Archives of Physical Medicine and Rehabilitation

journal homepage: www.archives-pmr.org

Archives of Physical Medicine and Rehabilitation 2018;99:834-42

ORIGINAL RESEARCH

Virtual Reality for Upper Limb Rehabilitation in

Subacute and Chronic Stroke: A Randomized
Controlled Trial
Pawel Kiper, PhD,a Andrzej Szczudlik, PhD,b Michela Agostini, MSc,a Jozef Opara, PhD,c
Roman Nowobilski, PhD,b Laura Ventura, PhD,d Paolo Tonin, MD,a Andrea Turolla, PhDa,e
From the aLaboratory of Kinematics and Robotics, IRCCS San Camillo Hospital Foundation, Venice, Italy; bDepartment of Neurology,
Jagiellonian University Medical College, Krakow, Poland; cDepartment of Physiotherapy, Academy of Physical Education, Katowice, Poland;
d
Department of Statistical Sciences, University of Padova, Padua, Italy; and eDepartment of Neuroscience, The University of Sheffield,
Sheffield, UK.

Abstract
Objective: To evaluate the effectiveness of reinforced feedback in virtual environment (RFVE) treatment combined with conventional
rehabilitation (CR) in comparison with CR alone, and to study whether changes are related to stroke etiology (ie, ischemic, hemorrhagic).
Design: Randomized controlled trial.
Setting: Hospital facility for intensive rehabilitation.
Participants: Patients (NZ136) within 1 year from onset of a single stroke (ischemic: nZ78, hemorrhagic: nZ58).
Interventions: The experimental treatment was based on the combination of RFVE with CR, whereas control treatment was based on the same
amount of CR. Both treatments lasted 2 hours daily, 5d/wk, for 4 weeks.
Main Outcome Measures: Fugl-Meyer upper extremity scale (F-M UE) (primary outcome), FIM, National Institutes of Health Stroke Scale
(NIHSS), and Edmonton Symptom Assessment Scale (ESAS) (secondary outcomes). Kinematic parameters of requested movements included
duration (time), mean linear velocity (speed), and number of submovements (peak) (secondary outcomes).
Results: Patients were randomized in 2 groups (RFVE with CR: nZ68, CR: nZ68) and stratified by stroke etiology (ischemic or hemorrhagic).
Both groups improved after treatment, but the experimental group had better results than the control group (Mann-Whitney U test) for F-M UE
(P<.001), FIM (P<.001), NIHSS (P.014), ESAS (P.022), time (P<.001), speed (P<.001), and peak (P<.001). Stroke etiology did not have
significant effects on patient outcomes.
Conclusions: The RFVE therapy combined with CR treatment promotes better outcomes for upper limb than the same amount of CR, regardless
of stroke etiology.
Archives of Physical Medicine and Rehabilitation 2018;99:834-42
ª 2018 by the American Congress of Rehabilitation Medicine

Stroke is the third cause of death and the first cause of disability
worldwide,1 regardless of ethnicity and citizenship, dramatically
affecting the independence and quality of life of survivors.
Impairment of motor function is commonly present after stroke
because of the involvement of the brain areas responsible for
planning and execution of movements (eg, primary motor cortex,
anterior frontal lobe).2,3 Recent studies have also shown that other
Clinical Trial Registration No.: NCT01955291.
Disclosures: none.
functions, mainly represented in the posterior parietal cortex, are
involved in decision-making for the production of voluntary
movements (eg, spatial attention, spatial awareness, multisensory
integration of feedback).4 Another important factor which in-
terferes with performance of motor tasks after stroke is the
severity of associated cognitive impairments.5 Regardless of
stroke etiology (ie, ischemic, hemorrhagic), clinical sequelae are
devastating. Therefore, rehabilitation is provided to patients
regardless of onset. Nevertheless, evidence has shown that patients
gain greater functional improvement after intracerebral hemor-
rhage (10%e15% of all strokes), than after ischemic infarct.6 This

0003-9993/18/$36 - see front matter ª 2018 by the American Congress of Rehabilitation Medicine
https://doi.org/10.1016/j.apmr.2018.01.023
Reinforced feedback in virtual environment
finding was also observed in patients with hemorrhagic stroke
treated in a virtual environment in the preliminary study of this
clinical trial.7
The main objective of motor rehabilitation after stroke is to
promote movement learning8 with the aim to reduce the effect of
disease on loss of function and to improve quality of life. Recent
advances to improve the efficacy of rehabilitation modalities
suggest to include the principles of motor learning in feasible
clinical treatments.9 Therefore, recent findings have demon-
strated that repetitive, intensive, and random practice of func-
tional tasks may be required to modify neural structures involved
in motor control and learning processes.1,10 A significant boost
to promote neuroplasticity during rehabilitation care came from
the use of innovative computer-based technologies (eg, virtual
reality, robotics).11-14 These technology-based modalities enrich
the traditional rehabilitation setting by increasing specific feed-
back on results and motor performance. Many studies have
shown that augmented feedback is able to modulate sensori-
motor cortical activity, by facilitating learning in selected brain
networks.15
Virtual reality acts as an augmented environment where feed-
back can be delivered in the form of enhanced information about
knowledge of results and knowledge of performance.7 Regarding
clinical applications for rehabilitation, the translation of functional
skills during the recovery process can be promoted by adapting the
complexity of requested motor tasks to the functional level of each
single patient.9 Several studies demonstrated that treatment
enriched by reinforced feedback in virtual environment (RFVE)
may be more effective than conventional rehabilitation (CR) to
improve motor function of the upper limb after stroke.7,11,16 RFVE
may contain elements useful to maximizing movement learning (eg,
providing repeated and varied task practice, augmenting feedback
on achieved results). RFVE has been tested in clinical trials
providing insights into specific stroke rehabilitation and benefits
that can be obtained by using enriched virtual environment.17-19
Although several studies have been conducted on this topic, there
is still a lack of large clinical and randomized trials that involve and
assess different stroke etiologies in relation to augmented virtual
feedback. Therefore, to overstep this literature limitation, we
compared effects on patients with ischemic stroke with effects on
patients with hemorrhagic stroke. There is also deficiency in the
literature on the potential of combining virtual therapy with other
treatments and its influence on poststroke recovery. Therefore, this
aspect has also been analyzed in this study.
In this trial, we hypothesized that performing a movement in
enhanced virtual environment could have significant effect on
upper limb motor function and that stroke etiology could be a
factor influencing the recovery process.
The aims of this study were to explore whether the RFVE
intervention, combined with CR, could improve motor function
more than the same amount of CR applied alone, after both
ischemic and hemorrhagic stroke, and to observe whether the type
List of abbreviations:
CR conventional rehabilitation
ESAS Edmonton Symptom Assessment scale
F-M UE Fugl-Meyer upper extremity scale
NIHSS National Institutes of Health Stroke Scale
RFVE reinforced feedback in virtual environment
VRRS Virtual Reality Rehabilitation System
www.archives-pmr.org
835
of stroke (ie, ischemic, hemorrhagic) may influence recov-
ery gains.
Methods
Participants and design
A single-blind, randomized controlled trial was completed among
inpatients affected by a first episode of stroke because of both
ischemic and hemorrhagic etiologies (fig 1). Subjects who had a
stroke onset up to 12 months before enrollment were included in the
study. The following criteria were considered for exclusion: cognitive
impairment (defined as score <24 on the Mini-Mental State Exam-
ination),20 having previously received RFVE treatment, presence of
apraxia (defined as a score <62 points on the De Renzi test),21
impairment of verbal comprehension (defined as a score >40 errors
on the Token test),22 evidence in clinical history of neglect, upper
extremity complete hemiplegia (score of 4 points on the upper limb
subitem of the National Institutes of Health Stroke Scale [NIHSS]),23
presence of upper limb sensory disorders (defined as <1 point on
items shoulder, elbow, wrist, and thumb in the sensitivity section of
the Fugl-Meyer scale), fracture, and joint dislocation.
The internal review board of the hospital approved the study,
and all patients were informed about the aim and experimental
procedures before enrollment; therefore, written informed consent
was obtained from all of them. Individuals included in the study
were randomly assigned to experimental or control group accord-
ing to a simple randomization technique (computerized random
numbers). The randomization sequence was generated at the start
of the trial using a computerized program (Microsoft Excela). The
allocation sequence was concealed from the principal investigator
enrolling patients in sequentially numbered, opaque, sealed enve-
lopes. The researcher responsible for randomization was indepen-
dent from the assessors, assuring blindness to treatment allocation
and randomization procedures, respectively. Because of the treat-
ment modality, it was not possible to blind patients regarding
treatment received. In the experimental group, patients received
treatment based on RFVE, whereas in the control group CR was
provided. Moreover, both groups received an additional hour per
day of CR. In this regard, the amount of therapy, defined as hours
of treatment provided, was comparable between groups (ie, 2h/d,
5d/wk for 4wk: 40h overall), but the therapeutic modalities (ie,
RFVE and CR, twice CR) representing the independent variable
determining the effects in the trial were not. Both therapeutic
modalities were provided by physiotherapists employed in the
institute and trained for RFVE and CR application. Moreover, each
single treatment was executed by different clinicians blind to study
aims. The CR treatment consisted of whole-body exercises which
were selected autonomously by the clinician. The additional hour
consisted of RFVE or CR. The exercises in the RFVE group were
selected from the library of the Virtual Reality Rehabilitation
System (VRRS) programb or created ad hoc, whereas exercises in
the CR group were replicated from the virtual scenarios. However,
both treatments were personalized and coherent to the patient’s
condition. Therefore, not all the tasks and movements were
matched in the 2 treatments (ie, CR, RFVE). All patients enrolled
in this study were inpatients of the rehabilitation institute and were
treated during their stay in the hospital. According to Italian health
policies, it is regular practice to admit patients >1 year after stroke
to rehabilitation hospitals, if they still present with rehabilita-
tion needs.
836 P. Kiper et al

