Accepted Manuscript

Upper limb robot-assisted rehabilitation versus physical therapy on subacute stroke

patients: a follow-up study

Marco Franceschini, Stefano Mazzoleni, Michela Goffredo, Sanaz Pournajaf, Daniele

Galafate, Simone Criscuolo, Maurizio Agosti, Federico Posteraro

PII: S1360-8592(19)30111-1
DOI: https://doi.org/10.1016/j.jbmt.2019.03.016
Reference: YJBMT 1807

To appear in: Journal of Bodywork & Movement Therapies

Received Date: 20 March 2019

Accepted Date: 29 March 2019

Please cite this article as: Franceschini, M., Mazzoleni, S., Goffredo, M., Pournajaf, S., Galafate, D.,
Criscuolo, S., Agosti, M., Posteraro, F., Upper limb robot-assisted rehabilitation versus physical therapy
on subacute stroke patients: a follow-up study, Journal of Bodywork & Movement Therapies, https://
doi.org/10.1016/j.jbmt.2019.03.016.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT
TITLE PAGE

Upper limb robot-assisted rehabilitation versus physical therapy on subacute

stroke patients: a follow-up study

Marco Franceschinia,b, Stefano Mazzolenic,d, Michela Goffredoa, Sanaz Pournajafa,

PT
Daniele Galafatea, Simone Criscuoloa, Maurizio Agostie, Federico Posterarod,f

RI
a
Department of Neurorehabilitation IRCCS San Raffaele Pisana, Rome, Italy; bSan Raffaele

SC
University, Rome, Italy; cThe BioRobotics Institute, Scuola Superiore Sant’Anna, Pisa, Italy;

U
d e
Rehabilitation Bioengineering Laboratory, Volterra, Italy; Passignano sul Trasimeno, Italy;
AN
f
Rehabilitation Department Versilia Hospital AUSL Tuscany North West, Camaiore, Italy.
M

Corresponding author
D

Sanaz Pournajaf
TE

Department of Neurorehabilitation
EP

IRCCS San Raffaele Pisana

Via della Pisana, 235

C
AC

00163 Rome, Italy

Ph: +39 0652252319

E-mail: [email protected]

Co-authors:

Marco Franceschini
 ACCEPTED MANUSCRIPT
Department of Neurorehabilitation and San Raffaele University

IRCCS San Raffaele Pisana

Via della Pisana, 235

00163 Rome, Italy

PT
Ph: +39 0652252319

RI
E-mail: [email protected]

SC
Stefano Mazzoleni

U
The BioRobotics Institute, Scuola Superiore Sant'Anna
AN
V.le R. Piaggio 34
M

56025 Pisa, Italy

Ph: +39 050883132

D
TE

E-mail: [email protected]
EP

Michela Goffredo
C

Department of Neurorehabilitation
AC

IRCCS San Raffaele Pisana

Via della Pisana, 235

00163 Rome, Italy

Ph: +39 0652252319

E-mail: [email protected]
 ACCEPTED MANUSCRIPT

Daniele Galafate

Department of Neurorehabilitation

IRCCS San Raffaele Pisana

PT
Via della Pisana, 235

RI
00163 Rome, Italy

Ph: +39 0652252319

SC
E-mail: [email protected]

U
AN
Simone Criscuolo
M

Department of Neurorehabilitation

IRCCS San Raffaele Pisana

D
TE

Via della Pisana, 235

00163 Rome, Italy

EP

Ph: +39 0652252319

C

E-mail: [email protected]
AC

Maurizio Agosti

Rehabilitation unit, Hospital of Passignano

Via Gramsci, 14

43126 Passignano sul Trasimeno, Italy

 ACCEPTED MANUSCRIPT
Ph: +39 0521703335

E-mail: [email protected]

Federico Posteraro

PT
Rehabilitation Department - Versilia Hospital - AUSL Tuscany North West

RI
Via Aurelia 335

Camaiore – Lucca, Italy

SC
Ph: +39 05846057062

U
E-mail: [email protected]
AN
M

Declarations of interest: none.

D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
TITLE PAGE

Upper limb robot-assisted rehabilitation versus physical therapy on

subacute stroke patients: a follow-up study

Marco Franceschinia,b, Stefano Mazzolenic,d, Michela Goffredoa, Sanaz

Pournajafa, Daniele Galafatea, Simone Criscuoloa, Maurizio Agostie,

PT
Federico Posterarod,f

RI
SC
a
Department of Neurorehabilitation IRCCS San Raffaele Pisana, Rome, Italy; bSan
c
Raffaele University, Rome, Italy; The BioRobotics Institute, Scuola Superiore

U
Sant’Anna, Pisa, Italy; dRehabilitation Bioengineering Laboratory, Volterra, Italy;
AN
e f
Department of Rehabilitation, Hospital-University Parma, Italy; Rehabilitation
M

Department Versilia Hospital AUSL Tuscany North West, Camaiore Italy.

