BH. Shortened           2003,17(1):14-20.
  constraint-induced movement therapy in
  subacute stroke - no effect of using a              Cardoso E, Rodrigues B, Lucena R, de Oliveira
  restraint: a randomized controlled study with         IR, Pedreira G, Melo A. Botulinum Toxin Type
  independent observers. J Rehabil Med                  A for the Treatment of the Upper Limb
  2009b;41:231-236.                                     Spasticity After Stroke. Arq Neuropsiquiatr
Brunnstrom S. Movement Therapy in
  Hemiplegia. New York: Harper & Row; 1970.           Carey JR. Manual stretch: effect on finger
                                                        movement control and force control in stroke
Burgar CG, Lum PS, Shor PC, Machiel Van der             subjects with spastic extrinsic finger flexor
  Loos HF. Development of robots for                    muscles. Arch Phys Med Rehabil
  rehabilitation therapy: the Palo Alto                 1990;71:888-894.
  VA/Stanford experience. J Rehabil Res Dev
  2000;37:663-673.                                    Carr J.H and Shepherd R.B. A Motor Relearning
                                                        Programme for Stroke 2nd ed. London:
Burge E, Kupper D, Finckh A, Ryerson S,                 William Heinemann Medical Books (1987).
  Schnider A, Leemann B. Neutral functional
  realignment orthosis prevents hand pain in          Carr M, Jones J. Physiological effects of exercise
  patients with subacute stroke: a randomized           on stroke survivors. Top Stroke Rehabil
  trial. Arch Phys Med Rehabil 2008;89:1857-            2003;9:57-64.
                                                      Caty GD, Detrembleur C, Bleyenheuft C,
Burns A, Burridge J, Pickering R. Does the use          Deltombe T, Lejeune TM. Effect of upper limb
  of a constraint mitten to encourage use of            botulinum toxin injections on impairment,
  the hemiplegic upper limb improve arm                 activity, participation, and quality of life
  function in adults with subacute stroke? Clin         among stroke patients. Stroke
  Rehabil 2007;21:895-904.                              2009;40:2589-2591.

Butler AJ, Page SJ. Mental practice with motor        Cauraugh JH, Coombes SA, Lodha N, Naik SK,
  imagery: evidence for motor recovery and              Summers JJ. Upper extremity improvements
  cortical reorganization after stroke. Arch            in chronic stroke: coupled bilateral load
  Phys Med Rehabil 2006;87:S2-11                        training. Restor Neurol Neurosci 2009;27:17-
Butefisch C, Hummelsheim H, Denzler P,
  Mauritz KH. Repetitive training of isolated         Cauraugh JH, Kim SB. Stroke motor recovery:
  movements improves the outcome of motor               active neuromuscular stimulation and
  rehabilitation of the centrally paretic hand. J       repetitive practice schedules. J Neurol
  Neurol Sci 1995;130(1):59-68.                         Neurosurg Psychiatry 2003;74(11):1562-6.

Byl N, Roderick J, Mohamed O, Hanny M, Kotler         Cauraugh J, Light K, Kim S, Thigpen M,
  J, Smith A, Tang M, Abrams G. Effectiveness           Behrman A. Chronic motor dysfunction after
  of sensory and motor rehabilitation of the            stroke: recovering wrist and finger extension
  upper limb following the principles of                by electromyography-triggered
  neuroplasticity: patients stable poststroke.          neuromuscular stimulation. Stroke
  Neurorehabil Neural Repair 2003;17(3):176-            2000;31(6):1360-4.
                                                      Cauraugh JH, Kim S. Two coupled motor
                                                        recovery protocols are better than one:

10. Upper Extremity Interventions                                                       pg. 133 of 150
  electromyogram-triggered neuromuscular               M, Cohen LG. Effects of somatosensory
  stimulation and bilateral movements. Stroke          stimulation on motor function after subacute
  2002; 33(6):1589-1594.                               stroke. Neurorehabil Neural Repair
Celnik P, Hummel F, Harris-Love M, Wolk R,
  Cohen LG. Somatosensory stimulation                Conforto AB, Kaelin-Lang A, Cohen LG. Increase
  enhances the effects of training functional          in hand muscle strength of stroke patients
  hand tasks in patients with chronic stroke.          after somatosensory stimulation. Ann Neurol
  Arch Phys Med Rehabil 2007;88:1369-1376              2002;51(1):122-125.

Chae J, Bethoux F, Bohine T, Dobos L, Davis T,       Conforto AB, Cohen LG, dos Santos RL, Scaff M,
  Friedl A. Neuromuscular stimulation for              Marie SK. Effects of somatosensory
  upper extremity motor and functional                 stimulation on motor function in chronic
  recovery in acute hemiplegia. Stroke                 cortico-subcortical strokes. J Neurol
  1998;29(5):975-979.                                  2007;254:333-339.

Chan MK, Tong RK, Chung KY. Bilateral upper          Connell LA, Lincoln NB, Radford KA.
  limb training with functional electric               Somatosensory impairment after stroke:
  stimulation in patients with chronic stroke.         frequency of different deficits and their
  Neurorehabil Neural Repair 2009;23:357-              recovery. Clin Rehabil 2008;22:758-767.
                                                     Coote S, Murphy B, Harwin W, Stokes E. The
Chen JC, Liang CC, Shaw FZ. Facilitation of            effect of the GENTLE/s robot-mediated
  sensory and motor recovery by thermal                therapy system on arm function after stroke.
  intervention for the hemiplegic upper limb in        Clin Rehabil 2008;22:395-405
  acute stroke patients: a single-blind
  randomized clinical trial. Stroke 2005;            Coupar F, Pollock A, van WF, Morris J,
  36(12):2665-2669.                                    Langhorne P. Simultaneous bilateral training
                                                       for improving arm function after stroke.
Cheng PT, Wang CM, Chung CY, Chen CL.                  Cochrane Database Syst Rev
  Effects of visual feedback rhythmic weight-          2010;4:CD006432.
  shift training on hemiplegic stroke patients.
  Clin Rehabil 2004;18:747-753.                      Cousins E, Ward A, Roffe C, Rimington L,
                                                       Pandyan A. Does low-dose botulinum toxin
Childers MK, Brashear A, Jozefczyk P, Reding M,        help the recovery of arm function when given
  Alexander D, Good D, Walcott JM, Jenkins             early after stroke? A phase II randomized
  SW, Turkel C, Molloy PT. Dose-dependent              controlled pilot study to estimate effect size.
  response to intramuscular botulinum toxin            Clin Rehabil 2010;24:501-513.
  type A for upper-limb spasticity in patients
  after a stroke. Arch Phys Med Rehabil              Cramer SC. The EXCITE trial: a major step
  2004; 85:1063-9.                                     forward for restorative therapies in stroke.
                                                       Stroke 2007;38:2204-2205.
Cirstea CM, Ptito A, Levin MF. Feedback and
   Cognition in Arm Motor Skill Reacquisition        Chae J, Harley MY, Hisel TZ, Corrigan CM,
   After Stroke. Stroke 2006;37:1237-1242.             Demchak JA, Wong YT, Fang ZP.
                                                       Intramuscular electrical stimulation for upper
Cirstea MC, Levin MF. Improvement of arm               limb recovery in chronic hemiparesis: an
   movement patterns and endpoint control              exploratory randomized clinical trial.
   depends on type of feedback during practice         Neurorehabil Neural Repair 2009;23:569-
   in stroke survivors. Neurorehabil Neural            578.
   Repair 2007;21:398-411
                                                     Crow JL, Lincoln NB, Nouri FM, DeWeerdt W.
Classen J, Liepert J, Wise SP, Hallett M, Cohen        The effectiveness of EMG biofeedback in the
   LG. Rapid plasticity of human cortical              treatment of arm function after stroke. Int
   movement representation induced by                  Disabil Stud 1989; 11:155-160.
   practice. J Neurophysiol 1998;79(2):1117-
   23.                                               Dahl A, Askim T, Stock R, Langorgen E,
                                                       Lydersen S, Indredavik B. Short- and long-
Conforto AB, Ferreiro KN, Tomasi C, dos Santos         term outcome of constraint-induced
  RL, Moreira VL, Marie SK, Baltieri SC, Scaff         movement therapy after stroke: a

10. Upper Extremity Interventions                                                     pg. 134 of 150
  randomized controlled feasibility trial. Clin         cycling and spasticity. J Neurol Sci
  Rehabil 2008;22:436-447.                              2007;253:18-24.

Daly JJ, Hogan N, Perepezko EM, Krebs HI,             Dohle C, Pullen J, Nakaten A, Kust J, Rietz C,
  Rogers JM, Goyal KS et al. Response to                Karbe H. Mirror therapy promotes recovery
  upper-limb robotics and functional                    from severe hemiparesis: a randomized
  neuromuscular stimulation following stroke. J         controlled trial. Neurorehabil Neural Repair
  Rehabil Res Dev 2005; 42(6):723-736.                  2009;23:209-217.

Dean CM, Shepherd RB. Task-related training           Dombovy ML. Rehabilitation and the course of
  improves performance of seated reaching               recovery after stroke. In Whisnant J (ed).
  tasks after stroke. A randomized controlled           Stroke: Populations, Cohorts and Clinical
  trial. Stroke 1997;28:722-728.                        Trials. Butterworth-Heinemann, 1993.

de Kroon JR, van der Lee JH, IJzerman MJ,             Dominkus M, Grisold W, Jelinek V. Transcranial
  Lankhorst GJ. Therapeutic electrical                  electrical motor evoked potentials as a
  stimulation to improve motor control and              prognostic indicator for motor recovery in
  functional abilities of the upper extremity           stroke patients. J Neurol Neurosurg
  after stroke: a systematic review. Clin               Psychiatry 1990;53(9):745-748.
  Rehabil 2002,16(4):350-360.
                                                      Donaldson C, Tallis R, Miller S, Sunderland A,
de Kroon JR, IJzerman MJ, Lankhorst GJ,                 Lemon R, Pomeroy V. Effects of conventional
  Zilvold G. Electrical stimulation of the upper        physical therapy and functional strength
  limb in stroke: stimulation of the extensors          training on upper limb motor recovery after
  of the hand vs. alternate stimulation of              stroke: a randomized phase II study.
  flexors and extensors. Am J Phys Med                  Neurorehabil Neural Repair 2009;23:389-
  Rehabil 2004;83(8):592-600.                           397.

de Kroon, Jr., IJzerman MJ. Electrical                Doyle S, Bennett S, Fasoli SE, McKenna KT.
  stimulation of the upper extremity in stroke:         Interventions for sensory impairment in the
  cyclic versus EMG-triggered stimulation. Clin         upper limb after stroke. Cochrane Database
  Rehabil 2008;22:690-697.                              Syst Rev 2010;6:CD006331.