Screened potentially eligible patients

with stroke (N = 205)

Excluded total (n = 66)

• Complete hemiplegia (n = 2)
• Refusing to participate (n = 2)
• Sensory disorders (n = 17)
• Apraxia (n = 10)
• Aphasia (n = 31)
• Other reason (n = 4)

Randomized
> 12 Month (n = 139)
RFVE experimental group CR control group

Randomized to Experimental Group: Control Group: Randomized to

intervention (n= 69) Treatment with Treatment with intervention (n= 70)
reinforced feedback traditional
Received in virtual rehabilitation Received
intervention (n = 68) environment program intervention (n = 68)
• 1 hour of CR and • 2 hours of CR
Drop out (n = 1) 1 hour of RFVE Drop out (n = 2)

Stratified by the kind of stroke

(n = 136)
RFVE experimental group CR control group

RFVE Ischemic group: RFVE Hemorrhagic CR Ischemic group: CR Hemorrhagic

Allocated to
intervention (n= 36)
Received allocated
intervention (n = 36)
Did not receive
allocated intervention
(n = 0)
group: group:
Allocated to
Allocated to intervention (n= 42) Allocated to
intervention (n= 32) intervention (n= 26)
Received allocated
Received allocated intervention (n = 42) Received allocated
intervention (n = 32) intervention (n = 26)
Did not receive
Did not receive allocated intervention Did not receive
allocated intervention (n = 0) allocated intervention
(n = 0) (n = 0)

Responses available for analysis

4 Weeks (n = 136)

Fig 1 Flowchart of participants through the study.

Interventions liquid crystal display projector displaying the virtual scenarios

on a large wall screen.
The patients randomized to the RFVE group were treated with the During the RFVE treatment, the subject was seated in front of
VRRS, which consisted of a computer connected to a a wall screen grasping a sensorized real object (eg, ball, disk,
3-dimensional motion tracking systemc and a high-resolution glass) with the paretic hand, or in case of severe grasping deficits