D
TE

Corresponding author

Sanaz Pournajaf
EP

Department of Neurorehabilitation
C

IRCCS San Raffaele Pisana

AC

Via della Pisana, 235

00163 Rome, Italy

Ph: +39 0652252319

E-mail: [email protected]
 ACCEPTED MANUSCRIPT
Co-authors:

Marco Franceschini

Department of Neurorehabilitation and San Raffaele University

IRCCS San Raffaele Pisana

PT
Via della Pisana, 235

00163 Rome, Italy

RI
Ph: +39 0652252319

SC
E-mail: [email protected]

U
AN
Stefano Mazzoleni

The BioRobotics Institute, Scuola Superiore Sant'Anna

M

V.le R. Piaggio 34
D

56025 Pisa, Italy

TE

Ph: +39 050883132

EP

E-mail: [email protected]
C

Michela Goffredo
AC

Department of Neurorehabilitation

IRCCS San Raffaele Pisana

Via della Pisana, 235

00163 Rome, Italy

 ACCEPTED MANUSCRIPT
Ph: +39 0652252319

E-mail: [email protected]

Daniele Galafate

PT
Department of Neurorehabilitation

IRCCS San Raffaele Pisana

RI
Via della Pisana, 235

SC
00163 Rome, Italy

Ph: +39 0652252319

U
AN
E-mail: [email protected]
M

Simone Criscuolo
D

Department of Neurorehabilitation
TE

IRCCS San Raffaele Pisana

EP

Via della Pisana, 235

00163 Rome, Italy

C

Ph: +39 0652252319

AC

E-mail: [email protected]

Maurizio Agosti

Department of Geriatrics and Rehabilitation, University Hospital Parma

 ACCEPTED MANUSCRIPT
Via Gramsci, 14

43126 Parma, Italy

Ph: +39 0521703335

E-mail: [email protected]

PT
Federico Posteraro

RI
Rehabilitation Department - Versilia Hospital - AUSL Tuscany North West

SC
Via Aurelia 335

Camaiore – Lucca, Italy

U
AN
Ph: +39 05846057062

E-mail: [email protected]
M
D

Declarations of interest: none.

TE
C EP
AC
 ACCEPTED MANUSCRIPT
ABSTRACT

Abstract

This study aims to analyse the long-term effects (6 months follow-up) of upper limb

Robot-assisted Therapy (RT) compared to a Traditional physical Therapy (TT), in

subacute stroke patients. Although the literature on upper-limb rehabilitation with

PT
robots shows increasing evidence of its effectiveness in stroke survivors, how long the

re-learned motor abilities could be maintained over time is still understudied. A

RI
randomized controlled follow-up study was conducted on 48 subacute stroke patients

SC
who performed the upper-limb therapy using a planar end-effector robotic system

(Experimental Group-EG) or TT (Control Group-CG). The clinical assessments were

U
collected at T0 (baseline), T1 (end of treatment) and T2 (6 months follow-up): Upper
AN
Limb part of Fugl-Meyer assessment (FM-UL), total passive Range Of Motion

(pROM), Modified Ashworth Scale Shoulder (MAS-S) and Elbow (MAS-E). At T1, the
M

intra-group analysis showed significant gain of FM-UL in both EG and CG, while
D

significant improvement in MAS-S, MAS-E, and pROM were found in the EG only. At
TE

T2, significant increase in MAS-S were revealed only in the CG. In FM-UL, pROM and

MAS-E the improvements obtained at the end of treatment seem to be maintained at 6

EP

months follow-up in both groups. The inter-groups analysis of FM-UL values at T1 and

T2 demonstrated significant differences in favour of EG. In conclusion, upper limb

C

Robot-assisted Therapy may lead a greater reduction of motor impairment in subacute

AC

stroke patients compared to Traditional Therapy. The gains observed at the end of

treatment persisted over time. No serious adverse event related to the study occurred.

Keywords

Robot-assisted Therapy, Follow-up, Rehabilitation, Stroke, Upper Limb

 ACCEPTED MANUSCRIPT
TEXT

Introduction

Stroke is one of the most common causes of adult disabilities with high incidence

rates (Béjot et al., 2016) and it has a strong impact on healthcare services and costs.

Impairments caused by stroke may result in a variety of symptoms such as sensory,

PT
motor, cognitive, and psychological deficits that need specific treatments to be restored.

Specially, the recovery of upper limb function requires prolonged period of

RI
rehabilitation treatment, even if started at an early stage, and its prognosis seems to be

SC
worse than the lower limbs impairments (Langhorne et al., 2009). Consequently,

impaired arm and hand functions induce considerable limitations in the Activities of

U
Daily Living (ADLs) performance, which are directly related to the quality of life (Rand
AN
& Eng, 2015).
M

Upper limb rehabilitation is often described as a process of active motor re-learning

that starts within the first few days after stroke and requires a complex integration of
D

muscle activities of both proximal and distal parts (Langhorne et al., 2011). Repetitive
TE

motor tasks and intensive treatments appear to have a prominent impact on promoting

the neuroplasticity and functional outcomes improvement (Lohse et al., 2014; van
EP

Kordelaar et al., 2014; Waddell et al., 2014). Studies on the time course of motor
C

recovery after stroke reported that the greatest gains in motor function occur within the
AC

first month post-stroke with some additional improvements up to 6 months after the

acute event (Buma et al., 2013). Limited capacity of neural mechanisms to reorganize

and recover missed functions after stroke has been considered to make a plateau in

motor abilities related to the timing and intensity of stroke rehabilitation treatments

(Kwakkel et al., 2004). A high-dose intensive training, through the specific functional
 ACCEPTED MANUSCRIPT
task repetitions, represents an important factor for enhancing motor recovery and

preserving it over time (Feys et al., 2004; Lohse et al., 2014).