Demeurisse G, Demol O, Robaye E. Motor                Dromerick AW, Edwards DF, Hahn M. Does the
  evaluation in vascular hemiplegia. Eur Neurol         application of constraint-induced movement
  1980;19(6):382-389.                                   therapy during acute rehabilitation reduce
                                                        arm impairment after ischemic stroke?
Desrosiers J, Bourbonnais D, Corriveau H,               Stroke 2000;31:2984-2988.
  Gosselin S, Bravo G. Effectiveness of
  unilateral and symmetrical bilateral task           Dromerick AW. The case for CI therapy:
  training for arm during the subacute phase            response. J Rehab Res Dev 2003a;40:xv-
  after stroke: a randomized controlled trial.          xvi.
  Clin Rehabil 2005;19:581-593.
                                                      Dromerick AW. Evidence-based rehabilitation:
Dickstein R, Hocherman S, Pillar T, Shaham R.           the case for and against constraint-induced
   Stroke rehabilitation. Three exercise therapy        movement therapy. J Rehab Res Dev
   approaches. Phys Ther 1986; 66(8):1233-              2003b;40:vii.
                                                      Dromerick AW, Lang CE, Birkenmeier RL,
Dickstein R, Heffes Y, Laufer Y, Abulaffio N,           Wagner JM, Miller JP, Videen TO, Powers WJ,
   Shabtai EL. Repetitive practice of a single          Wolf SL, Edwards DF. Very Early Constraint-
   joint movement for enhancing elbow function          Induced Movement during Stroke
   in hemiparetic patients. Percept Mot Skills          Rehabilitation (VECTORS): A single-center
   1997;85:771-785.                                     RCT. Neurology 2009;73:195-201.

Diserens K, Perret N, Chatelain S, Bashir S,          Duncan PW, Lai SM. Stroke recovery. Topics in
   Ruegg D, Vuadens P, Vingerhoets F. The               Stroke Rehabilitation 1997;4:51-58.
   effect of repetitive arm cycling on post stroke
   spasticity and motor control Repetitive arm        Duncan P, Studenski S, Richards L, Gollub S,
                                                        Lai SM, Reker D, Perera S, Yates J, Koch V,

10. Upper Extremity Interventions                                                      pg. 135 of 150
  Rigler S, Johnson D. Randomized clinical trial        Bever CT, Krebs HI, Hogan N. Short-duration
  of therapeutic exercise in subacute stroke.           robotic therapy in stroke patients with severe
  Stroke 2003;34(9):2173-80.                            upper-limb motor impairment. J Rehabil Res
                                                        Dev 2005;42:683-692.
Ertelt D, Small S, Solodkin A, Dettmers C,
   McNamara A, Binkofski F. Action Observation        Fischer HC, Stubblefield K, Kline T, Luo T,
   has positive impact on rehabilitation of motor        Kenyon RV, Kamper DG. Hand Rehabilitation
   deficits after stroke. Neuroimage                     Following Stroke: A Pilot Study of Assisted
   2007;36:164-173.                                      Finger Extension Training in a Virtual
                                                         Environment. Top Stroke Rehabil 2007;14:1-
Faghri PD, Rodgers MM, Glaser RM, Bors JG, Ho            12.
  C, Akuthota P. The effects of functional
  electrical stimulation on shoulder                  Francis HP, Wade DT, Turner-Stokes L,
  subluxation, arm function recovery, and                Kingswell RS, Dott CS, Coxon EA. Does
  shoulder pain in hemiplegic stroke patients.           reducing spasticity translate into functional
  Arch Phys Med Rehabil 1994;75(1):73-9.                 benefit? An exploratory meta-analysis. J
                                                         Neurol Neurosurg Psychiatry 2004;
Faghri PD, Rodgers MM. The effects of                    75(11):1547-1551.
  functional neuromuscular stimulation-
  augmented physical therapy program in the           Francisco G, Chae J, Chawla H, Kirshblum S,
  functional recovery of hemiplegic arm in               Zorowitz R, Lewis G, Pang S.
  stroke patients. Clin Kinesiol 1997;51(1):9-           Electromyogram-triggered neuromuscular
  15.                                                    stimulation for improving the arm function of
                                                         acute stroke survivors: a randomized pilot
Faghri PD, The effects of neuromuscular                  study. Arch Phys Med Rehabil
  stimulation-induced muscle contraction                 1998;79(5):570-575.
  versus elevation on hand edema in CVA
  patients. J Hand Ther 1997;10:29-34.                Francisco GE, Boake C, Vaughn A. Botulinum
                                                         toxin in upper limb spasticity after acquired
Fasoli, HI. Krebs, J Stein, WR. Frontera, N              brain injury: a randomized trial comparing
  Hogan, Effects of Robotic Therapy on Motor             dilution techniques. Am J Phys Med Rehabil
  Impairment and Recovery in Chronic Stroke.             2002; 81(5):355-363.
  Arch. Phys. Med. Rehab 2003;84:477-482.
                                                      Fritz SL, Light KE, Patterson TS, Behrman AL,
Fasoli SE, Krebs HI, Stein J, Frontera WR,               Davis SB. Active finger extension predicts
  Hughes R, Hogan N. Robotic therapy for                 outcomes after constraint-induced movement
  chronic motor impairments after stroke:                therapy for individuals with hemiparesis after
  Follow-up results. Arch Phys Med Rehabil               stroke. Stroke 2005;36:1172-1177.
  2004; 85(7):1106-1111.
                                                      Gabr U, Levine P, Page SJ. Home-based
Feeney DM, Gonzales A, Law W. Amphetamine,              electromyography-triggered stimulation in
  haloperidol, and experience interact to affect        chronic stroke. Clin Rehabil 2005;
  rate of recovery after motor cortex injury.           19(7):737-745.
  Science 1982;217(4562):855-7.
                                                      Gallichio JE. Pharmacologic management of
Ferraro M, Palazzolo JJ, Krol J et al. Robot-           spasticity following stroke. Phys Ther
  aided sensorimotor arm training improves              2004;84:973-981.
  outcome in patients with chronic stroke.
  Neurology. 2003;61:1604-1607.                       Galea MP, Miller KJ, Kilbreath SL. Early Task-
                                                        Related Training Enhances Upper Limb
Feys HM, De Weerdt WJ, Selz BE, Cox SG,                 Function Following Stroke. Poster presented
  Spichiger R, Vereeck LE, Putman KD, Van               at the annual meeting of the Society for
  Hoydonck GA. Effect of a therapeutic                  Neural Control of Movement, Sevilla, Spain,
  intervention for the hemiplegic upper limb in         2001.
  the acute phase after stroke: a single-blind,
  randomized, controlled multicenter trial.           Gelber DA. Comparison of two therapy
  Stroke 1998;29:785-792.                               approaches in the rehabilitation of the pure
                                                        motor hemiparetic stroke patient. J Neurol
Finley MA, Fasoli SE, Dipietro L, Ohlhoff J,            Rehabil 1995,9:191-196.
   MacClellan L, Meister C, Whitall J, Macko R,

10. Upper Extremity Interventions                                                       pg. 136 of 150
Gelber DA, Good DC, Dromerick A, Sergay S,               depression, shoulder pain and health-related
  Richardson M. Open-label dose-titration                quality of life in patients after stroke. J
  safety and efficacy study of tizanidine                Rehabil Med 2007;39:627-632
  hydrochloride in the treatment of spasticity
  associated with chronic stroke. Stroke               Hakkenes S, Keating JL. Constraint-Induced
  2001;32:1841-1846.                                     movement therapy following stroke
                                                         systematic review of randomised controlled
Gilmore PE, Spaulding SJ. Motor learning and             trials. Australian Journal of Physiotherapy
   the use of videotape feedback after stroke.           2005;51:221-231.
   Top Stroke Rehabil 2007;14:28-36.
                                                       Hammer A, Lindmark B. Is forced use of the
Giudice M. Effects of continuous passive motion          paretic upper limb beneficial? A randomized
  and elevation on hand edema. Am J. Occup               pilot study during subacute post-stroke
  Therap 1990;44 (10):914-921.                           recovery. Clin Rehabil 2009a;23:424-433.