www.archives-pmr.org
Reinforced feedback in virtual environment
the sensor was fixed to a fingerless glove (with elastic hook and
loop fastener) worn by the patient. The real object held by the
subject and equipped with an electromagnetic sensor was
matched to the virtual handling object, representing the end
effector of the biomechanical system. A library of virtual sce-
narios and tasks was created by physiotherapists involved in
previous studies of our team, based on patients’ motor impair-
ments.17,24-26 Then, the physiotherapists involved in the current
study had the possibility to choose the most appropriate exer-
cises for each patient from the existing library and to create new
virtual scenarios by the editor application of the VRRS program.
The virtual object was displayed in a shape as close as possible
to the real one, and its displacement was synchronized in time
and space between real and virtual environments; therefore, no
delay or spatial distortion affected the emulation in the virtual
environment of the movement executed in the physical space.
For each motor task, the starting and final positions and move-
ment complexity were determined (supplemental appendix S1,
available online only at http://www.archives-pmr.org/). Acoustic
signals and a digital voice provided information on the cor-
rectness of movement execution online (reinforcement learning),
and after each trial feedback was provided in the form of a score
which was proportional to the amount of spatial error made
(knowledge of results). Virtual tasks consisted of both simple
movements (eg, moving a ball by executing elbow flexion in
vertical plane) and complex movements that involved multiple
muscle synergies (eg, pouring water from a glass, using a
hammer, using a toothbrush). The characteristics of the target
were determined in each virtual scenario (eg, task orientation,
position). Moreover, additional virtual objects were added in the
scenario, with the aim to increase task’s complexity. These
objects were not present in the real world or sensorized as the
end effector but were just visually displayed in the virtual
environment. In addition, reinforced visual feedback was visu-
alized in the form of a virtual teacher (ie, nonsensorized virtual
object) with the same shape but different color of virtual rep-
resentation of the end effector, which automatically executed the
correct trajectory (supervised learning). The forward, backward,
or lateral trunk movement was advised by the physiotherapist in
both training groups, and it was limited or favored according to
specific task requirements. At the end of the therapy session, the
physiotherapist showed patients the result of their own move-
ments (knowledge of performance) by visualizing all recorded
trajectories compared with the requested one.
Patients allocated to the CR group were treated in a 1-to-1
setting in the hospital rehabilitation service. Patients received
specific rehabilitation of the upper limb consisting of exercises in
many directions of the upper limb workspace (eg, shoulder flexion
and extension, shoulder abduction and adduction, shoulder inter-
nal and external rotation, elbow flexion and extension, forearm
pronation and supination) and hand grasping-release tasks. In the
CR group, all exercises were performed in the sitting position and
replicated as much as possible coherently to the RFVE tasks, to
maintain comparable exercise content in both treatments. How-
ever, each training program was personalized to patient’s motor
capacities. Therefore, individual task-oriented exercises were
selected from the exercise booklet, for each patient (eg, exercises
for postural control, including reaching movement, exercises of
elbow flexion and extension or shoulder abduction and adduction).
Then, the exercise program was progressively increased in terms
of complexity by the physiotherapist in charge of the treatment
(eg, exercises for hand posture, exercises for manipulation and
www.archives-pmr.org
837
hand grasping, exercises to improve proximal and distal coordi-
nation), according to results from the functional assessment.
Therefore, exercises performed by patients in the CR group were
addressed to achieve the best functional skills according to each
patient’s current capacity.
The main role of the physiotherapist during the therapeutic
session in both training groups was to advise correct movement
execution according to motor learning principles (eg, smoothness
of movement, reduced compensatory behaviors, to favor func-
tional movement).27 The comprehensive time spent for treatment
in both groups was 2 hours. The duration and amount of single
exercise and number of repetitions were always adapted to the
patients’ physical efficiency.
Outcome measures
At enrollment, a detailed review of the medical history of each
patient was collected to check the presence of inclusion/exclusion
criteria. Further, the following outcome measures were assessed:
the Fugl-Meyer upper extremity scale (F-M UE)28 (primary
outcome), which assesses upper limb motor function, balance,
sensitivity, and joint range of movement, determining severity of
impairment; the FIM29; the NIHSS23; and the Edmonton Symp-
tom Assessment Scale (ESAS)30 (secondary outcomes). Further-
more, the following kinematic parameters were measured during
the execution of requested tasks: mean linear velocity (speed),
mean duration of movements (time), and mean number of sub-
movements (peak) (secondary outcomes). The assessment script
was composed of 8 motor tasks covering the main axes of motion
(eg, flexion, abduction, adduction, rotations, pronation, supina-
tion) and involving multiple joints (eg, shoulder, elbow) of the
upper limb. Each task was performed for 10 consecutive trials
(fig 2); therefore, 80 trials overall were recorded for every patient,
whose features represented the kinematic assessment. During the
kinematic assessment, participants were seated with their back
supported and hips and knees flexed to 90 , the hand was placed
on the hip or knee depending on the initial movement position
requested. Prior to assessment, each patient had the possibility to
familiarize themselves with the system by performing 2 to 3 at-
tempts. Afterward, the task performance was recorded and data
from all 80 trials were processed to calculate each kinematic
parameter. All participants performed the kinematic assessment
protocol before and after the treatment, but the same exercises
were never used during the training. Both functional and kine-
matic assessments were carried out for all patients at the begin-
ning and end of treatment (ie, after 4wk).
Statistical analysis
The sample size was calculated based on expected difference be-
tween experimental and control groups on the Fugl-Meyer test
(Student t test) from a previous study.31 The sample size calculation
indicated that 68 patients per group should have been enrolled for
aZ.05 and 1ebZ.95, considering a 10% possible dropout rate,
with the standardized effect size of 3.296. Therefore, the indicated
overall number of 136 participants was achieved in this trial. Data
distribution for all the variables was tested by the Shapiro-Wilk test,
and all of them resulted in being not normally distributed; therefore
the Wilcoxon test for paired samples was used for comparisons
within groups. The Wilcoxon-Mann-Whitney test was used for
comparison between groups both at the baseline and after treatment.
Moreover, patients enrolled were stratified post hoc according to
838 P. Kiper et al

Fig 2 The 8 virtual scenarios representing motor tasks performed during kinematic assessment: (A) reaching on horizontal planedstarting
hand position: one-third of proximal femur on lateral side, movement: toward the knee on ipsilateral side; (B) elbow flexiondstarting hand
position: knee on hemiparetic side, movement: toward the mouth; (C) elbow extension on horizontal planedstarting hand position: hip
contralateral to hemiparetic side, movement: toward hemiparetic knee; (D) reaching movementdstarting hand position: one-third of proximal
femur on dorsal side, movement: forward with elbow extension toward a target; (E) forearm pronation and supinationdstarting position with
hand pronated: contralateral one-half of femur, movement: supination toward ipsilateral one-half of the femur; (F) arm abductiondstarting hand
position: one-third of proximal femur on lateral side, movement: abduction up to 90 ; (G) shoulder flexiondstarting hand position: knee on
hemiparetic side, movement: shoulder flexion up to 90 with straight elbow; and (H) shoulder external and internal rotationdstarting hand
position: one-third of proximal femur on dorsal side, movement: external rotation with elbow flexed at 90 .

their stroke etiology (ie, ischemic, hemorrhagic) in both groups.
Then, a subgroup analysis was run to compare differences because
of the kind of stroke (Wilcoxon-Mann-Whitney test). The
chi-square test was used to analyze the therapies effect related to the
minimal detectable change, which is established at 7.9%. Finally, a
skew-normal regression model was used to infer any potential
relation between dependent and independent variables. Based on
previous studies a gain of 13.6% at the F-M UE was considered as
clinically important.32 The statistical significance level was set
at P<.05.
Results
Among the cohort of patients accepted at the cerebrovascular
disease unit of a rehabilitation institute, 205 patients were
screened for eligibility between years 2008 and 2015: 66 (32%)
did not meet the inclusion criteria and 3 (1.5%) dropped out of the
clinical trial (did not complete the full training sessions). Overall,
136 (66.5%) patients were enrolled (80 men and 56 women), who
survived a first episode of ischemic (nZ78) or hemorrhagic
(nZ58) stroke. The average time since stroke onset was 4.23.0
months, and the mean age was 63.914.1 years. All patients
experienced the RFVE intervention for the first time, and all of
them reported to be comfortable throughout the training. More-
over, none of them experienced any side effects because of
interaction with the virtual environment (eg, nausea, dizziness,
headache, disorientation).33 Demographic characteristics did not
differ significantly at baseline (table 1), nor did the functional and
kinematic outcomes (F-M UE: PZ.080; FIM: PZ.853; NIHSS:
PZ.916; ESAS: PZ.508; time: PZ.210; speed: PZ.603; peak:
PZ.118) (table 2).
Effect of virtual reality and CR on motor function
Functional and kinematic outcomes improved in both groups
(table 2) after treatment, but the gain in the CR group was lower
than in the RFVE group. Indeed, the F-M UE score gained 4.77% in
the CR group and 14.73% in the RFVE group (P<.001). All out-
comes were significantly higher in the RFVE group than the CR
group, with a consistently better gain after treatment (FIM: RFVE:
8.42%, CR: 4.82%, P<.001; NIHSS: RFVE: 2.49%, CR: 1.26%,
PZ.014; ESAS: RFVE: 4.70%, CR: 1.53%, PZ.022; time: RFVE:
97.35%, CR: 26.78%, P<.001; speed: RFVE: 90.14%, CR:
25.60%, P<.001; peak: RFVE: 98.67%, CR: 36.08%, P<.001).
In the RFVE group, 54 out of 68 (79.4%) patients demonstrated
motor improvements bigger than the established minimal detectable
change (7.9%) on the F-M UE, whereas in the CR group only 9 of 68
(13.2%) patients exceeded this level (P<.001; c2 test).
Effects of RFVE and CR therapy regarding stroke
etiology
In patients with ischemic stroke, we observed better results in the
RFVE group than the CR group (table 3), as demonstrated by