In this context, robotic devices could be considered as suitable tools for achieving

this kind of goals due to their ability to control and quantify the intensity of the motor

tasks. They are also able to measure and control kinematics and kinetics providing

PT
repetitive, adaptive, and intensive treatments. In literature, there is an increasing

evidence of the effectiveness of robotics for rehabilitation in terms of motor impairment

RI
reduction in subacute stroke patients (Babaiasl et al., 2016; Mehrholz et al., 2015; Yoo

SC
& Kim, 2015). These findings have been reported in several studies (Klamroth-

Marganska et al., 2014; Masiero et al., 2014; Prange et al., 2015) revealing that patients

U
who received robot-assisted therapy had significant short-term gains in motor
AN
coordination and muscle strength of the exercised shoulder and elbow. In a recent study

on subacute stroke patients (Sale et al., 2014) significant improvements in upper limb
M

total passive Range Of Motion (pROM) and muscle tones, assessed by Modified
D

Ashworth Scale Shoulder (MAS-S) and Elbow (MAS-E), have been found in patients
TE

treated with Robot-assisted Therapy (RT) compared to Traditional Therapy (TT). Thus,

robotic technology provides new options for rehabilitation and it may increase the
EP

knowledge of post-stroke neural reorganization mechanisms in terms of motor recovery

and motor function’s maintenance (Krebs et al., 2000; Veerbeek et al., 2017). However,
C

the optimization of RT for maximising functional improvements (i.e., ADLs, quality of

AC

life, activities, and participation) and preserving and/or increasing such progress over

time is still an open issue (Fasoli et al., 2004).

There is a limited number of follow-up studies on subacute stroke patients who had

conducted a robot-assisted therapy (Burgar et al., 2011; Krebs et al., 2000; Lum et al.,

2006; Masiero et al., 2007; Taveggia et al., 2016). These trials are heterogeneous
 ACCEPTED MANUSCRIPT
because of different devices used for RT, different definition of the follow-up terms

(e.g. 3, 6, 8 months) and different aims of the studies.

Considering that studies on short-term efficacy of RT often employed planar end-

effector robots (Rabadi et al., 2008; Sale et al., 2014; Yoo & Kim, 2015; Mehrholz et

al., 2015; Babaiasl et al., 2016) and literature on long-term effects in subacute stroke

PT
patients is variegated (Burgar et al., 2011; Krebs et al., 2000; Lum et al., 2006; Masiero

et al., 2007; Taveggia et al., 2016), further follow-up investigations on the effects of

RI
upper-limb RT in subacute phase are needed.

SC
This study aims to extend our previous analysis on short-term effects of upper limb

robot-assisted therapy in subacute stroke patients (Sale et al., 2014), with the purpose to

U
investigate whether and how motor recovery could be modified at 6 months after upper
AN
limb physical therapy with a planar end-effector robot. We hypothesise that the

rehabilitation treatment with the robotic system would introduce continued and
M

sustained improvements of the upper-limb functional status in subacute stroke patients.

D
TE

Materials and methods

EP

patients

A Randomized Controlled Trial (RCT) has been carried out, in two Italian
C

rehabilitation centres having the same facilities, by enrolling eligible hemiparetic

AC

ischemic and haemorrhagic stroke survivors 30 ± 7 days after the acute event and with

moderate to severe upper limb impairment. The recruitment were performed from July

2013 to May 2014. The diagnosis were confirmed by computed tomography scan and/or

magnetic resonance imaging exams.

 ACCEPTED MANUSCRIPT
Inclusion criteria were: (a) first acute event of cerebrovascular stroke; (b) unilateral

moderate to severe paresis: score between 7 and 38 of the upper limb part of Fugl-

Meyer assessment test (Fugl-Meyer AR. et al. 1975), which has a maximum of 66

points; (c) ability to understand and follow simple instructions; (d) ability to maintain

the sitting position.

PT
Exclusion criteria were: (e) bilateral impairment; (f) severe sensory deficits in the

paretic upper limb; (g) cognitive impairment or behavioural dysfunction which could

RI
influence the ability to comprehend or perform the experiment; (h) interruption of the

SC
treatment for more than 3 consecutive days; (i) refusal or inability to provide informed

consent; (l) presence of other severe medical conditions. The local Ethics Committee of

U
the coordinator centre approved the study and all patients signed the informed consent.
AN
Since this follow-up study aims to extend our previous analysis, 53 patients (from

Sale et al., 2014) who met the inclusion criteria had been involved in the study.
M

However, 13 patients withdrawn at follow-up. In order to make this study consistent

D

with our previous one, we recruited other 13 new patients who were eligible for the
TE

study.

The subset of patients in common with the original study were randomly assigned to
EP

the control and experimental groups based upon dedicated software, while the new

patients were blindly assigned by a physician, not involved in this study, to the groups
C

in order to balance the number of subjects as follows: 27 to the Experimental Group

AC

(EG) and 26 to the Control Group (CG). 5 patients (2 EG and 3 CG) were excluded

from the analysis because they used robotic systems for rehabilitation during the follow-

up period.