Glanz M, Klawansky S, Stason W, Berkey C,              Hammer AM, Lindmark B. Effects of forced use
  Chalmers TC. Functional electrostimulation in          on arm function in the subacute phase after
  poststroke rehabilitation: a meta-analysis of          stroke: a randomized, clinical pilot study.
  the randomized controlled trials. Arch Phys            Phys Ther 2009b;89:526-539.
  Med Rehabil 1996;77(6) :549-53.
                                                       Hanlon RE. Motor learning following unilateral
Gordon MF, Brashear A, Elovic E, Kassicieh D,            stroke. Arch Phys Med Rehabil
  Marciniak C, Liu J et al. Repeated dosing of           1996;77(8):811-5.
  botulinum toxin type A for upper limb
  spasticity following stroke. Neurology 2004;         Hara Y, Ogawa S, Muraoka Y. Hybrid Power-
  63(10):1971-1973.                                      Assisted Functional Electrical Stimulation to
                                                         Improve Hemiparetic Upper-Extremity
Gowland C, Stratford P, Ward M, Moreland J,              Function. Am J Phys Med Rehabil
  Torresin W, Van Hullenaar S, Sanford J,                2006;85:977-985.
  Barreca S, Vanspall B, Plews N. Measuring
  physical impairment and disability with the          Hara Y, Ogawa S, Tsujiuchi K, Muraoka Y. A
  Chedoke-McMaster Stroke Assessment.                    home-based rehabilitation program for the
  Stroke 1993;24(1):58-63.                               hemiplegic upper extremity by power-
                                                         assisted functional electrical stimulation.
Gowland C. Recovery of motor function                    Disabil Rehabil 2008;30:296-304.
  following stroke: Profile and prediction.
  Physiotherapy Canada 1982; 34, 77-84.                Harris JE, Eng JJ, Miller WC, Dawson AS. A self-
                                                         administered Graded Repetitive Arm
Gracies JM, Marosszeky JE, Renton R,                     Supplementary Program (GRASP) improves
  Sandanam J, Gandevia SC, Burke D. Short-               arm function during inpatient stroke
  term effects of dynamic lycra splints on               rehabilitation: a multi-site randomized
  upper limb in hemiplegic patients. Arch Phys           controlled trial. Stroke 2009;40:2123-2128.
  Med Rehabil 2000;81(12):1547-55.
                                                       Heckermann J, Mokrusch T, Krockel A, Warnke
Greenberg S, Fowler RS Jr. Kinesthetic                   S, von Stockert T, Neundorfer B. EMG-
  biofeedback: a treatment modality for elbow            triggered electrical muscle stimulation in the
  range of motion in hemiplegia. Am J Occup              treatment of central hemiparesis after a
  Ther 1980;34(11):738-743.                              stroke. Eur J Phys Med Rehabil 1997;
Grotta JC, Noser EA, Ro T, et al. Constraint-
  induced movement therapy. Stroke                     Hemmen B, Seelen HA. Effects of movement
  2004;35:2699-2701.                                     imagery and electromyography-triggered
                                                         feedback on arm hand function in stroke
Hafsteinsdóttir TB, Algra A, Kappelle LJ,                patients in the subacute phase. Clin Rehabil
  Grypdonck. Neurodevelopmental treatment                2007;21:587-594.
  after stroke: a comparative study. J Neurol
  Neurosurg Psychiatry 2005; 76:788-792.               Henderson A, Korner-Bitensky N, Levin M.
                                                         Virtual Reality in Stroke Rehabilitation: A
Hafsteinsdottir TB, Kappelle J, Grypdonck MH,            Systematic Review of its Effectiveness for
  Algra A. Effects of Bobath-based therapy on

10. Upper Extremity Interventions                                                       pg. 137 of 150
  Upper Limb Motor Recovery. Top Stroke                Function in Patients With Stroke. IEEE Trans
  Rehabil 2007;14:52-61.                               Neural Syst Rehabil Eng 2007;15:36-42.

Hendricks HT, van Limbeek J, Geurts AC,              Horowitz R. Road to recovery: drugs used in
  Zwarts MJ. Motor recovery after stroke: a            stroke rehabilitation Expert Rev
  systematic review of the literature. Arch Phys       Neurotherapeutics 2004;4(2):219-231
  Med Rehabil 2002;83(11):1629-37.
                                                     Horsley SA, Herbert RD, Ada L. Four weeks of
Hesse S, Reiter F, Konrad M, Jahnke MT.                daily stretch has little or no effect on wrist
  Botulinum toxin type A and short-term                contracture after stroke: a randomised
  electrical stimulation in the treatment of           controlled trial. Aust J Physiother
  upper limb flexor spasticity after stroke: a         2007;53:239-245.
  randomized, double-blind, placebo-controlled
  trial. Clin Rehabil 1998;12:381-388.               Housman SJ, Scott KM, Reinkensmeyer DJ. A
                                                       randomized controlled trial of gravity-
Hesse S, Schulte-Tigges G, Konrad M,                   supported, computer-enhanced arm exercise
  Bardeleben A, Werner C. Robot-assisted arm           for individuals with severe hemiparesis.
  trainer for the passive and active practice of       Neurorehabil Neural Repair 2009;23:505-
  bilateral forearm and wrist movements in             514.
  hemiparetic subjects. Arch Phys Med Rehabil
  2003;84:915-20.                                    Hsu SS, Hu MH, Wang YH, Yip PK, Chiu JW,
                                                       Hsieh CL. Dose-response relation between
Hesse S, Kuhlmann H, Wilk J, Tomelleri C,              neuromuscular electrical stimulation and
  Kirker SG. A new electromechanical trainer           upper-extremity function in patients with
  for sensorimotor rehabilitation of paralysed         stroke. Stroke 2010;41:821-824.
  fingers: a case series in chronic and acute
  stroke patients. J Neuroeng Rehabil                Hu XL, Tong KY, Song R, Zheng XJ, Leung WW.
  2008;5:21.                                           A comparison between electromyography-
                                                       driven robot and passive motion device on
Hesse S, Werner C, Pohl M, Mehrholz J, Puzich          wrist rehabilitation for chronic stroke.
  U, Krebs HI. Mechanical arm trainer for the          Neurorehabil Neural Repair 2009;23:837-
  treatment of the severely affected arm after         846.
  a stroke: a single-blinded randomized trial in
  two centers. Am J Phys Med Rehabil                 Hummel F, Celnik P, Giraux P, Floel A, Wu WH,
  2008;87:779-788.                                     Gerloff C et al. Effects of non-invasive
                                                       cortical stimulation on skilled motor function
Higgins J, Salbach NM, Wood-Dauphinee S,               in chronic stroke. Brain 2005; 128(Pt
  Richards CL, Cote R, Mayo NE. The effect of          3):490-499.
  a task-oriented intervention on arm function
  in people with stroke: a randomized                Hummelsheim H, Eickhof C. Repetitive
  controlled trial. Clin Rehabil 2006;                 sensorimotor training for arm and hand in a
  20(4):296-310.                                       patient with locked-in syndrome. Scand.J
                                                       Rehabil Med 1999,31(4):250-256.
Hiraoka K. Rehabilitation effort to improve
   upper extremity function in post-stroke           Hummelsheim H, Maier-Loth ML, Eickhof C. The
   patients:A meta-analysis. J Phys Ther Sci           functional value of electrical muscle
   2005;13:5-9.                                        stimulation for the rehabilitation of the hand
                                                       in stroke patients. Scand J Rehabil Med
Holden M, Todorov E, Callahan J, Bizzi E. Virtual      1997;29(1):3-10.
  environment training improves motor
  performance with stroke: case report. Neuro.       Hurd WW, Pegram V, Nepomuceno C.
  Rep 1999; 23(2): 57-67.                              Comparison of actual and simulated EMG
                                                       biofeedback in the treatment of hemiplegic
Holden M, Dyar T. Virtual environment training:        patients. Am J Phys Med 1980;59(2):73-82.
  a new tool for neurorehabilitation. Neurol
  Rep. 2002;26:62-71.                                Inglis J, Donald MW, Monga TN, Sproule M,
                                                       Young MJ. Electromyographic biofeedback
Holden MK, Dyar TA, Dyan-Cimadoro L.                   and physical therapy of the hemiplegic upper
  Telerehabilitation Using a Virtual                   limb. Arch Phys Med Rehabil 1984;65:755-
  Environment Improves Upper Extremity                 759.

10. Upper Extremity Interventions                                                      pg. 138 of 150
Jack D, Boian R, Merians A, Tremaine M,                 and cortical changes in subjects with stroke.
   Burdea G, Adamovich S, Recce M, Poizner H.           Exp Brain Res 2004;154:450-460.
   Virtual Reality-Enhanced Stroke
   Rehabilitation. IEEE Trans Neural Syst             King I. The effect of neuromuscular electrical
   Rehabil Eng 2001;9(3), 308-318.                       stimulation in reducing tone. Am J Occup
                                                         Ther 1996; 50(1):62-64.
Jahangir AW, Tan HJ, Norlinah MI, Nafisah WY,
  Ramesh S, Hamidon BB, Raymond AA.                   Klaiput A, Kitisomprayoonkul W. Increased
  Intramuscular injection of botulinum toxin for         pinch strength in acute and subacute stroke
  the treatment of wrist and finger spasticity           patients after simultaneous median and ulnar
  after stroke. Med J Malaysia 2007;62:319-              sensory stimulation. Neurorehabil Neural
  322.                                                   Repair 2009;23:351-356.

Jang SH, You SH, Hallett M, Cho YW, Park CM,          Kong KH, Chua KS. Neurolysis of the
  Cho SH et al. Cortical reorganization and             musculocutaneous nerve with alcohol to treat
  associated functional motor recovery after            poststroke elbow flexor spasticity. Arch Phys
  virtual reality in patients with chronic stroke:      Med Rehabil 1999;80:1234-1236.
  an experimenter-blind preliminary study.
  Arch Phys Med Rehabil 2005; 86(11):2218-            Kopp B, Kunkel A, Muhlnickel W, Villringer K,
  2223.                                                 Taub E, Flor H. Plasticity in the motor system
                                                        related to therapy-induced improvement of
Johansson K, Lindgren I, Widner H, Wiklund I,           movement after stroke. Neuroreport
  Johansson BB. Can sensory stimulation                 1999;10:807-810.
  improve the functional outcome in stroke
  patients? Neurology 1993;43(11):2189-               Kowalczewski J, Gritsenko V, Ashworth N,
  2192.                                                 Ellaway P, Prochazka A. Upper-extremity
                                                        functional electric stimulation-assisted
Jongbloed L, Stacey S, Brighton C.Stroke                exercises on a workstation in the subacute
  rehabilitation: sensorimotor integrative              phase of stroke recovery. Arch Phys Med
  treatment versus functional treatment. Am J           Rehabil 2007;88:833-839.
  Occup Ther 1989; 43(6):391-397.
                                                      Kraft GH, Fitts SS, Hammond MC. Techniques
Kahn LE, Zygman ML, Rymer WZ,                           to improve function of the arm and hand in
  Reinkensmeyer DJ. Robot-assisted reaching             chronic hemiplegia. Arch Phys Med Rehabil
  exercise promotes arm movement recovery               1992;73(3):220-227.
  in chronic hemiparetic stroke: a randomized
  controlled pilot study. J Neuro Eng Rehab           Krebs, H.I., Volpe, B.T., Ferraro, M., Fasoli, S.,
  2005;3:12.                                            Palazzolo, J., Rohrer, B., Edelstein, L.,
                                                        Hogan, N. Robot-Aided Neuro-Rehabilitation:
Kaji R, Osako Y, Suyama K, Maeda T, Uechi Y,            From Evidence-Based to Science-Based
  Iwasaki M. Botulinum toxin type A in post-            Rehabilitation. Topics in Stroke Rehabilitation
  stroke upper limb spasticity. Curr Med Res            2002;8(4):54-70.
  Opin 2010;26:1983-1992.
                                                      Kunkel A, Kopp B, Muller G, Villringer K,
Kanovsky P, Slawek J, Denes Z, Platz T, Sassin          Villringer A, Taub E, Flor H. Constraint-
  I, Comes G, Grafe S. Efficacy and safety of           induced movement therapy for motor
  botulinum neurotoxin NT 201 in poststroke             recovery in chronic stroke patients. Arch
  upper limb spasticity. Clin Neuropharmacol            Phys Med Rehabil 1999,80(6):624-628.
                                                      Kutner NG, Zhang R, Butler AJ, Wolf SL, Alberts
Karni A, Meyer G, Jezzard P, Adams MM, Turner           JL. Quality-of-life change associated with
  R, Ungerleider LG. Functional MRI evidence            robotic-assisted therapy to improve hand
  for adult motor cortex plasticity during motor        motor function in patients with subacute
  skill learning. Nature 1995                           stroke: a randomized clinical trial. Phys Ther
  Sep;377(6545):155-8.                                  2010;90:493-504.