Table 1 Demographic and clinical characteristics of patients at baseline

Demographic and Clinical Parameters Overall (NZ136) RFVE Group (nZ68) CR Group (nZ68) P
Sex, men/women 80 (58.8)/56 (41.2) 37 (54.4)/31 (45.6) 43 (63.2)/25 (36.8) .299
Age, y 63.914.1 62.515.2 66.012.9 .132
Time from stroke, mo 4.23.0 4.42.8 4.13.2 .570
Lesion side, right/left 69 (50.7)/67 (49.3) 33 (48.5)/35 (51.5) 36 (53.0)/32 (47.0) .666
Type of stroke, ischemic/hemorrhagic 78 (57.4)/58 (42.6) 36 (53.0)/32 (47.0) 42 (61.8)/26 (38.2) .302
NOTE. Values are mean  SD, n (%), or as otherwise indicated.

www.archives-pmr.org
Reinforced feedback in virtual environment 839

bigger gains for the following outcomes: F-M UE (RFVE:

P (Mann-Whitney U )
14.69%, CR: 4.91%, P<.001), FIM (RFVE: 7.41%, CR: 3.50%,
PZ.004), NIHSS (RFVE: 2.58%, CR: 1.13%, PZ.010), time
RFVE Versus CR (RFVE: 101.50%, CR: 21.80%, P<.001), speed (RFVE: 103.35%,
CR: 15.12%, P<.001), and peak (RFVE: 113.34%, CR: 35.46%,
PZ.002). However, the ESAS did not differ significantly (RFVE:

<.001
<.001
.014
.022

<.001
<.001
<.001
5.89%, CR: 3.67%, PZ.382).
In patients with hemorrhagic stroke, the following outcomes
were significantly higher in the RFVE group: F-M UE (RFVE:

173.14124.12* (125.95e205.15)
(93.06e102.39)
14.77%, CR: 4.55%, P<.001), FIM (RFVE: 9.52%, CR: 6.96%,
(39.24e49.48)

(15.77e22.70)

29.6626.28* (29.47e45.06)
9.325.92* (8.30e12.04)
(2.00e3.11)
PZ.008), ESAS (RFVE: 3.18%, CR: e1.92%; PZ.003), time
(RFVE: 92.68%, CR: 34.83%, P<.001), and peak (RFVE:
83.09%, CR: 37.09%, PZ.002). On the contrary, the NIHSS
(RFVE: 2.38%, CR: 1.47%, PZ.348) and speed (RFVE: 75.29%,
CR: 42.52%, PZ.093) were comparable between groups
18.9411.70
46.2917.25*
100.6617.53*
2.521.90*

after treatment.
The comparison between ischemic and hemorrhagic subgroups
CR Group (nZ68)

within the RFVE group did not show any significant difference.
After

Relations between outcomes and clinical variables

Abbreviations: Peak, mean number of submovements; Speed, mean velocity of movement (cm/s); Time, mean duration of movement (s).
143.38106.53 (102.33e174.95)

NOTE. Results are displayed as mean  SD (95% confidence interval) or as otherwise indicated. Statistical threshold was set at P.05.

The skew-normal regression model showed that none of the

(36.50e46.40)
(86.39e96.50)

(17.80e23.99)

33.3326.17 (34.12e49.23)
10.555.50 (9.62e13.10)
(2.54e4.05)

demographic (ie, age, sex) characteristics were statistically related

to motor improvements among patients treated by RFVE.
Furthermore, the estimated regression coefficients showed that the
improvement noted in the F-M UE and NIHSS was dependent
from time since stroke and was higher in patients within 6 months
3.212.52
43.1517.21
94.5920.32

20.9411.38

(nZ102) than >6 months (nZ34) after stroke (PZ.04 and

PZ.05, respectively). Further, the regression analysis showed that
patients with right hemiparesis gained better functional skills on
Before

the FIM (PZ.02) than patients with left hemiparesis.

216.55129.15* (171.89e242.54)

Discussion
(42.22e50.87)

(12.06e18.36)

25.1221.53* (22.68e34.05)
(98.74e108)
(1.57e2.74)

7.385.16* (6.66e9.49)

The results of this study showed a positive effect of RFVE

treatment, intended as enriched environment, for rehabilitation of
upper limb after stroke. Data demonstrated that recovery is
Effect of RFVE and CR on outcomes in all patients with stroke

achievable also with CR but to a lesser extent. This study

47.7115.74*
104.4018.50*
2.172.23*
15.2311.82*

demonstrated an added therapeutic effect of RFVE treatment,

supporting a beneficial integration with CR treatment. When pa-
RFVE Group (nZ68)

tients undergo combined CR and RFVE treatments they seem to

After

regain better motor function than the recovery induced by the

augmented intensity provided by CR program. Previous clinical
studies have indicated that patients with hemorrhagic stroke may
128.6569.36 (102.52e138.12)

gain better functional improvement than patients with ischemic

(32.24e41.79)
(88.19e98.33)

(16.98e23.82)

12.197.04 (11.12e14.92)

39.5126.79 (37.40e51.08)

stroke after CR.6 However, in this study, we found that the

(2.65e3.99)

application of RFVE is effective after both ischemic and hemor-

* Comparison within group (Wilcoxon test).

rhagic stroke without contraindications because of stroke etiology;

this result provides additional evidence regarding the effect of
stroke etiology in virtual reality. Current evidence has shown that
37.9917.76
93.7421.08

20.2512.70
3.282.52

treatment in virtual environment should exceed 15 hours of

training to achieve clinical improvement12,34; in this study, both
Before

treatments were provided for 2h/d, 5d/wk for 4 weeks, which

represents a new insight on poststroke recovery on the advantage
of mixed intensive training. Despite exercises used for both
therapies were based on an exercise booklet, physiotherapists were
Speed (cm/s)

not limited in the exercises selection or restricted to number of

Peak (n)
Kinematics
Time (s)
Functional

exercises provided during one therapy session. As demonstrated in

Outcomes

F-M UE

NIHSS
Table 2

ESAS

previous studies, the virtual reality training seems to be effective

FIM

for patients with different demographic and clinical characteris-

tics, and those findings were confirmed in this trial.