Thus, data processing was carried out on 48 people (22 women and 26 men; 42 with

ischemic lesions and 6 with haemorrhagic stroke lesions; 28 right and 20 left lesion
 ACCEPTED MANUSCRIPT
side): 25 (74.0 [68.3;80.3] years old) in the EG, and 23 (70.0 [60.3;72.0] years old) in

the CG.

procedure

In the previous study, data was collected at the beginning, after 15 sessions of

PT
treatment, and at the end of the treatment (after 30 sessions).

RI
To satisfy this long-term study, the clinical assessments carried out previously at the

beginning (T0) and at the end of the treatment (T1) were analyzed in addition to

SC
clinical assessments collected at 6-months follow-up (T2). Follow-up assessments were

U
performed either at the hospital or at the patient’s home, in case of inability to reach the
AN
hospital.

The random allocation to treatment was concealed and based upon dedicated
M

software. A Lehemer algorithm was applied to achieve a balanced allocation in the EG

D

and CG groups. Therapists were randomly assigned to patients within each group using
TE

the same algorithm. The clinical assessments were carried out by blinded evaluator.
EP

intervention
C

Between T0 and T1, all subjects underwent conventional physiotherapy sessions

AC

according to an individually tailored exercise scheduling. In addition, patients also

received one daily session of upper limb training.

Subjects in the EG performed 30 sessions (5 days/week for 6 weeks) of goal-

directed, planar reaching tasks involving shoulder and elbow movements, moving from

a central target to each of the 8 peripheral targets, equally spaced on a .14 m radius

circumference, and vice versa, using the planar end-effector InMotion2 robotic system
 ACCEPTED MANUSCRIPT
(Interactive Motion Technologies, Inc., Cambridge, MA, USA). During each session, a

series of 16 unassisted clockwise repetitions to each target followed by 3 series of 320

assisted (using an adaptive control technique) clockwise repetitions were performed by

patients. At the end of each assisted series, an additional series of 16 unassisted

clockwise movements was performed. After 45 minutes the session was stopped.

PT
Subjects in the CG performed 30 sessions (5 days/week for 6 weeks) of upper limb

traditional physical therapy, with amount and type of movements matched to the robot-

RI
assisted therapy: assisted stretching, shoulder and arm exercises, and functional

SC
reaching tasks. Treatments were provided by senior physiotherapists. In both groups,

every lost session was retrieved and subjects who were not able to retrieve sessions or

U
interrupted the treatment for more than 3 consecutive days were excluded from the
AN
study. Between T1 and T2, all patients conducted a further personalized rehabilitation

programme including a specific upper limb TT, targeted at motor function improvement
M

(i.e. soft tissue and joint mobilization, neuromuscular re-education strategies,

D

progressive resistive restorative methods, etc.). All patients guaranteed they would not
TE

perform any RT during the follow-up period.

EP

clinical outcome measures

C

Clinical outcomes were evaluated using a set of valid and reliable assessment scales.
AC

The primary outcome measure was the upper limb portion of Fugl-Meyer assessment

test (FM-UL) (Fugl-Meyer et al., 1975). The secondary outcome measures were: the

pROM (Riddle et al., 1987) as sum of shoulder and elbow movements (shoulder

flexion/extension, abduction, intra/extra rotation and elbow extension) in order to assess

joints ranges correlated to spasticity (range 0-910°); and the MAS-S and MAS-E as
 ACCEPTED MANUSCRIPT
spasticity evaluation tool (Sommerfeld et al., 2012). Cognitive impairment and

behaviour dysfunction were assessed by clinical examinations.

sample size calculation and statistical analysis

The sample size was determined by assuming a 2-tailed independent t-test with

PT
power equal to .80 and alpha equal to .05, considering a delta value of .15 on the

RI
improvement of FM-UL from baseline to 6-months follow-up, as a primary outcome

measure, with respective standard deviations value of .21, in order to detect a Minimally

SC
Clinically Important Difference (Shelton et al., 2001). The sample of 53 subjects

resulted using the Lehr’s formula (Lehr, 1992) and considering a 10% drop-out rate

subjects.
U
AN
To assess homogeneity of the two groups for socio-demographic, clinical data and
M

outcome measures, the Mann-Whitney U Test was utilized as a nonparametric test to

compare median scores. The Fisher's Exact Test was used to test frequencies.
D
TE

The outcome values at T1 and at T2 were compared in a whiting object design by the

Wilcoxon’s signed-rank test in EG and CG.

EP

The Mann-Whitney U test was applied to compare the median outcome value of the

two groups at T1 and T2.

C
AC

The alpha level for significance was set at p-value<0.05. Statistical analysis was

performed using the "IBM SPSS version 23.0" software packages.

Results

No serious adverse effects occurred during or after the experimental treatment.

 ACCEPTED MANUSCRIPT
Table 1 describes the sample (n=48) at baseline and no significant differences

between the two groups were shown in terms of gender, aetiology, lesion side, age,

onset-admission interval, and clinical scales at T0.