Kimberley TJ, Lewis SM, Auerbach EJ, Dorsey           Kwakkel G, Wagenaar RC, Twisk JW, Lankhorst
  LL, Lojovich JM, Carey JR. Electrical                 GJ, Koetsier JC. Intensity of leg and arm
  stimulation driving functional improvements           training after primary middle-cerebral-artery

10. Upper Extremity Interventions                                                       pg. 139 of 150
  stroke: a randomised trial. Lancet                  Lannin NA, Horsley SA, Herbert R, McCluskey A,
  1999;354(9174):191-196.                               Cusick A. Splinting the hand in the functional
                                                        position after brain impairment: a
Kwakkel G, van Dijk GM, Wagenaar RC.                    randomized, controlled trial. Arch Phys Med
  Accuracy of physical and occupational                 Rehabil. 2003;84(2):297-302.
  therapists' early predictions of recovery after
  severe middle cerebral artery stroke. Clin          Lannin NA, Cusick A, McCluskey A, Herbert RD.
  Rehabil 2000;14(1):28-41.                             Effects of Splinting on Wrist Contracture
                                                        After Stroke: A Randomized Controlled Trial.
Kwakkel G, Kollen BJ, van der GJ, Prevo AJ.             Stroke 2007;38:111-116.
  Probability of regaining dexterity in the
  flaccid upper limb: impact of severity of           Latimer CP, Keeling J, Lin B, Henderson M, Hale
  paresis and time since onset in acute stroke.         LA. The impact of bilateral therapy on upper
  Stroke 2003;34:2181-2186.                             limb function after chronic stroke: a
                                                        systematic review. Disabil Rehabil
Kwakkel G, Kollen BJ, Krebs HI. Effects of              2010;32:1221-1231.
  robot-assisted therapy on upper limb
  recovery after stroke: a systematic review.         Lee KH, Hill E, Johnston R, Smiehorowski T.
  Neurorehabil Neural Repair 2008;22:111-               Myofeedback for muscle retraining in
  121.                                                  hemiplegic patients. Arch Phys Med Rehabil
Lagalla G, Danni M, Reiter F, Ceravolo MG,
  Provinciali L. Post-stroke spasticity               Leibovitz A, Baumoehl Y, Roginsky Y et al.
  management with repeated botulinum toxin               Edema of the paretic hand in elderly post-
  injections in the upper limb. Am J Phys Med            stroke nursing patients. Arch Gerontol
  Rehabil 2000; 79(4):377-384.                           Geriatr. 2007;44:37-42.

Lai JM, Francisco GE, Willis FB. Dynamic              Levy CE, Giuffrida C, Richards L, Wu S, Davis S,
   splinting after treatment with botulinum             Nadeau SE. Botulinum toxin a, evidence-
   toxin type-A: A randomized controlled pilot          based exercise therapy, and constraint-
   study. Adv Ther 2009;26:241-248.                     induced movement therapy for upper-limb
                                                        hemiparesis attributable to stroke: a
Lam YS, Man DWK, Tam SF, Weiss PL. Virtual              preliminary study. Am J Phys Med Rehabil
  reality training for stroke rehabilitation.           2007;86:696-706.
  Neurorehabilitation 2006;21:245-253.
                                                      Lin KC, Chung HY, Wu CY, Liu HL, Hsieh YW,
Langhammer B, Stanghelle JK. Bobath or motor             Chen IH, Chen CL, Chuang LL, Liu JS, Wai
  relearning programme? A comparison of two              YY. Constraint-induced therapy versus
  different approaches of physiotherapy in               control intervention in patients with stroke: a
  stroke rehabilitation: a randomized                    functional magnetic resonance imaging
  controlled study. Clin Rehabil                         study. Am J Phys Med Rehabil 2010;89:177-
  2000;14(4):361-9.                                      185.

Langhammer B, Stanghelle JK. Bobath or motor          Lin KC, Chen YA, Chen CL, Wu CY, Chang YF.
  relearning programme? A follow-up one and              The effects of bilateral arm training on motor
  four years post stroke. Clin Rehabil                   control and functional performance in chronic
  2003;17(7):731-734.                                    stroke: a randomized controlled study.
                                                         Neurorehabil Neural Repair 2010;24:42-51.
Langlois S, Pederson L, Mackinnon JR. The
  effects of splinting on the spastic hemiplegic      Lin KC, Chang YF, Wu CY, Chen YA. Effects of
  hand: report of a feasibility study. Can J             constraint-induced therapy versus bilateral
  Occup Ther 1991;58:17-25.                              arm training on motor performance, daily
                                                         functions, and quality of life in stroke
Lannin NA, Herbert RD. Is hand splinting                 survivors. Neurorehabil Neural Repair
  effective for adults following stroke? A               2009;23:441-448.
  systematic review and methodologic critique
  of published research. Clin Rehabil.                Lin KC, Wu CY, Wei TH, Lee CY, Liu JS. Effects
  2003;17(8):807-16.                                     of modified constraint-induced movement
                                                         therapy on reach-to-grasp movements and
                                                         functional performance after chronic stroke:

10. Upper Extremity Interventions                                                       pg. 140 of 150
  a randomized controlled study. Clin Rehabil            subacute stroke subjects: A follow-up study.
  2007;21:1075-1086.                                     J Rehabil Res Dev 2006;43:631-642.

Lin KC, Wu CY, Liu JS. A randomized controlled         Macclellan LR, Bradham DD, Whitall J, Volpe B,
   trial of constraint-induced movement therapy          Wilson PD, Ohlhoff J et al. Robotic upper-
   after stroke. Acta Neurochir Suppl                    limb neurorehabilitation in chronic stroke
   2008;101:61-64.                                       patients. J Rehabil Res Dev 2005;
Lin KC, Wu CY, Liu JS, Chen YT, Hsu CJ.
   Constraint-induced therapy versus dose-             Manganotti P, Amelio E. Long-term effect of
   matched control intervention to improve               shock wave therapy on upper limb
   motor ability, basic/extended daily functions,        hypertonia in patients affected by stroke.
   and quality of life in stroke. Neurorehabil           Stroke 2005; 36(9):1967-1971.
   Neural Repair 2009;23:160-165.
                                                       Mangold S, Schuster C, Keller T, Zimmermann-
Lincoln NB, Parry RH, Vass CD. Randomized,               Schlatter A, Ettlin T. Motor training of upper
   controlled trial to evaluate increased                extremity with functional electrical
   intensity of physiotherapy treatment of arm           stimulation in early stroke rehabilitation.
   function after stroke. Stroke 1999;30:573-            Neurorehabil Neural Repair 2009;23:184-
   579.                                                  190.

Lo AC, Guarino PD, Richards LG, Haselkorn JK,          Mann G, Burridge JH, Malone L, Strike PW. A
  Wittenberg GF, Federman DG, Ringer RJ,                 pilot study to investigate the effects of
  Wagner TH, Krebs HI, Volpe BT, Bever CT,               electrical stimulation on recovery of hand
  Jr., Bravata DM, Duncan PW, Corn BH,                   function and sensation in subacute stroke
  Maffucci AD, Nadeau SE, Conroy SS, Powell              patients. Neuromodulation 2005;8:193-202.
  JM, Huang GD, Peduzzi P. Robot-Assisted
  Therapy for Long-Term Upper-Limb                     Masiero S, Celia A, Armani M, Rosati G. A novel
  Impairment after Stroke. N Engl J Med 2010.            robot device in rehabilitation of post-stroke
                                                         hemiplegia upper limbs. Aging Clin Exp Res
Logigian MK, Samuels MA, Falconer J, Zagar R.            2006;18:531-535.
  Clinical exercise trial for stroke patients. Arch
  Phys Med Rehabil 1983;64(8):364-367.                 Masiero S, Celia A, Rosita G, Armani M.
                                                         Robotic-Assisted Rehabilitation of the Upper
Lord JP, Hall K. Neuromuscular reeducation               Limb After Acute Stroke. Arch Phys Med
  versus traditional programs for stroke                 Rehabil 2007;88:142-149.
  rehabilitation. Arch Phys Med Rehabil
  1986;67(2):88-91.                                    McCrory P, Turner-Stokes L, Baguley IJ, De GS,
                                                         Katrak P, Sandanam J, Davies L, Munns M,
Luft AR, McCombe-Waller S, Whitall J, et al.             Hughes A. Botulinum toxin A for treatment of
  Repetitive bilateral arm training and motor            upper limb spasticity following stroke: a
  cortex activation in chronic stroke: a                 multi-centre randomized placebo-controlled
  randomized controlled trial. JAMA                      study of the effects on quality of life and
  2004;292:1853-1861.                                    other person-centred outcomes. J Rehabil
                                                         Med 2009;41:536-544.
Luke C, Dodd KJ, Brock K. Outcomes of the
  Bobath concept on upper limb recovery                McDonnell MN, Hillier SL, Miles TS, Thompson
  following stroke. Clin Rehabil 2004;18:888-            PD, Ridding MC. Influence of combined
  898.                                                   afferent stimulation and task-specific training
                                                         following stroke: a pilot randomized
Lum PS, Burgar CG, Shor PC, Majmundar M,                 controlled trial. Neurorehabil Neural Repair
  Van der Loos M. Robot-assisted movement                2007;21:435-443
  training compared with conventional therapy
  techniques for the rehabilitation of upper-          McPherson JJ, Kreimeyer D, Aalderks M,
  limb motor function after stroke. Arch Phys            Gallagher T. A comparison of dorsal and
  Med Rehabil 2002;83(7):952-9.                          volar resting hand splints in the reduction of
                                                         hypertonus. Am J Occup Ther
Lum PS, Burgar CG, Loos MV, Shor PC,                     1982;36(10):664-70.
  Majmundar M, Yap R. MIME robotic device
  for upper-limb neurorehabilitation in