www.archives-pmr.org
 840
Table 3 Effect of RFVE and CR treatments on patients with ischemic and hemorrhagic stroke
Patients With Ischemic Stroke Patients With Hemorrhagic Stroke
RFVE RFVE
RFVE Group (nZ36) CR Group (nZ42) Versus CR RFVE Group (nZ32) CR Group (nZ26) Versus CR
P (Mann- P (Mann-
Whitney Whitney
Outcomes Before After Before After U) Before After Before After U)
Functional
F-M UE 37.9217.81 47.6115.76* 43.1416.81 46.3816.60* <.001 38.0617.99 47.8115.96* 43.1518.18 46.1518.60* <.001
(28.31e42.04) (38.84e51.16) (33.23e45.49) (36.41e49.16) (31.92e45.93) (41.72e54.56) (35.70e53.35) (37.50e55.87)
FIM 100.4717.16 109.9414.26* 96.2120.50 100.6218.01* .004 86.1622.72 98.1620.83* 91.9620.13 100.7317.08* .008
(91.78e104.43) (101.62e112.58) (86.11e100.24) (92.75e105.10) (80.31e96.19) (91.89e107.11) (81.32e96.47) (88.20e103.70)
NIHSS 3.132.39 1.912.11* 3.032.37 2.411.81* .010 3.462.70 2.462.36* 3.502.78 2.702.08* .348
(2.27e4.08) (1.08e2.64) (2.25e4.11) (1.81e3.19) (2.42e4.51) (1.55e3.38) (2.10e4.85) (1.61e3.65)
ESAS 21.5914.53 14.9711.25* 22.7211.98 17.918.31* .382 18.7110.28 15.5412.63* 18.109.98 20.6015.82 .003
(16.33e27.74) (10.59e19.21) (17.88e26.55) (14.46e20.83) (14.73e22.70) (10.64e20.44) (14.37e23.52) (14.05e29.11)
Kinematics
Time (s) 11.507.34 6.715.06* 10.825.40 9.675.69* <.001 12.976.72 8.145.25* 10.115.72 8.776.35* <.001
(9.66e15.43) (5.62e9.61) (9.40e13.88) (8.02e12.71) (10.88e16.15) (6.44e10.67) (7.92e13.98) (6.51e13.25)
Speed 130.3576.83 223.03126.30* 153.25126.03 170.57135.68* <.001 126.7461.03 209.27133.93* 127.4562.76 177.30105.20* .093
(cm/s) (92.48e147.81) (158.07e254.09) (90.86e207.82) (107.96e230.10) (96.44e144.55) (153.22e263.55) (90.44e155.31) (144.11e206.72)
Peak (n) 36.9127.20 22.4220.97* 34.3525.76 30.3125.29* .002 42.2826.49 27.9922.09* 31.7427.25 28.6428.24* <.001
(31.25e51.68) (17.75e33.72) (33.89e53.17) (28.52e48.12) (37.53e56.71) (22.60e39.57) (25.73e52.15) (21.76e49.66)
NOTE. Results are displayed as mean  SD (95% confidence interval) or as otherwise indicated. Statistical threshold was set at P.05.
Abbreviations: Peak, mean number of submovements; Speed, mean velocity of movement (cm/s); Time, mean duration of movement (s).
* Comparison within group (Wilcoxon test).
www.archives-pmr.org