The intra-group analysis between T0 and T1 (treatment change) and between T1 and

T2 (follow-up change) is represented in table 2. Treatment changes demonstrate

PT
significant improvements in FM-UL in both EG (p-value < .001) and CG (p-value =

.010), with a higher increase in the EG. On the other hand, MAS-S (p-value = .20),

RI
MAS-E (p-value = .27) and pROM (p-value < .001) change significantly in the EG

SC
only. The follow-up changes are significantly different in MAS-S only: the spasticity at

shoulder level increases in the CG only (p-value= .042). At T2, the FM-UL, pROM and

U
MAS-E maintain the results obtained at T1 in both groups.
AN
Table 3 shows the inter-group analysis at T1 and T2. The variations of FM-UL at T1

(p-value = .011) and T2 (p-value = .010) are significantly different between groups and
M

the EG reports higher scores at T2 than the CG. Significant differences between groups
D

in pROM and MAS-S occurred at the follow-up only. In detail, the EG shows higher
TE

values of pROM, while the CG reveals higher MAS-S. The spasticity at elbow level do

not present any substantial between group changes.

C EP

Discussion
AC

This paper presents the results of a RCT study carried out by two Italian

rehabilitation centres with the aim of analysing the long-term effects of upper limb RT

in subacute stroke patients. Previous studies on the time course of motor recovery after

stroke reported that the gains in motor function occur up to 6 months after the acute

event (Buma et al., 2013), and such increases may be related to the timing and intensity

of stroke rehabilitation services (Kwakkel et al., 2004). The recent introduction of

 ACCEPTED MANUSCRIPT
robotics in upper limb rehabilitation produced a large amount of data on short-term

effects of this kind of therapy on stroke survivors (Babaiasl et al., 2016; Mehrholz et al.,

2015; Kwakkel and Meskers, 2014; Yoo and Kim, 2015), while literature on long-term

effects in subacute stroke patients is still heterogeneous. In some cases, patients

receiving RT obtained greater improvements both at the end of treatment and at follow-

PT
up (Krebs et al., 2000; Taveggia et al., 2016), while others did not reported any further

long-term benefits (Burgar et al., 2011; Lum et al., 2006; Masiero et al., 2007). In this

RI
context, our study aims to investigate whether and how RT changes the time course of

SC
upper limb motor recovery in subacute stroke patients, extending our previous study of

robot-assisted therapy short-term efficacy (Sale et al., 2014).

U
The primary outcome (FM-UL) showed a significant increment at the end of the
AN
treatment in both CG and EG, with higher values in the EG. The obtained results at T1

affirm the findings in literature (Burgar et al., 2011; Masiero et al., 2007; Taveggia et
M

al., 2016; Yoo and Kim, 2015), and they are in accordance with our previous study
D

(Sale et al., 2014), as expected. The results of motor impairment did not demonstrate
TE

worsening in any patients of both groups, between the end of the treatment and the

follow-up, since maintained the achieved benefits following the treatments. The inter-
EP

group differences of FM-UL scores at the end of the treatment were maintained at

follow-up. Such results are in agreement with Lum et al., (2006) and Masiero et al.,
C

(2007) that revealed a plateau trend between the end of the treatment and the follow-up
AC

in subacute patients.

Significant variations of pROM has been revealed in the EG at the end of the therapy

only, and the inter-group analysis shows significant differences at T2. The increase of

the range of motion of shoulder and elbow after an upper limb RT is confirmed by
 ACCEPTED MANUSCRIPT
Krebs et al. (2000), where improvements of the muscle groups trained by the RT were

sustained at follow-up.

A considerable finding is represented by the statistically significant increase of

spasticity at the shoulder level in the CG between T1 and T2. The group treated using

the robotic system, on the other hand, is characterised by a small increase of muscular

PT
tone between T1 and T2, which is not statistically significant. However, the MAS-S

variations between the end of the treatment and the follow-up confirms that the RT

RI
might adversely affect the spasticity (Bovolenta et al., 2011; Posteraro et al., 2010;

SC
Taveggia et al., 2016). The stability of muscular tone in the EG at T2 is probably

responsible for better pROM values compared to the CG, whereas worsening in

U
shoulder muscular tone interferes with pROM modifications (Lum et al., 2006).
AN
Despite the growing number of published clinical trials and meta-analysis of upper

limb RT, which confirmed the efficacy of robotic systems as therapy for motor
M

recovery, many issues were identified to be adequately addressed in future clinical

D

trials. Specifically, the outcomes from studies on the use of RT in chronic patients (Lo
TE

et al., 2000; Lum et al., 2006; Klamroth-Marganska et al., 2014) suggest further

investigations on the continuous use of RT along the follow-up period. This last may
EP

lead to increase motor function improvement through the RT even after the acute phase.
C

Limitations of this study are associated with the number of participating patients and
AC

with the high variability of motor impairment scores at baseline. At T0, the EG had

higher FM-UL scores than the CG. Although such difference was not statistically

different and was smaller than the clinically important difference (Shelton et al., 2001),

it could have influenced the obtained outcomes. In the recruitment phase, we found

difficulties in enrolling subjects who satisfied the inclusion criteria (especially 30 ± 7

days after the acute event and the ability to remain in sitting posture). Although the
 ACCEPTED MANUSCRIPT
sample size calculation showed that a population of 53 subjects would be sufficient to

have significant results, a higher number of patients and multi-centric international RCT

studies would be desirable in the future.