10. Upper Extremity Interventions                                                       pg. 141 of 150
Mehrholz J, Platz T, Kugler J, Pohl M.                  controlled trial. Arch Phys Med Rehabil
  Electromechanical and robot-assisted arm              2008;89:1237-1245.
  training for improving arm function and
  activities of daily living after stroke.            Mroczek N, Halpern D, McHugh R.
  Cochrane Database Syst Rev                            Electromyographic feedback and physical
  2008;CD006876.                                        therapy for neuromuscular retraining in
                                                        hemiplegia. Arch Phys Med Rehabil
Meilink A, Hemmen B, Seelen HA, Kwakkel G.              1978;59:258-267.
  Impact of EMG-triggered neuromuscular
  stimulation of the wrist and finger extensors       Mudie MH, Matyas TA. Can simultaneous
  of the paretic hand after stroke: a systematic        bilateral movement involve the undamaged
  review of the literature. Clin Rehabil                hemisphere in reconstruction of neural
  2008;22:291-305.                                      networks damaged by stroke? Disabil Rehabil
Merians AS, Jack D, Boian R, Tremaine M,
  Burdea GC, Adamovich SV, Recce M, Poizner           Myers BJ. Proprioceptive neuromuscular
  H. Virtual reality-augmented rehabilitation           facilitation approach. In Trombly CA, ed.
  for patients following stroke. Phys Ther.             (Ed.), Occupational Therapy for Physical
  2002 Sep;82(9):898-915.                               Dysfunction: 474-498. Baltimore, MD:
                                                        Williams & Wilkins; 1995.
Merians AS, Poizner H, Boian R, Burdea G,
  Adamovich S. Sensorimotor Training in a             Myint JM, Yuen GF, Yu TK, et al. A study of
  Virtual Reality Environment: Does It Improve          constraint-induced movement therapy in
  Functional Recovery Poststroke?                       subacute stroke patients in Hong Kong. Clin
  Neurorehabil Neurol Repair 2006;20:252-               Rehabil 2008;22:112-124
                                                      Nadeau SE, Behrman AL, Davis SE, Reid K, Wu
Meythaler JM, Vogtle L, Brunner RC. A                   SS, Stidham RN, Helms KM, Gonzalez LJ.
  preliminary assessment of the benefits of the         Donepezil as an adjuvant to constraint-
  addition of botulinum toxin a to a                    induced therapy for upper-limb dysfunction
  conventional therapy program on the                   after stroke: an exploratory randomized
  function of people with longstanding stroke.          clinical trial. 2004;41:525-534.
  Arch Phys Med Rehabil 2009;90:1453-1461.
                                                      Nakayama H, Jorgensen HS, Raaschou HO,
Michaelsen SM, Levin MF. Short-term effects of          Olsen TS. Recovery of upper extremity
  practice with trunk restraint on reaching             function in stroke patients: the Copenhagen
  movements in patients with chronic stroke: a          Stroke Study. Arch Phys Med Rehabil
  controlled trial. Stroke 2004;35:1914-1919.           1994;75:394-398.

Michaelsen SM, Dannenbaum R, Levin MF.                O’Brian C. Treatment of spasticity with
  Task-specific Training With Trunk Restraint           botulinum toxin. Clin J Pain 2002; 18:S182-
  on Arm Recovery in Stroke. Stroke                     190.
                                                      Paci M. Physiotherapy based on the Bobath
Miller GJ, Light KE. Strength training in spastic       Concept for adults with post-stroke
   hemiparesis? Should it be avoided?                   hemiplegia: a review of effectivenedd
   NeuroRehabilitation 1997;9:17-28.                    studies. J Rehabil Med 2003; 35:2-7.

Moreland J, Thomson MA. Efficacy of                   Page SJ. Intensity versus task-specificity after
  electromyographic biofeedback compared                stroke: how important is intensity? Am J
  with conventional physical therapy for upper-         Phys Med Rehabil 2003;82(9):730-2.
  extremity function in patients following
  stroke: a research overview and meta-               Page SJ, Sisto S, Johnston MV, Levine P.
  analysis. Phys Ther 1994;74(6):534-43;                Modified constraint-induced therapy after
  discussion 544-7.                                     subacute stroke: a preliminary study.
                                                        Neurorehabil Neural Repair 2002; 16(3):290-
Morris JH, van WF, Joice S, Ogston SA, Cole I,          295.
  MacWalter RS. A comparison of bilateral and
  unilateral upper-limb task training in early        Page SJ, Sisto S, Levine P, McGrath RE.
  poststroke rehabilitation: a randomized               Efficacy of modified constraint-induced

10. Upper Extremity Interventions                                                       pg. 142 of 150
  movement therapy in chronic stroke: a                 enhance sensorimotor recovery in chronic
  single-blinded randomized controlled trial.           stroke. Clin Rehabil 2002;16:709-716.
  Arch Phys Med Rehabil. 2004;85(1):14-8.
                                                      Piron L, Turolla A, Agostini M, Zucconi CS,
Page SJ, Levine P, Leonard AC. Modified                  Ventura L, Tonin P, Dam M. Motor learning
  Constraint-Induced Therapy in Acute Stroke:            principles for rehabilitation: a pilot
  A Randomized Controlled Pilot Study.                   randomized controlled study in poststroke
  Neurorehabil Neural Repair 2005;19:27-32.              patients. Neurorehabil Neural Repair
Page SJ, Levine P, Leonard AC. Effects of
  mental practice on affected limb use and            Piron L, Turolla A, Agostini M, Zucconi C,
  function in chronic stroke. Arch Phys Med              Cortese F, Zampolini M, Zannini M, Dam M,
  Rehabil 2005; 86(3):399-402.                           Ventura L, Battauz M, Tonin P. Exercises for
                                                         paretic upper limb after stroke: a combined
Page SJ, Levine P. Back From the Brink:                  virtual-reality and telemedicine approach. J
  Electromyography-Triggered Stimulation                 Rehabil Med 2009;41:1016-1102.
  Combined with Constraint-Induced
  Movement Therapy in Chronic Stroke. Arch            Piron L, Toin P, Atzori AM, Zucci Z, Massaro C,
  Phys Med Rehabil 2006;87:27-31.                        Trivello E, Dam M. The augumented-
                                                         feedback rehabilitation technique facilitates
Page SJ, Levine P, Leonard A. Mental Practice in         the arm motor recovery in patients after a
  Chronic Stroke: Results of a Randomized,               recent stroke. Stud Health Technol Inform
  Placebo-Controlled Trial. Stroke                       2003;94:265-297.
                                                      Piron L, Tonin P, Piccione F, Iaia V, Trivello E,
Page SJ, Levine P, Leonard A, Szaflarski JP,             Dam M. Virtual environment training therapy
  Kissela BM. Modified constraint-induced                for arm motor rehabilitation. Presense
  therapy in chronic stroke: results of a single-        2005;14:732-740.
  blinded randomized controlled trial. Phys
  Ther 2008;88:333-340.                               Pizzi A, Carlucci G, Falsini C, Verdesca S, Grippo
                                                         A. Application of a volar static splint in
Page SJ, Levine P, Khoury JC. Modified                   poststroke spasticity of the upper limb. Arch
  constraint-induced therapy combined with               Phys Med Rehabil 2005; 86(9):1855-1859.
  mental practice: thinking through better
  motor outcomes. Stroke 2009;40:551-554.             Platz T, van KS, Mehrholz J, Leidner O, Eickhof
                                                         C, Pohl M. Best conventional therapy versus
Pandyan AD, Granat MH, Stott DJ.Effects of               modular impairment-oriented training for
  electrical stimulation on flexion contractures         arm paresis after stroke: a single-blind,
  in the hemiplegic wrist. Clin Rehabil                  multicenter randomized controlled trial.
  1997;11(2):123-130.                                    Neurorehabil Neural Repair 2009;23:706-
Pandyan AD, Vuadens P, van Wijck FMJ, Stark
  S, Johnson GR, Barnes M. Are we                     Platz T, Eickhof C, van Kaick S, et al.
  underestimating the clinical efficacy of               Impairment-oriented training or Bobath
  botulinum toxin (type A)? Quantifying                  therapy for severe arm paresis after stroke:
  changes in spasticity, strength and upper              a single-blind, multicentre randomized
  limb function after injections of Botox to the         controlled trial. Clin Rehabil 2005;19:714-
  elbow flexors in a unilateral stroke                   724.
  population. Clin Rehabil 2002;16:654-660.
                                                      Platz T, Kim IH, Engel U, Pinkowski C, Eickhof
Partridge C, Mackenzie M, Edwards S, Reid A,             C, Kutzner M. Amphetamine fails to facilitate
  Jayawardena S, Guck N, Potter J. Is dosage             motor performance and to enhance motor
  of physiotherapy a critical factor in deciding         recovery among stroke patients with mild
  patterns of recovery from stroke: a                    arm paresis: interim analysis and
  pragmatic randomized controlled trial.                 termination of a double blind, randomised,
  Physiother Res Int 2000;5(4):230-240.                  placebo-controlled trial. Restor Neurol
                                                         Neurosci 2005;23:271-280.
Peurala SH, Pitkanen K, Sivenius J, Tarkka IM.
  Cutaneous electrical stimulation may                Platz T, Winter T, Muller N, Pinkowski C, Eickhof
                                                         C, Mauritz KH. Arm ability training for stroke

10. Upper Extremity Interventions                                                       pg. 143 of 150
  and traumatic brain injury patients with mild       Prange GB, Jannink MJ, Groothuis-Oudshoorn
  arm paresis: a single-blind, randomized,               CG et al. Systematic review of the effect of
  controlled trial. Arch Phys Med Rehabil                robot-aided therapy on recovery of the
  2001;82:961-968.                                       hemiparetic arm after stroke. J Rehabil Res
                                                         Dev. 2006;43:171-184.
Ploughman M, Corbett D. Can forced-use
   therapy be clinically applied after stroke? An     Prevo AJ, Visser SL, Vogelaar TW. Effect of EMG
   exploratory randomized controlled trial. Arch         feedback on paretic muscles and abnormal
   Phys Med Rehabil. 2004;85(9):1417-23.                 co-contraction in the hemiplegic arm,
                                                         compared with conventional physical
Pohl PS, Winstein CJ. Practice effects on the            therapy. Scand J Rehabil Med
  less-affected upper extremity after stroke.            1982;14(3):121-131.
  Arch Phys Med Rehabil 1999;80:668-675.
                                                      Quaney BM, He J, Timberlake G, Dodd K, Carr
Poole JL, Whitney SL, Hangeland N, Baker C.             C. Visuomotor training improves stroke-
  The effectiveness of inflatable pressure              related ipsilesional upper extremity
  splints on motor function in stroke patients.         impairments. Neurorehabil Neural Repair
  Occup Ther J Res 1990;10:360-366.                     2010;24:52-61.