P. Kiper et al
Reinforced feedback in virtual environment
Although the neural mechanisms associated with practice-
dependent motor recovery are not clearly understood, it has been
suggested that intensive and repetitive use of the affected limb, as
those stimulated by RFVE, may induce a positive effect on neu-
roplasticity and improvement of motor function.7,8,35 Motor
learning implies a process of motor action selection to perform a
requested task, and theoretically the movement should be repeated
exactly to obtain the best relearning of one specific movement.36
We have argued that in our proposal for RFVE treatment,
different paradigms (eg, reinforcement learning, supervised
learning) operate to promote motor learning, based on the feed-
back received from the artificial environment during
motor practice.
Several studies have demonstrated greater improvement of
motor function in patients treated with virtual environment.11,12
This was confirmed by the results of this study because the
patients from the RFVE group improved their motor functions by
14.7% on average on the F-M UE, which is more than the
established minimal clinically important difference (13.6%).32
Moreover, the minimal detectable change level (7.9%)37 was
exceeded by 79.4% of participants. Furthermore, the analysis
suggests that RFVE can be provided regardless of the subject’s
age and sex, and stroke etiology. However, as demonstrated by
results on the F-M UE and NIHSS, the time since stroke onset still
remains a major factor determining functional recovery, with the
biggest improvements occurring closer to stroke onset. In fact, the
greatest functional improvement was found in patients within 6
months of stroke onset. The results of the FIM showed greater
improvements in patients affected by right hemiparesis. This could
be because of right-hand dominance in most of the population or
because left hemiparesis (caused mostly by lesions in the right
hemisphere) can provoke severe impairments in behavioral
domains, or cognitive and attentive functions. Finally, patients did
not report significant changes in their general health conditions
(ESAS) after the introduction of double CR, mainly based on
intensive repetition of exercises. Such a result suggests that
massed practice should be mixed with task-oriented and variable
motor exercises, such as those provided by RFVE.38,39
The benefits coming from the use of enhanced feedback in
poststroke patients has been reported in various studies, showing
that when provided through virtual environment, it can be bene-
ficial both in the subacute and chronic phases after stroke.40 Our
findings support the evidence that the process of motor recovery in
the subacute phase after stroke is enhanced effectively by the use
of virtual realityebased therapy,12 potentially because of variable
and random tasks delivered with sustained intensity in virtual
environment. Moreover, an adjunctive important factor influ-
encing motor recovery may rely on the frequency of feedback
known to promote motor learning (ie, knowledge of performance,
knowledge of results). It is also known that the visual and acoustic
feedback are provided in high-quality systems for the interaction
with virtual environments, and this has a potential role for
regaining a better motor outcome. It also has been shown that
enhanced feedback is important to increase patient’s motivation.41
Therefore, in this study, we also provided the sensory feedback
throughout the real object coherent with the virtual one, which
was held during task execution. However, this was limited to the
patients with residual grasp capacity. In patients with stroke, vir-
tual training may stimulate movement relearning by underpinning
the recovery of previous motor skills or facilitating the acquisition
of new ones, to be translated into activities of daily living.
www.archives-pmr.org
841
Study limitations
The differences observed between motor improvements of the upper
extremity may be the result of intensive virtual training and intensive
conventional therapy, both of which were introduced in the RFVE
group. Therefore, it was not possible to infer directly to what extent
the motor function improvement was because of virtual or con-
ventional treatment. Nevertheless, the combination of RFVE with
CR treatment resulted in significant improvement of upper limb
motor function and kinematic parameters. Moreover, the specific
content of every exercise was not controlled either in the RFVE or
CR group. Therefore, it is not possible to conclude which specific
parameters were clinically useful or which patients better exploited
the proposed stimuli. Patients in both groups performed exercises
for the upper extremity for 2 hours in total, and each training was
selected appropriately in accordance with patient’s deficits. How-
ever, the intensity, intended as several repetitions, was not recorded,
which is another limitation of the study. Moreover, because not all
tasks and movements were matched in the 2 treatments (ie, CR,
RFVE), it was not possible to infer whether the intensity of training
could explain the difference between groups. Patients with upper
limb sensory disorders, neglect, apraxia, severe cognitive impair-
ment, and verbal comprehension were excluded from this study,
therefore a large number of patients could not be analyzed. Further
studies should control the intensity of training provided in both the
experimental and control treatments and should take into consid-
eration such characteristics.
Conclusions
The RFVE training combined with CR treatment provided better
and clinically meaningful results compared with the same amount
of CR. Moreover, observed effectiveness of virtual therapy was
comparable for patients with ischemic and hemorrhagic stroke.
The effectiveness of virtual therapy was not dependent on age and
sex, but it remains sensitive to time since stroke onset. Virtual
therapy by means of RFVE can be considered as a safe treatment
and feasible for clinical use in most of the patients experiencing
hemiparesis after stroke, without severe cognitive impairments,
neglect, apraxia, and sensory disorders.
Suppliers
a. Excel 2007; Microsoft.
b. Virtual Reality Rehabilitation System (VRRS); Khymeia
Group.
c. LIBERTY; Polhemus.
Keywords
Feedback; Rehabilitation; Stroke; Virtual reality
Corresponding author
Pawel Kiper, PhD, Fondazione Ospedale San Camillo IRCCS, via
Alberoni 70, 30126 Venezia, Italy. E-mail address: pawel.kiper@
ospedalesancamillo.net.
842
Acknowledgment
We thank Aneta Kiper, MSc, PT, for her support and assistance
throughout the study.
References
1. Arya KN, Pandian S, Verma R, Garg RK. Movement therapy induced
neural reorganization and motor recovery in stroke: a review. J Bodyw
Mov Ther 2011;15:528-37.
2. Zang Y, Jia F, Weng X, et al. Functional organization of the primary
motor cortex characterized by event-related fMRI during movement
preparation and execution. Neurosci Lett 2003;337:69-72.
3. Teka WW, Hamade KC, Barnett WH, et al. From the motor cortex to
the movement and back again. PLoS One 2017;12:e0179288.
4. Xu GQ, Lan Y, Zhang Q, Liu DX, He XF, Lin T. 1-Hz repetitive
transcranial magnetic stimulation over the posterior parietal cortex
modulates spatial attention. Front Hum Neurosci 2016;10:38.
5. Dennis A, Bosnell R, Dawes H, et al. Cognitive context determines
dorsal premotor cortical activity during hand movement in patients
after stroke. Stroke 2011;42:1056-61.
6. Kelly PJ, Furie KL, Shafqat S, Rallis N, Chang Y, Stein J. Functional
recovery following rehabilitation after hemorrhagic and ischemic
stroke. Arch Phys Med Rehabil 2003;84:968-72.
7. Kiper P, Agostini M, Luque-Moreno C, Tonin P, Turolla A. Reinforced
feedback in virtual environment for rehabilitation of upper extremity
dysfunction after stroke: preliminary data from a randomized
controlled trial. Biomed Res Int 2014;2014:752128.
8. Krakauer JW. Motor learning: its relevance to stroke recovery and
neurorehabilitation. Curr Opin Neurol 2006;19:84-90.
9. Kitago T, Krakauer JW. Motor learning principles for neuro-
rehabilitation. Handb Clin Neurol 2013;110:93-103.
10. Vearrier LA, Langan J, Shumway-Cook A, Woollacott M. An inten-
sive massed practice approach to retraining balance post-stroke. Gait
Posture 2005;22:154-63.
11. Pollock A, Farmer SE, Brady MC, et al. Interventions for improving
upper limb function after stroke. Cochrane Database Syst Rev
2014;(11):CD010820.
12. Laver KE, George S, Thomas S, Deutsch JE, Crotty M. Virtual reality for
stroke rehabilitation. Cochrane Database Syst Rev 2015;(2):CD008349.
13. Bermudez i Badia S, Garcia Morgade A, Samaha H, Verschure PF.
Using a hybrid brain computer interface and virtual reality system to
monitor and promote cortical reorganization through motor activity
and motor imagery training. IEEE Trans Neural Syst Rehabil Eng
2013;21:174-81.
14. Merians AS, Tunik E, Adamovich SV. Virtual reality to maximize
function for hand and arm rehabilitation: exploration of neural
mechanisms. Stud Health Technol Inform 2009;145:109-25.
15. Tunik E, Saleh S, Adamovich SV. Visuomotor discordance during
visually-guided hand movement in virtual reality modulates sensori-
motor cortical activity in healthy and hemiparetic subjects. IEEE
Trans Neural Syst Rehabil Eng 2013;21:198-207.
16. Lucca LF. Virtual reality and motor rehabilitation of the upper limb after
stroke: a generation of progress? J Rehabil Med 2009;41:1003-100.
17. Turolla A, Dam M, Ventura L, et al. Virtual reality for the rehabili-
tation of the upper limb motor function after stroke: a prospective
controlled trial. J Neuroeng Rehabil 2013;10:85.
18. Kiper P, Piron L, Turolla A, Stozek J, Tonin P. The effectiveness of
reinforced feedback in virtual environment in the first 12 months after
stroke. Neurol Neurochir Pol 2011;45:436-44.
19. Luque-Moreno C, Oliva-Pascual-Vaca A, Kiper P, Rodriguez-
Blanco C, Agostini M, Turolla A. Virtual reality to assess and treat
P. Kiper et al
lower extremity disorders in post-stroke patients. Methods Inf Med
2016;55:89-92.
20. Folstein MF, Folstein SE, McHugh PR. Mini-mental state”. A prac-
tical method for grading the cognitive state of patients for the clini-
cian. J Psychiatr Res 1975;12:189-98.
21. De Renzi E, Motti F, Nichelli P. Imitating gestures. A quantitative
approach to ideomotor apraxia. Arch Neurol 1980;37:6-10.
22. De Renzi E, Vignolo LA. The token test: a sensitive test to detect
receptive disturbances in aphasics. Brain 1962;85:665-78.
23. Brott T, Adams HP Jr, Olinger CP, et al. Measurements of acute
cerebral infarction: a clinical examination scale. Stroke 1989;20:
864-70.
24. Piron L, Turolla A, Agostini M, et al. Motor learning principles for
rehabilitation: a pilot randomized controlled study in poststroke pa-
tients. Neurorehabil Neural Repair 2010;24:501-8.
25. Luque-Moreno C, Ferragut-Garcias A, Rodriguez-Blanco C, et al. A
decade of progress using virtual reality for poststroke lower extremity
rehabilitation: systematic review of the intervention methods. Biomed
Res Int 2015;2015:342529.
26. Kiper P, Zucconi C, Agostini M, et al. Assessment of virtual
teacher feedback for the recovery of the upper limb after a stroke.
Study protocol for a randomized controlled trial. Med Rehabil
2016;20:13-20.
27. Levin MF, Kleim JA, Wolf SL. What do motor “recovery” and
“compensation” mean in patients following stroke? Neurorehabil
Neural Repair 2009;23:313-9.
28. Fugl-Meyer AR, Jaasko L, Leyman I, Olsson S, Steglind S. The post-
stroke hemiplegic patient. 1. A method for evaluation of physical
performance. Scand J Rehabil Med 1975;7:13-31.
29. Keith RA, Granger CV, Hamilton BB, Sherwin FS. The functional
independence measure: a new tool for rehabilitation. Adv Clin Rehabil
1987;1:6-18.
30. Bruera E, Kuehn N, Miller MJ, Selmser P, Macmillan K. The
Edmonton Symptom Assessment System (ESAS): a simple method for
the assessment of palliative care patients. J Palliat Care 1991;7:6-9.
31. Piron L, Tonin P, Cortese F, et al. Post-stroke arm motor tele-
rehabilitation web-based. In: International Workshop on Virtual
Rehabilitation. August 29-30, 2006. New York, NY: IEEE; 2006.
32. Shelton FD, Volpe BT, Reding M. Motor impairment as a predictor of
functional recovery and guide to rehabilitation treatment after stroke.
Neurorehabil Neural Repair 2001;15:229-37.
33. Sveistrup H. Motor rehabilitation using virtual reality. J Neuroeng
Rehabil 2004;1:10.
34. Colomer C, Llorens R, Noe E, Alcaniz M. Effect of a mixed reality-
based intervention on arm, hand, and finger function on chronic
stroke. J Neuroeng Rehabil 2016;13:45.
35. van Dokkum L, Mottet D, Bonnin-Koang HY, et al. People post-stroke
perceive movement fluency in virtual reality. Exp Brain Res 2012;218:
1-8.
36. Doya K. What are the computations of the cerebellum, the basal
ganglia and the cerebral cortex? Neural Netw 1999;12:961-74.
37. Wagner JM, Rhodes JA, Patten C. Reproducibility and minimal detectable
change of three-dimensional kinematic analysis of reaching tasks in
people with hemiparesis after stroke. Phys Ther 2008;88:652-63.
38. Hanlon RE. Motor learning following unilateral stroke. Arch Phys
Med Rehabil 1996;77:811-5.
39. Shea CH, Kohl RM. Composition of practice: influence on the
retention of motor skills. Res Q Exerc Sport 1991;62:187-95.
40. Subramanian SK, Lourenco CB, Chilingaryan G, Sveistrup H,
Levin MF. Arm motor recovery using a virtual reality intervention in
chronic stroke: randomized control trial. Neurorehabil Neural Repair
2013;27:13-23.
41. Piron L, Turolla A, Tonin P, Piccione F, Lain L, Dam M. Satisfaction
with care in post-stroke patients undergoing a telerehabilitation pro-
gramme at home. J Telemed Telecare 2008;14:257-60.
www.archives-pmr.org
Reinforced feedback in virtual environment 842.e1