Considering these potential limitations, our results justify the need of further studies

of short and long-term effects of RT, on the factors that may influence them, and on the

PT
use of RT in chronic phase.

RI
Conclusions

SC
This paper aims to analyse the long-term effects of upper limb therapy with a planar

U
end-effector robot in subacute stroke patients. Our results confirm that upper limb RT
AN
may reduce mostly the motor impairment in patients at the subacute stage compared to a

TT. The functional and motor improvements, obtained at the end of treatment persist in
M

the EG over time. Therefore, the robotic system could be considered as a suitable tool
D

for promoting an active motor re-learning and intensive treatments, with a potential
TE

positive economic impact on healthcare systems in terms of a better cost/effectiveness

ratio (Wagner et al., 2011).

EP

In conclusion, this study amplifies the knowledge of the changes in motor function of

subacute stroke patients over time, since it highlights the potential role of RT in neuro-
C

rehabilitation. It could suggest that only the translational research could lead to
AC

significant improvement in terms of effectiveness of rehabilitation programmes through

precise and reproducible scientific methods (Franceschini, 2012).

Found source

This research was funded by the Italian Ministry of Health (Ricerca Corrente).
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT
REFERENCES

References

Babaiasl M, Mahdioun SH, Jaryani P, and Yazdani M 2016 A review of

technological and clinical aspects of robot-aided rehabilitation of upper-extremity after

stroke. Disability and Rehabilitation: Assistive Technology 11(4): 263-280

PT
Béjot Y, Bailly H, Durier J, and Giroud M 2016 Epidemiology of stroke in Europe

RI
and trends for the 21st century. La Presse Médicale 45(12): e391-e398

Bovolenta F, Sale P, Dall'Armi V, Clerici P, and Franceschini M 2011 Robot-aided

SC
therapy for upper limbs in patients with stroke-related lesions. Brief report of a clinical

U
experience. Journal of neuroengineering and rehabilitation 8(1): 18
AN
Buma F, Kwakkel G, and Ramsey N 2013 Understanding upper limb recovery after

stroke. Restorative neurology and neuroscience 31(6): 707-722

M

Burgar CG, Garber SL, and Van der Loos HM 2011 Robot-assisted upper-limb
D

therapy in acute rehabilitation setting following stroke: Department of Veterans Affairs

TE

multisite clinical trial. Journal of rehabilitation research and development 48(4): 445

Fasoli SE, Krebs HI, and Hogan N 2004 Robotic technology and stroke
EP

rehabilitation: translating research into practice. Topics in stroke rehabilitation 11(4):

11-19
C
AC

Feys H, De Weerdt W, Verbeke G, Steck GC, Capiau C, Kiekens C, ... and Cras P

2004 Early and repetitive stimulation of the arm can substantially improve the long-

term outcome after stroke: a 5-year follow-up study of a randomized trial. Stroke 35(4):

924-929
 ACCEPTED MANUSCRIPT
Franceschini M 2012 Clinical rehabilitation needs the translation from basic

neuroscience in daily practice. European Journal of Physical and Rehabilitation

Medicine 48 (1): 87-89

Fugl-Meyer AR, Jääskö L, Leyman, I, Olsson S, and Steglind S 1975 The post-stroke

hemiplegic patient. 1. a method for evaluation of physical performance. Scandinavian

PT
journal of rehabilitation medicine 7(1): 13-31

Klamroth-Marganska V, Blanco J, Campen, K, Curt A, Dietz V, Ettlin T, ... and Luft

RI
A 2014 Three-dimensional, task-specific robot therapy of the arm after stroke: a

SC
multicentre, parallel-group randomised trial. The Lancet Neurology 13(2): 159-166

Krebs HI, Volpe BT, Aisen ML, and Hogan N 2000 Increasing productivity and

U
quality of care: Robot-aided neuro-rehabilitation. Journal of rehabilitation research and
AN
development 37(6): 639
M

Kwakkel G, Kollen B, and Lindeman E 2004 Understanding the pattern of functional

recovery after stroke: facts and theories. Restorative neurology and neuroscience 22(3-
D

5): 281-299
TE

Kwakkel G and Meskers CG 2014 Effects of robotic therapy of the arm after stroke.
EP

The Lancet Neurology 13(2): 132-133

Langhorne P, Coupar F, and Pollock A 2009 Motor recovery after stroke: a

C

systematic review. The Lancet Neurology 8(8): 741-754

AC

Langhorne P, Bernhardt J and Kwakkel G 2011 Stroke rehabilitation. The Lancet

377(9778): 1693-1702

Lehr R 1992 Sixteen S squared over D squared: A relation for crude sample size

estimates. Statistics in medicine 11(8): 1099-1102

 ACCEPTED MANUSCRIPT
Lo AC, Guarino PD, Richards LG, Haselkorn JK, Wittenberg GF, Federman DG, ...

and Bever Jr CT 2010 Robot-assisted therapy for long-term upper-limb impairment

after stroke. New England Journal of Medicine 362(19): 1772-1783

Lohse KR, Lang CE, and Boyd LA 2014 Is more better? Using metadata to explore

dose–response relationships in stroke rehabilitation. Stroke 45(7): 2053-2058

PT
Lum PS, Burgar CG, Van der Loos M, and Shor PC 2006 MIME robotic device for

upper-limb neurorehabilitation in subacute stroke subjects: A follow-up study. Journal