Pomeroy VM, King L, Pollock A, Baily-Hallam A,        Rabadi M, Galgano M, Lynch D, Akerman M,
  Langhorne P. Electrostimulation for                   Lesser M, Volpe B. A pilot study of activity-
  promoting recovery of movement or                     based therapy in the arm motor recovery
  functional ability after stroke. Cochrane             post stroke: a randomized controlled trial.
  Database Syst Rev 2006;(2):CD003241.                  Clin Rehabil 2008;22:1071-1082.

Popovic MR, Popovic DB, Keller T.                     Restemeyer C, Weiller C, Liepert J. No effect of
  Neuroprostheses for grasping. Neurol Res              a levodopa single dose on motor
  2002,24(5):443-452.                                   performance and motor excitability in chronic
                                                        stroke. A double-blind placebo-controlled
Popovic MB, Popovic DB, Sinkjaer T, Stefanovic          cross-over pilot study. Restor Neurol
  A, Schwirtlich L. Clinical evaluation of              Neurosci 2007;25:143-150
  Functional Electrical Therapy in acute
  hemiplegic subjects. J Rehabil Res Dev              Riccio I, Iolascon G, Barillari MR, Gimigliano R,
  2003;40:443-453.                                       Gimigliano F. Mental practice is effective in
                                                         upper limb recovery after stroke: a
Popovic DB, Popovic MB, Sinkjaer T, Stefanovic           randomized single-blind cross-over study.
  A, Schwirtlich L. Therapy of paretic arm in            Eur J Phys Rehabil Med 2010;46:19-25.
  hemiplegic subjects augmented with a neural
  prosthesis: a cross-over study. Can J Physiol       Richards L, Pohl P. Therapeutic interventions to
  Pharmacol 2004;82:749-756.                             improve upper extremity recovery and
                                                         function. Clin Geriatr Med 1999: 15: 819-
Post MW, Visser- Meily JM, Boomkamp-Koppen               832.
  HG, Prevo AJ. Assessment of oedema in
  stroke patients: comparison of visual               Richards L, Rothi LJG, Davis S, Wu AA, Nadeau
  inspection by therapists and volumetric                SE. Limited dose response to Constraint-
  assessment. Disabil Rehabil 2003;25:1265-              Induced Movement therapy in patients with
  1270.                                                  chronic stroke. Clin Rehabil 2006;20: 1066-
Potisk KP, Gregorič M, Vodovik L, Ing D. Effects
  of Transcutaneous Electrical Nerve                  Rijntjes M, Hobbeling V, Hamzei F, et al.
  Stimulation (TENS) on Spasticity in Patients           Individual factors in constraint-induced
  with Hemiplegia. Scand J Rehab Med                     movement therapy after stroke. Neurorehabil
  1995;27:169-174.                                       Neural Repair 2005;19:238-249.

Powell J, Pandyan AD, Granat M, Cameron M,            Risedal A, Zeng J and Johansson BB. Early
  Stott DJ. Electrical stimulation of wrist              training may exacerbate brain damage after
  extensors in poststroke hemiplegia. Stroke             focal brain ischemia in the rat. Journal of
  1999;30(7):1384-1389.                                  Cerebral Blood Flow and Metabolism

10. Upper Extremity Interventions                                                       pg. 144 of 150
Ro T, Noser E, Boake C, Johnson R, Gaber M,           Saposnik G, Teasell R, Mamdani M, Hall J,
  Speroni A, Bernstein M, De Joya A, Burgin             McIlroy W, Cheung D, Thorpe KE, Cohen LG,
  WS, Zhang L, Taub E, Grotta JC, Levin HS.             Bayley M. Effectiveness of virtual reality
  Functional Reorganization and Recovery After          using Wii gaming technology in stroke
  Constraint-Induced Movement Therapy in                rehabilitation: a pilot randomized clinical trial
  Subacute Stroke: Case Reports. Neurocase              and proof of principle. Stroke 2010;41:1477-
  2006;12:50-60.                                        1484.

Robbins SM, Houghton PE, Woodbury G, Brown            Sawaki L, Wu CW, Kaelin-Lang A, Cohen LG.
  JL. The Therapeutic Effect of Functional and          Effects of somatosensory stimulation on use-
  Transcutaneous Electrical Stimulation on              dependent plasticity in chronic stroke. Stroke
  Improving Gait Speed in Stroke Patients: A            2006;37:246-247.
  Meta-Analysis. Arch Phys Med Rehabil 2006;
  87:853-859.                                         Sawaki L, Butler AJ, Xiaoyan L, Wassenaar PA,
                                                        Mohammad YM, Blanton S, Sathian K,
Rodgers H, Mackintosh J, Price C, Wood R,               Nichols-Larsen DS, Wolf SL, Good DC,
  McNamee P, Fearon T, Marritt A, Curless R.            Wittenberg GF. Constraint-induced
  Does an early increased-intensity                     movement therapy results in increased
  interdisciplinary upper limb therapy                  motor map area in subjects 3 to 9 months
  programme following acute stroke improve              after stroke. Neurorehabil Neural Repair
  outcome? Clin Rehabil 2003;17(6):579-89.              2008;22:505-513.

Roper TA, Redford S, Tallis RC. Intermittent          Schabrun SM, Hillier S. Evidence for the
  compression for the treatment of the                  retraining of sensation after stroke: a
  oedematous hand in hemiplegic stroke: a               systematic review. Clin Rehabil 2009;23:27-
  randomized controlled trial. Age Ageing               39.
                                                      Schmidt RA, Wrisberg CA Motor Learning and
Rosman l, Johansson B. Can electroacupunture            Performance. From Principles to Practice,
  or transcutaneous nerve stimulation                   1999, 2nd edition. Champaign, Illinois:
  influence cognition and emotional outcome             Human Kinetics.
  after stroke. J Rehabil Med 2006;38:13-19.
                                                      Sheehan JL, Winzeler-Merçay U, Mudie MH. A
Rose V, Shah S. A comparative study on the              randomized controlled pilot study to obtain
  immediate effects of orthoses in reduction of         the best estimate of the size of the effect of
  hypertonus. Australian Occupational Therapy           a thermoplastic resting splint on spasticity in
  Journal 1987;34,59-64.                                the stroke-affected wrist and fingers. Clin
                                                        Rehabil 2006;20:1032-1037.
Ross LF, Harvey LA, Lannin NA. Do people with
  acquired brain impairment benefit from              Simpson DM, Alexander DN, O'Brien CF, Tagliati
  additional therapy specifically directed at the       M, Aswad AS, Leon JM, Gibson J, Mordaunt
  hand? A randomized controlled trial. Clin             JM, Monaghan EP. Botulinum toxin type A in
  Rehabil 2009;23:492-503.                              the treatment of upper extremity spasticity:
                                                        a randomized, double-blind, placebo-
Rosser N, Heuschmann P, Wersching H,                    controlled trial. Neurology 1996;46:1306-
  Breitenstein C, Knecht S, Floel A. Levodopa           1310.
  improves procedural motor learning in
  chronic stroke patients. Arch Phys Med              Simpson DM, Gracies JM, Yablon SA, Barbano
  Rehabil 2008;89:1633-1641.                            R, Brashear A. Botulinum neurotoxin versus
                                                        tizanidine in upper limb spasticity: a placebo-
Sanchez RJ, Liu J, Rao S, Shah P, Smith R,              controlled study. J Neurol Neurosurg
  Rahman T, Cramer SC, Bobrow JEB,                      Psychiatry 2009;80:380-385.
  Reinkensmeyer DJ. Automating Arm
  Movement training Following Severe Stroke:          Sirtori V, Corbetta D, Moja L, Gatti R.
  Fucntional Exercises With Quantitative                 Constraint-induced movement therapy for
  Feedback in a Gravity-Reduced Environment.             upper extremities in stroke patients.
  IEEE Trans Neural Syst Rehabil Eng                     Cochrane Database Syst Rev
  2006;14:378-389.                                       2009;CD004433.

10. Upper Extremity Interventions                                                       pg. 145 of 150
Sisto SA, Forrest GF, Glendinning D. Virtual           Stewart KC, Cauraugh JH, Summers JJ.
   reality applications for motor rehabilitation         Bilateral movement training and stroke
   after stroke. Top Stroke Rehabil.                     rehabilitation: a systematic review and meta-
   2002;8(4):11-23.                                      analysis. J Neurol Sci 2006;244:89-95.