Supplemental Appendix S1 Exercise Client’s aim

Booklet To strengthen the muscles that straighten your elbow.
Bending the wrist to a target
Therapist’s instructions
Therapist’s aim
Position the patient in supine with their arm extended vertically.
To improve the ability to extend the wrist. Instruct the patient to bend their elbow to bring their palm to touch
their forehead and then straighten it again. Ensure that the upper
arm remains vertical and only the forearm moves.
Client’s aim
To improve your ability to straighten your wrist. Client’s instructions
Position yourself lying on your back with your arm held vertically.
Therapist’s instructions Start with your elbow straight. Finish with your elbow bent and
Position the patient in sitting with their forearm pronated and resting your palm touching your forehead. Ensure that your upper arm
on a block on a table in front of them. Place a cup in their hand, a does not move and only your forearm moves.
sandbag on their forearm, and tape a flexible straw to the dorsal
surface of their forearm with the joint of the straw aligned with their
wrist joint. Instruct the patient to extend their wrist to touch the tip of Elbow extensor strengthening in sitting
the straw and then relax their wrist back into flexion. using free weights
Client’s instructions Therapist’s aim
Position yourself sitting with your forearm resting on a block on a To strengthen the elbow extensors.
table in front of you and your palm facing down. Place a cup in
your hand, a sandbag on your forearm, and tape a flexible straw to Client’s aim
the top of your forearm so that the bend of the straw is over your
wrist. Practice straightening your wrist so that your hand touches To strengthen your triceps.
the tip of the straw and then relaxing your hand back down again.
Therapist’s instructions
Bending the wrist while holding a cup Position the patient in sitting with their shoulder flexed and elbow
flexed. Instruct the patient to extend their elbow.
Therapist’s aim
Client’s instructions
To improve the ability to radially deviate the wrist.
Position yourself sitting on a chair. Start with your arm above your
head and your elbow bent. Finish with your arm above your head
Client’s aim and your elbow straight.
To improve your ability to bend your wrist.
Precautions
Therapist’s instructions
1. Ensure that wheelchair does not flip backward.
Position the patient in sitting with their hand grasping a cup and
hanging over the edge of a table. Instruct the patient to tilt the cup
up by radially deviating their wrist to a point level with, or higher Elbow extensor strengthening in sitting
than, the table.
using pulleys
Client’s instructions
Therapist’s aim
Position yourself sitting with your hand grasping a cup and
To strengthen the elbow extensors.
hanging over the edge of a table. Practice tilting the cup up by
bending your wrist to a point level with, or higher than, the table.
Client’s aim
To strengthen your triceps.
Elbow extensor strengthening in lying
without weights Therapist’s instructions

Therapist’s aim Position the patient in sitting facing the pulleys. Adjust the pulley
system so that the direction of pull opposes elbow extension.
To strengthen the elbow extensors. Instruct the patient to extend their elbow.

www.archives-pmr.org
842.e2 P. Kiper et al

Client’s instructions Lifting up an object

Position yourself sitting facing the pulleys. Adjust the pulley so
that the direction of pull is downward from the ceiling. Start with Therapist’s aim
your elbow bent and tucked in beside your body. Finish with your
To improve the ability to reach.
elbow straight.