RI
of rehabilitation research and development 43(5): 631

SC
Masiero S, Celia A, Rosati G, and Armani M 2007 Robotic-assisted rehabilitation of

the upper limb after acute stroke. Archives of physical medicine and rehabilitation

88(2): 142-149
U
AN
Masiero S, Armani M, Ferlini G, Rosati G, and Rossi A 2014 Randomized trial of a
M

robotic assistive device for the upper extremity during early inpatient stroke

rehabilitation. Neurorehabilitation and neural repair 28(4): 377-386

D
TE

Mehrholz J, Pohl M, Platz T, Kugler J, and Elsner B 2015 Electromechanical and

robot assisted arm training for improving activities of daily living, arm function, and
EP

arm muscle strength after stroke. The Cochrane Library

Posteraro F, Mazzoleni, Aliboni S, Cesqui B, Battaglia A, Carrozza MC, ... and

C

Micera S 2010 Upper limb spasticity reduction following active training: a robot-
AC

mediated study in patients with chronic hemiparesis. Journal of rehabilitation medicine:

42(3), 279-281

Prange GB, Kottink AI, Buurke JH, Eckhardt MM, van Keulen-Rouweler BJ,

Ribbers, GM, and Rietman JS 2015 The effect of arm support combined with
 ACCEPTED MANUSCRIPT
rehabilitation games on upper-extremity function in subacute stroke: a randomized

controlled trial. Neurorehabilitation and neural repair 29(2): 174-182

Rabadi MH, Galgano M, Lynch D, Akerman M, Lesser M, and Volpe BT 2008 A

pilot study of activity-based therapy in the arm motor recovery post stroke: a

randomized controlled trial. Clinical Rehabilitation 22(12): 1071-1082

PT
Rand D and Eng JJ 2015 Predicting daily use of the affected upper extremity 1 year

after stroke. Journal of Stroke and Cerebrovascular Diseases 24(2): 274-283

RI
Riddle DL, Rothstein JM, and Lamb RL 1987 Goniometric reliability in a clinical

SC
setting: shoulder measurements. Physical therapy 67(5): 668-673

U
Sale P, Franceschini M, Mazzoleni S, Palma E, Agosti M, and Posteraro F 2014
AN
Effects of upper limb robot-assisted therapy on motor recovery in subacute stroke

patients. Journal of neuroengineering and rehabilitation 11(1): 104

M

Shelton FDN, Volpe BT, and Reding M 2001 Motor impairment as a predictor of
D

functional recovery and guide to rehabilitation treatment after stroke.

TE

Neurorehabilitation and neural repair 15(3): 229-237

Sommerfeld DK, Gripenstedt U, and Welmer AK 2012 Spasticity after stroke: an

EP

overview of prevalence, test instruments, and treatments. American Journal of Physical

Medicine & Rehabilitation 91(9): 814-820

C
AC

Taveggia G, Borboni A, Salvi L, Mule C, Fogliaresi S, Villafañe JH, and Casale R

2016 Efficacy of robot-assisted rehabilitation for the functional recovery of the upper

limb in post-stroke patients: a randomized controlled study. European journal of

physical and rehabilitation medicine 52(6): 767-773

 ACCEPTED MANUSCRIPT
van Kordelaar J, van Wegen E, and Kwakkel G 2014 Impact of time on quality of

motor control of the paretic upper limb after stroke. Archives of physical medicine and

rehabilitation 95(2): 338-344

Veerbeek JM, Langbroek-Amersfoort AC, van Wegen EE, Meskers CG, and

Kwakkel G 2017 Effects of Robot-Assisted Therapy for the Upper Limb After Stroke A

PT
Systematic Review and Meta-analysis. Neurorehabilitation and Neural Repair 31(2):

107-121

RI
Waddell KJ, Birkenmeier RL, Moore JL, Hornby TG, and Lang CE 2014 Feasibility

SC
of high-repetition, task-specific training for individuals with upper-extremity paresis.

American Journal of Occupational Therapy 68(4): 444-453

U
Wagner TH, Lo AC, Peduzzi P, Bravata DM, Huang GD, Krebs HI, ... and
AN
Wittenberg GF 2011 An economic analysis of robot-assisted therapy for long-term
M

upper-limb impairment after stroke. Stroke 42(9): 2630-2632

Yoo DH and Kim SY 2015 Effects of upper limb robot-assisted therapy in the
D

rehabilitation of stroke patients. Journal of physical therapy science 27(3): 677-679

TE
C EP
AC
 ACCEPTED MANUSCRIPT
TABLES

Titles of tables

table 1 - Socio-demographic and clinical data of the patients at baseline T0.

table 2 – intra-group changes in clinical outcome measures.

PT
table 3 – inter-group changes in clinical outcome measures.

RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

table 1 - Participant characteristics at baseline (n = 48).