Slawek J, Bogucki A, Reclawowicz D. Botulinum          Stinear CM, Barber PA, Coxon JP, Fleming MK,
   toxin type A for upper limb spasticity                 Byblow WD. Priming the motor system
   following stroke: an open-label study with             enhances the effects of upper limb therapy in
   individualised, flexible injection regimens.           chronic stroke. Brain 2008;131:1381-1390.
   Neurol Sci 2005; 26(1):32-39.
                                                       Stoykov ME, Lewis GN, Corcos DM. Comparison
Smith KN (1979) Biofeedback in strokes.                  of bilateral and unilateral training for upper
  Australian Journal of Physiotherapy 25: 155-           extremity hemiparesis in stroke.
  61.                                                    Neurorehabil Neural Repair 2009;23:945-
Smith SJ, Ellis E, White S, Moore AP. A double-
  blind placebo-controlled study of botulinum          Subramanian SK, Massie CL, Malcolm MP, Levin
  toxin in upper limb spasticity after stroke or         MF. Does provision of extrinsic feedback
  head injury. Clin Rehabil 2000;14:5-13.                result in improved motor learning in the
                                                         upper limb poststroke? A systematic review
Smith GV, Silver KH, Goldberg AP, Macko RF.              of the evidence. Neurorehabil Neural Repair
  "Task-oriented" exercise improves hamstring            2010;24:113-124.
  strength and spastic reflexes in chronic
  stroke patients. Stroke 1999;30(10):2112-8.          Sullivan JE, Hedman LD. Effects of home-based
                                                         sesory and motor amplitude electrical
Sonde L, Gip C, Fernaeus SE, Nilsson CG,                 stimulation on arm dysfunction in chronic
  Viitanen M. Stimulation with low frequency             stroke. Clin Rehabil 2007;21:142-150.
  (1.7 Hz) transcutaneous electric nerve
  stimulation (low-tens) increases motor               Summers JJ, Kagerer FA, Garry MI, Hiraga CY,
  function of the post-stroke paretic arm.               Loftus A, Cauraugh JH. Bilateral and
  Scand J Rehabil Med 1998;30(2):95-99.                  unilateral movement training on upper limb
                                                         function in chronic stroke patients: A TMS
Sonde L, Kalimo H, Fernaeus SE, Viitanen M.              study. J Neurol Sci 2007;252:76-82.
  Low TENS treatment on post-stroke paretic
  arm: a three-year follow-up. Clin Rehabil            Sun SF, Hsu CW, Sun HP, Hwang CW, Yang CL,
  2000;14(1):14-19.                                      Wang JL. Combined botulinum toxin type A
                                                         with modified constraint-induced movement
Stamenova P, Koytchev R, Kuhn K, Hansen C,               therapy for chronic stroke patients with
  Horvath F, Ramm S et al. A randomized,                 upper extremity spasticity: a randomized
  double-blind, placebo-controlled study of the          controlled study. Neurorehabil Neural Repair
  efficacy and safety of tolperisone in spasticity       2010;24:34-41.
  following cerebral stroke. Eur J Neurol 2005;
  12(6):453-461.                                       Sunderland A, Tinson DJ, Bradley EL, Fletcher
                                                         D, Langton Hewer R, Wade DT. Enhanced
Stein J, Narendran K, McBean J, Krebs K,                 physical therapy improves recovery of arm
  Hughes R. Electromyography-controlled                  function after stroke. A randomised
  exoskeletal upper-limb—powered orthosis for            controlled trial. J Neurol Neurosurg
  exercise training after stroke. Am J Phys Med          Psychiatry 1992;55(7):530-5.
  Rehabil 2007;86:255-261.
                                                       Sunderland A, Fletcher D, Bradley L, Tinson D,
Sterr A, Elbert T, Berthold I et al. Longer versus       Hewer RL, Wade DT. Enhanced physical
  shorter daily constraint-induced movement              therapy for arm function after stroke: a one
  therapy of chronic hemiparesis: an                     year follow up study. J Neurol Neurosurg
  exploratory study. Arch Phys Med Rehabil               Psychiatry 1994;57:856-858.
  2002; 83(10):1374-1377.
                                                       Suputtitada A, Suwanwela NC, Tumvitee S.
Steultjens EM, Dekker J, Bouter LM, van de Nes           Effectiveness of constraint-induced
  JC, Cup EH, van den Ende CH. Occupational              movement therapy in chronic stroke
  therapy for stroke patients: a systematic              patients. J Med Assoc Thai 2004;87:1482-
  review. Stroke 2003;34:676-87.                         1490.

10. Upper Extremity Interventions                                                       pg. 146 of 150
Suputtitada A, Suwanwela NC. The lowest                 Movement therapy. Phys Med Rehabil Clin N
  effective dose of botulinum A toxin in adult          Am 2003;14(1 Suppl):S77-91, ix.
  patients with upper limb spasticity. Disabil        Taub E, Uswatte G. The case for CI therapy. J
  Rehabil 2005;27:176-184.                              Rehab Res Dev 2003;40:xiii-xv.

Szaflarski JP, Page SJ, Kissela BM, Lee JH,           Taub E, Lum PS, Hardin P, Mark VW, Uswatte
  Levine P, Strakowski SM. Cortical                     G. AutoCITE: automated delivery of CI
  Reorganization Following Modified Constrain-          therapy with reduced effort by therapists.
  Induced Movement Therapy: A Study of 4                Stroke 2005;36:1301-1304.
  Patients With Chronic Stroke. Arch Phys Med
  Rehabil 2006;87:1052-1058.                          Taub E, Uswatte G, Kay King D, Morris D, Crago
                                                        JE, Chatterjee A. A Placebo-Controlled Trial
Tardy J, Pariente J, Leger A, Dechaumont-               of Constraint-Induced Movement Therapy for
  Palacin S, Gerdelat A, Guiraud V, Conchou F,          Upper Extremity After Stroke. Stroke
  Albucher JF, Marque P, Franceries X,                  2006;37:1045-1049.
  Cognard C, Rascol O, Chollet F, Loubinoux I.
  Methylphenidate modulates cerebral post-            Tekeoglu Y, Adak B, Goksoy T. Effect of
  stroke reorganization. Neuroimage                     transcutaneous electrical nerve stimulation
  2006;33:913-922.                                      (TENS) on Barthel Activities of Daily Living
                                                        (ADL) index score following stroke. Clin
Tariq M, Akhtar N, Ali M, Rao S, Badshah M,             Rehabil 1998;12(4):277-280.
  Irshad M. Eperisone compared to
  physiotherapy on muscular tone of stroke            Thielman GT, Dean CM, Gentile AM.
  patients: a prospective randomized open               Rehabilitation of reaching after stroke: task-
  study. J Pak Med Assoc 2005; 55(5):202-               related training versus progressive resistive
  204.                                                  exercise. Arch Phys Med Rehabil
Tarkka IM, Pitkanen K, Sivenius J. Paretic hand
  rehabilitation with constraint-induced              Thrasher TA, Zivanovic V, McIlroy W, Popovic
  movement therapy after stroke. Am J Phys              MR. Rehabilitation of reaching and grasping
  Med Rehabil 2005;84:501-505.                          function in severe hemiplegic patients using
                                                        functional electrical stimulation therapy.
Taub E. Somatosensory deafferentation                   Neurorehabil Neural Repair 2008;22:706-
  research with monkeys: Implications for               714.
  rehabilitation medicine. In: LP Ince (Ed.),
  Behavioral Psychology in Rehabilitation             Trombly CA, Thayer-Nason L, Bliss G, Girard
  Medicine: Clinical Applications (pp.371-401).         CA, Lyrist LA, Brexa-Hooson A. The
  Baltimore: Williams and Wilkins, 1980.                effectiveness of therapy in improving finger
                                                        extension in stroke patients. Am J Occup
Taub E, Miller NE, Novack TA, Cook EW,                  Ther 1986;40:612-617.
  Fleming WC, Nepomuceno CS, Connell JS,
  Crago JE. Technique to improve chronic              Tseng CN, Chen CC, Wu SC, Lin LC. Effects of a
  motor deficit after stroke. Arch Phys Med             range-of-motion exercise programme. J Adv
  Rehabil 1993;74:347-354.                              Nurs 2007;57:181-191.

Taub E, Morris DM. Constraint-induced                 Turczynski BE, Harte W, Sturm W.
  movement therapy to enhance recovery after            Electromygraphic Feedback Treatment of
  stroke. Curr Atheroscler Rep 2001;3(4):279-           Chronic Hemiparesis: An attempt to Quantify
  86.                                                   Treatment Effects. Arch Phys Med Rehabil
Taub E, Uswatte G, Pidikiti R. Constraint-
  Induced Movement Therapy: a new family of           Turner-Stokes L, Baguley IJ, De GS, Katrak P,
  techniques with broad application to physical         Davies L, McCrory P, Hughes A. Goal
  rehabilitation--a clinical review. J Rehabil Res      attainment scaling in the evaluation of
  Dev 1999;36(3):237-51. Review.                        treatment of upper limb spasticity with
                                                        botulinum toxin: a secondary analysis from a
Taub E, Uswatte G, Morris DM. Improved motor            double-blind placebo-controlled randomized
  recovery after stroke and massive cortical            clinical trial. J Rehabil Med 2010;42:81-89.
  reorganization following Constraint-Induced

10. Upper Extremity Interventions                                                      pg. 147 of 150
Turton A, Fraser C. The use of home therapy           van Vliet PM, Wulf G. Extrinsic feedback for
  programmes for improving recovery of the              motor learning after stroke: what is the
  upper limb following stroke. Br J Occup Ther          evidence? Disabil Rehabil 2006;28:831-840.
  1990; 53: 457-62.
                                                      van Vliet PM, Lincoln NB, Foxall A. Comparison
Turton A, Pomeroy V. When should upper limb             of Bobath based and movement science
  function be trained after stroke? Evidence for        based treatment for stroke: a randomised
  and against early intervention.                       controlled trial. J Neurol Neurosurg
  NeuroRehabilitation 2002;17(3):215-24.                Psychiatry 2005;76:503-508.