Client’s aim
Forearm supination and pronation to a wall To improve your ability to reach.
target
Therapist’s instructions
Therapist’s aim
Position the patient in sitting with an object on a table in front of
To improve the ability to reach for objects.
them. Instruct the patient to lift the object off the table. Ensure that
the elbow is extended.
Client’s aim
To improve your ability to reach for objects. Client’s instructions
Position yourself sitting with an object on a table in front of you.
Therapist’s instructions Practice lifting the object off the table. Ensure that your elbow
stays straight.
Position the patient in sitting in front of a table and wall with their
elbow bent and a ruler taped to their hand. Draw 2 target lines on
some paper on the wall. Instruct the patient to supinate and pro-
nate their forearm until the ruler reaches the target lines. Ensure Picking an object off the floor
that internal rotation does not occur at the shoulder.
Therapist’s aim
Client’s instructions
To improve the ability to sit and reach for an object.
Position yourself sitting in front of a table and wall with your
elbow bent and a ruler taped to your hand. Draw 2 target lines on
Client’s aim
some paper on the wall. Practice rotating your forearm back and
forth until the ruler reaches the target lines. Ensure that the hand To improve your ability to sit and reach for an object.
does not move sideways.

Lifting an object from the floor to the table
Therapist’s aim
To improve the ability to sit and reach for an object.
Therapist’s instructions
Position the patient in sitting with an object placed on the floor
on their affected side. Instruct the patient to reach down and pick
up the object with either hand. Ensure that the affected knee
remains over a flat foot so that weight is borne through the
affected side.

Client’s aim Client’s instructions

To improve your ability to sit and reach for an object.
Therapist’s instructions
Position the patient in sitting with an object on the floor on their
affected side and a table nearby. Instruct the patient to reach down
and pick up the object with either hand and place it on the table.
Ensure that the affected knee remains over a flat foot so that
weight is borne through the affected side.
Position yourself sitting with an object placed on the floor on your
affected side. Practice reaching down to pick up the object with
either hand. Ensure that your affected knee remains over your foot
so that you weight-bear through your affected side.
Reaching from a low surface to a high
surface

Client’s instructions Therapist’s aim

Position yourself sitting with an object on the floor on your To improve the ability to reach.
affected side and a table nearby. Practice reaching down to pick up
the object with either hand and placing it on the table. Ensure that Client’s aim
your affected knee remains over your foot so that you weight-bear
through your affected side. To improve your ability to reach.

www.archives-pmr.org
Reinforced feedback in virtual environment
Therapist’s instructions
Position the patient in sitting with a block on a table in front of
them. Instruct the patient to lift a cup from the table to the top of
the block. Ensure that the arm flexes forward rather than abducts
and the shoulder does not elevate.
Client’s instructions
Position yourself sitting with a block on a table in front of you.
Practice lifting a cup from the table to the top of the block. Ensure
that your elbow stays tucked in and your shoulder does not hitch.
Reaching from side to side in standing
Therapist’s aim
To improve the ability to hold and move an object.
Client’s aim
To improve your ability to hold and move and object.
Therapist’s instructions
Position the patient in standing with a table in front of them and
another table to their side. Instruct the patient to lift an object with
their hand from one table to the other without dropping it.
Client’s instructions
Position yourself standing with a table in front of you and another
table to your side. Practice lifting an object with your hand from
one table to the other without dropping it.
Rotating the shoulder outward while
holding a cup
Therapist’s aim
To improve the ability to reach.
Client’s aim
To improve your ability to reach.
Therapist’s instructions
Position the patient in sitting with their elbow bent at 90 and a
piece of paper in between their upper arm and chest. Position a
table on the outside of the patient and place a cup of water in their
hand. Instruct the patient to move the cup outward to the table
without spilling the water. Ensure that the back is kept against the
chair and the piece of paper is not dropped.
Client’s instructions
Position yourself sitting with your elbow bent at right angles and a
piece of paper in between your upper arm and the side of your
www.archives-pmr.org
842.e3
chest. Practice moving a cup of water outward to a table beside
you without spilling it. Ensure that your back is kept against the
chair and the piece of paper is not dropped.
Shoulder abductor strengthening in sitting
using free weights
Therapist’s aim
To strengthen the shoulder abductors.
Client’s aim
To strengthen the muscles at the side and top of your shoulder.
Therapist’s instructions
Position the patient in sitting with their shoulder adducted. Instruct
the patient to abduct their shoulder with their elbow extended.
Client’s instructions
Position yourself sitting in a chair. Start with your arm down
beside your body. Finish with your arm above your head. Ensure
that you keep your elbow straight.
Precautions
1. Ensure that the chair does not tip backward.
Shoulder abductor strengthening in sitting
using pulleys
Therapist’s aim
To strengthen the shoulder abductors.
Client’s aim
To strengthen the muscles at the side and top of your shoulder.
Therapist’s instructions
Position the patient in sitting perpendicular to the pulleys. Adjust
the pulley so that the direction of pull opposes shoulder abduction.
Instruct the patient to abduct their shoulder with their
elbow extended.
Client’s instructions
Position yourself sitting side-on to the pulley. Adjust the pulley so
that the direction of pull is upward from the floor. Start with your
arm beside your body. Finish with your arm out to the side. Ensure
that you keep your elbow straight.
Precautions
1. Ensure that the chair does not tip backward.
842.e4 P. Kiper et al

Shoulder extensor strengthening in sitting Standing and reaching to the side

using uppertone
Therapist’s aim
Therapist’s aim To improve the ability to shift weight sideways when reaching to
To strengthen the shoulder extensors. the side in standing.

Client’s aim
To strengthen the muscles at the back of your shoulder.
Therapist’s instructions
Position the patient in sitting facing the frame of the Uppertone
with their hands wedged in the Uppertone handles. Instruct the
patient to extend their shoulders.
Client’s instructions
Position yourself sitting facing the frame of the Uppertone with
your hands wedged in the Uppertone handles. Start with your arms
out straight in front of you. Finish with your elbows tucked beside
your body.
Precautions
Client’s aim
To improve your ability to reach to the side in standing.
Therapist’s instructions
Position the patient in standing with their feet slightly apart and
a target placed at shoulder height out to their side. Position a
table at hip level beside them. Instruct the patient to reach
sideways to touch the target with their hand and the table with
their hips.
Client’s instructions
Position yourself standing with your feet slightly apart and a target
placed at shoulder height out to your side. Position a table at hip
level beside you. Practice reaching sideways to touch the target.
Ensure that your hips move slightly sideways to make contact with
the table.

1. Ensure that the chair does not tip backward. Taking a cup to the mouth
Therapist’s aim
Shoulder flexor strengthening in supine
To improve the ability to bring a cup to the mouth.
using free weights
Client’s aim
Therapist’s aim
To improve your ability to bring a cup to your mouth.
To strengthen the shoulder flexors.

Client’s aim
To strengthen the muscles at the front of your shoulder.
Therapist’s instructions
Position the patient in supine with their elbow extended. Instruct
the patient to flex their shoulder.
Client’s instructions
Position yourself lying on your back. Start with your arm beside
your body. Finish with your arm at 90 to your body. Ensure that
you keep your elbow straight.
Therapist’s instructions
Position the patient in sitting with a cup on a table in front of
them. Instruct the patient to lift the cup up to their mouth. Ensure
that their head remains erect.
Client’s instructions
Position yourself sitting with a cup on a table in front of you.
Practice lifting the cup up to your mouth. Ensure that you keep
your head up straight.
Precautions
1. Ensure appropriate for patients with swallowing difficulties.

www.archives-pmr.org