CG EG
Characteristics p-value
n (%) Median [IQR]
Subject 23 (47.9) 25 (52.1)

PT
Gender, female/male 9 (39.1)/14 (60.9) 13 (52.0)/12 (48.0) .401a

Aetiology,

RI
19 (82.6)/4 (17.4) 23 (92.0)/2 (8.0) .407a
ischemic/haemorrhagic

SC
Lesion Side, right/left 15 (65.2)/8 (34.8) 13 (52.0)/12 (48.0) .394a

Age, years 74.0 [68.3;80.3] 70.0 [60.3;72.0] .051b

U .733b
AN
OAI, days 31.0 [29.0;33.0] 31.0 [28.0;33.0]

FM-UL 24.0 [7.5;38.3] 32.0 [17.8;38.0] .231b

pROM 820.0 [742.5;860.0] 780.0 [707.5;857.5] .457b
M

MAS-S 1.0 [.3;2.0] 1.0 [.0;2.0] 1.000b

MAS-E 1.0 [.0;1.8] 1.0 [.0;2.0] .337b
D

a
Fisher's exact test; b Mann-Whitney U test (two-tailed).
TE

Abbreviations: CG, Control Group; CG, Experimental Group; IQR, Interquartile Range; OAI, Onset
Admission Interval; FM-UL, Fugl-Meyer upper limb; pROM, passive Range Of Motion; MAS-S,
Modified Ashworth Scale-Shoulder; MAS-E, Modified Ashworth Scale-Elbow.
C EP
AC
 ACCEPTED MANUSCRIPT

table 2 – Intra-group changes in clinical outcome measures.

Follow-up
Treatment change
T1 T2 change
(T1-T0)
Outcome (T2-T1)
Group
measures intra-group estimated median of
Median [IQR] difference (95% CI)
[p-value]a
2.0 (.0;3.0) 3.0 (.0;11.0)
FM-UL CG 23.0 [10.3;44.3] 31.0 [15.5;39.3]

PT
[.010] [.123]

9.0 (7.2;12.0) 1.0 (-1.2;3.9)

EG 43.0 [28.5;48.0] 45.0 [33.8;50.0]
[<.001] [.465]

RI
.0 (-61.1;20.0
810.0 767.6 .0 (.0;13.0 )
pROM CG )
[745.0;860.0] [732.5;837.5] [.765]

SC
[.651]

860.0 860.0 80.0 (40.7;104.3 ) .0 (-55.5;9.3)

EG
[775.0;910.0] [805.6;912.5] [<.001] [.396]

U
.0 (.0;.0) .0 (.0;.9)
MAS-S CG 1.0 [.0;2.0] 1.6 [1.0;2.0]
[1.000] [.042]
AN
.0 (-.9;.0) .0 (.0;.0)
EG 1 [.0;1.3] .9 [.0;1.3]
[.020] [ .791]
.0 (.0;.0) .0 (-1.0;.0)
MAS-E CG 1.0 [.0;2.0] 1.2 [1.0;2.0]
[.461] [.131]
M

.0 (.0;.7) .0 (.0;1.0)
EG 1.0 [.0;1.0] 1.0 [.0;2.0]
[.027] [.095]
D

a
Wilcoxon’s signed-rank test (two tailed), in bold statistical significance at p ≤ .05.
TE

Abbreviations: T0, Baseline; T1, End of the treatment; T2, 6-months follow-up; CG, Control Group; EG,
Experimental Group; IQR, Interquartile Range; CI, Confidence Interval; FM-UL, Fugl-Meyer upper limb;
pROM, passive Range Of Motion; MAS-S, Modified Ashworth Scale-Shoulder; MAS-E, Modified
Ashworth Scale-Elbow.
C EP
AC
 ACCEPTED MANUSCRIPT
table 3 – Inter-group changes in clinical outcome measures.

CG (n=23) EG (n=25)
Outcome measures Time ES p-valuea
Median [IQR]

FM-UL T1 23.0 [10.3;44.3] 43.0 [28.5;48.0] -.878 .011

T2 31.0 [15.5;39.3] 45.0 [33.8;50.0] -.733 .010

pROM T1 810.0 [745.0;860.0] 860.0 [775.0;910.0] -.593 .088

PT
T2 767.6 [732.5;837.5] 860.0 [805.6;912.5] -.399 .003

MAS-S T1 1.0 [.0;2.0] 1 [.0;1.3] .364 .287

RI
T2 1.6 [1.0;2.0] .9 [.0;1.3] .606 .032

MAS-E T1 1.0 [.0;2.0] 1.0 [.0;1.0] .276 .534

SC
T2 1.2 [1.0;2.0] 1.0 [.0;2.0] .224 .229
a
Mann-Whitney U test (two tailed), in bold statistical significance at p ≤ .05.

U
Abbreviations: T1, End of the treatment; T2, 6-months follow-up; CG, Control Group; EG, Experimental
AN
Group; IQR, Interquartile Range; ES, Effect Size; FM-UL, Fugl-Meyer upper limb; pROM, passive
Range Of Motion; MAS-S, Modified Ashworth Scale-Shoulder; MAS-E, Modified Ashworth Scale-
Elbow.
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
Authors Declaration

All authors have seen and approved the final version of the manuscript submitted. They warrant that
the article is the authors' original work, hasn't received prior publication and isn't under
consideration for publication elsewhere.

Moreover, the authors declare that there's no financial/personal interest or belief that could affect
their objectivity.

PT
Conflict of Interest: none

RI
Sanaz Pournajaf

SC
On behalf of authors

U
AN
M
D
TE
C EP
AC