Turton AJ, Britton E. A pilot randomized              Verbunt JA, Seelen HA, Ramos FP, Michielsen
  controlled trial of a daily muscle stretch            BH, Wetzelaer WL, Moennekens M. Mental
  regime to prevent contractures in the arm             practice-based rehabilitation training to
  after stroke. Clin Rehabil 2005;19:600-612.           improve arm function and daily activity
                                                        performance in stroke patients: a
Underwood J, Clark PC, Blanton S, Ayick DM,             randomized clinical trial. BMC Neurol
  Wolf SL. Pain, Fatigue and Intensity of               2008;8:7
  Practice in People With Stroke Who Are
  Receiving Constraint-Induced Movement               Volpe BT, Krebs HI, Hogan N, Edelsteinn L,
  Therapy. Physical Therapy 2006;86:1241-               Diels CM, Aisen ML. Robot training enhanced
  1250.                                                 motor outcome in patients with stroke
                                                        maintained over 3 years. Neurology
Uswatte G, Taub E, Morris D, Barman J, Crago            1999,53(8):1874-1876.
  J. Contribution of the shaping and restraint
  components of Constraint-Induced                    Volpe BT, Krebs HI, Hogan N, Edelstein OL,
  Movement therapy to Treatment Outcome.                Diels C, Aisen M. A novel approach to stroke
  Neurorehabilitation 2006;21:147-156.                  rehabilitation: robot-aided sensorimotor
                                                        stimulation. Neurology 2000,54(10):1938-
van der Lee JH, Wagenaar RC, Lankhorst GJ,              1944.
  Vogelaar TW, Deville WL, Bouter LM. Forced
  use of the upper extremity in chronic stroke        Volpe BT, Ferraro M, Lynch D, et al. Robotics
  patients: results from a single-blind                 and other devices in the treatment of
  randomized clinical trial. Stroke                     patients recovering from stroke. Curr
  1999;30:2369-2375.                                    Atheroscler Rep 2004;6:314-319.

van der Lee JH, De G, V, Beckerman H,                 Volpe BT, Lynch D, Rykman-Berland A, et al.
  Wagenaar RC, Lankhorst GJ, Bouter LM. The             Intensive sensorimotor arm training
  intra- and interrater reliability of the action       mediated by therapist or robot improves
  research arm test: a practical test of upper          hemiparesis in patients with chronic stroke.
  extremity function in patients with stroke.           Neurorehabil Neural Repair 2008;22:305-310
  Arch Phys Med Rehabil 2001;82:14-19.
                                                      Wagenaar RC, Meijer OG, van Wieringen PC,
van der Lee JH. Constraint-induced therapy for          Kuik DJ, Hazenberg GJ, Lindeboom J,
  stroke: more of the same or something                 Wichers F, Rijswijk H. The functional
  completely different? Curr Opin Neurol                recovery of stroke: a comparison between
  2001;14:741-744.                                      neuro-developmental treatment and the
                                                        Brunnstrom method. Scand J Rehabil Med
van Kuijk AA, Geurts AC, Bevaart BJ, van                1990;22(1):1-8.
  Limbeek J. Treatment of upper extremity
  spasticity in stroke patients by focal neuronal     Wang RY, Yang YR, Tsai MW, Wang WT, Chan
  or neuromuscular blockade: a systematic               RC. Effects of functional electric stimulation
  review of the literature. J Rehabil Med               on upper limb motor function and shoulder
  2002;34(2):51-61.                                     range of motion in hemiplegic patients. Am J
                                                        Phys Med Rehabil. 2002;81(4):283-90.
Van Peppen RP, Kwakkel G, Wood-Dauphinee
  S, Hendriks HJ, Van der Wees PJ, Dekker J.          Wang RY, Chan RC, Tsai MW. Functional
  The impact of physical therapy on functional          electrical stimulation on chronic and acute
  outcomes after stroke: what's the evidence?           hemiplegic shoulder subluxation. Am J Phys
  Clin Rehabil 2004;18:833-862.                         Med Rehabil. 2000;79(4):385-90.

10. Upper Extremity Interventions                                                      pg. 148 of 150
Watkins CL, Leathley MJ, Gregson JM, Moore             Wolf SL, Lecraw DE, Barton LA, Jann BB. Forced
  AP, Smith TL, Sharma AK. Prevalence of                 use of hemiplegic upper extremities to
  spasticity post stroke. Clin Rehabil                   reverse the effect of learned nonuse among
  2002,16(5):515-522.                                    chronic stroke and head-injured patients.
                                                         Exp Neurol 1989;104:125-132.
Weber DJ, Skidmore ER, Niyonkuru C, Chang
  CL, Huber LM, Munin MC. Cyclic functional            Wolf SL, Winstein CJ, Miller JP, Taub E, Uswatte
  electrical stimulation does not enhance gains          G, Morris D, Giuliani C, Light KE, Nichols-
  in hand grasp function when used as an                 Larsen D. Effect of Constraint-Induced
  adjunct to onabotulinumtoxinA and task                 Movement Therapy on Upper Extremity
  practice therapy: a single-blind, randomized           Function 3 to 9 months after stroke. JAMA
  controlled pilot study. Arch Phys Med Rehabil          2006; 296: 2095-2104.
                                                       Wolf SL, Winstein CJ, Miller JP, et al. Retention
Whitall J, McCombe WS, Silver KH, Macko RF.              of upper limb function in stroke survivors
  Repetitive bilateral arm training with                 who have received constraint-induced
  rhythmic auditory cueing improves motor                movement therapy: the EXCITE randomised
  function in chronic hemiparetic stroke.                trial. Lancet Neurol 2008;7:33-40.
  Stroke 2000; 31:2390-2395.
                                                       Wolf SL, Thompson PA, Winstein CJ, Miller JP,
Winstein CJ, Merians AS, Sullivan KJ. Motor              Blanton SR, Nichols-Larsen DS, Morris DM,
  learning after unilateral brain damage.                Uswatte G, Taub E, Light KE, Sawaki L. The
  Neuropsychologia 1999;37:975-987.                      EXCITE Stroke Trial. Comparing Early and
                                                         Delayed Constraint-Induced Movement
Wisntein CJ, Rose DK. Recovery and arm use               Therapy. Stroke 2010 (e-publ).
  after stroke. J Cerebrovasc Dis 2001;10:197.
                                                       Woodbury ML, Howland DR, McGuirk TE, Davis
Winstein CJ, Rose DK, Tan SM, Lewthwaite R,              SB, Senesac CR, Kautz S, Richards LG.
  Chui HC, Azen SP. A randomized controlled              Effects of trunk restraint combined with
  comparison of upper-extremity rehabilitation           intensive task practice on poststroke upper
  strategies in acute stroke: A pilot study of           extremity reach and function: a pilot study.
  immediate and long-term outcomes. Arch                 Neurorehabil Neural Repair 2009;23:78-91.
  Phys Med Rehabil 2004;85:620-628.
                                                       Wu CW, Sea HJ, Cohen LG. Influence of Electric
Wittenberg GF, Chen R, Ishii K, Bushara KO,              Somatosenory Stimulation on Paretic-Hand
  Eckloff S, Croarkin E, Taub E, Gerber LH,              Function in Chronic Stroke. Arch Phys Med
  Hallett M, Cohen LG. Constraint-induced                Rehabil 2006;87:351-357.
  therapy in stroke: magnetic-stimulation
  motor maps and cerebral activation.                  Wu A, Chen C, Tsai W, Lin K, Chou S. A
  Neurorehabil Neural Repair 2003;17(1):48-              randomized controlled trial of modified
  57.                                                    constraint-induced movement therapy for
                                                         elderly stroke survivors: changes in motor
Woldag H, Hummelsheim H. Evidence-based                  impairment, daily functioning, and quality of
  physiotherapeutic concepts for improving               life. Arch Phys Med Rehabil 2007a);88:273-
  arm and hand function in stroke patients: a            278.
  review. J Neurol 2002;249(5):518-28.
                                                       Wu CY, Lin KC, Chen HC, Chen IH, Hong WH.
Woldag H, Waldmann G, Heuschkel G,                       Effects of modified constraint-induced
  Hummelsheim H. Is the repetitive training of           movement therapy on movement kinematics
  complex hand and arm movements beneficial              and daily function in patients with stroke: a
  for motor recovery in stroke patients? Clin            kinematic study of motor control
  Rehabil 2003;17:723-730.                               mechanisms. Neurorehabil Neural Repair
Wolf SL, Binder-MacLeod SA. Electromyographic
  biofeedback applications to the hemiplegic           Wu CY, Chen CL, Tang SF, Lin KC, Huang YY.
  patient. Changes in upper extremity                    Kinematic and clinical analyses of upper-
  neuromuscular and functional status. Phys              extremity movements after constraint-
  Ther 1983;63:1393-1403.                                induced movement therapy in patients with
                                                         stroke: a randomized controlled trial. Arch
                                                         Phys Med Rehabil 2007c);88:964-970.

10. Upper Extremity Interventions                                                        pg. 149 of 150
Yavuzer G, Selles R, Sezer N, et al. Mirror
  therapy improves hand function in subacute
  stroke: a randomized controlled trial. Arch
  Phys Med Rehabil 2008;89:393-398.

Yavuzer G, Senel A, Atay MB, Stam HJ.
  ''Playstation eyetoy games'' improve upper
  extremity-related motor functioning in
  subacute stroke: a randomized controlled
  clinical trial. Eur J Phys Rehabil Med

Yekutiel M, Guttman E. A controlled trial of the
  retraining of the sensory function of the hand
  in stroke patients. J Neurol Neurosurg
  Psychiatry 1993;56:241-244.

Yozbatiran N, Donmez B, Kayak N, Bozan O.
  Electrical stimulation of wrist and fingers for
  sensory and functional recovery in acute
  hemiplegia. Clin Rehabil 2006;20:4-11.

You SH, Jang SH, Kim Y, Hallett M, Ahn SH,
  Kwon Y, Kim JH, Young Lee M. Virtual
  Reality-Induced Cortical Reorganization and
  Associated Locomotor Recovery in Chronic
  Stroke. Stroke 2005;36 :1166-1171.

Zimmermann-Schlatter A, Schuster C, Puhan
  MA, Siekierka E, Steurer J. Efficacy of motor
  imagery in post-stroke rehabilitation: a
  systematic review. J Neuroeng Rehabil

Zittel S, Weiller C, Liepert J. Reboxetine
   improves motor function in chronic stroke. A
   pilot study. J Neurol 2007;254:197-201.

Zittel S, Weiller C, Liepert J. Citalopram
   improves dexterity in chronic stroke patients.
   Neurorehabil Neural Repair 2008;22:311-

10. Upper Extremity Interventions                            pg. 150 of 150