Research ArticleAction Observation Treatment Improves Upper LimbMotor Functions in Children with Cerebral Palsy: A CombinedClinical and Brain Imaging Study

Giovanni Buccino ,1 AnnaMolinaro,2,3 Claudia Ambrosi,4Daniele Arisi,5 Lorella Mascaro,6

Chiara Pinardi,7 Andrea Rossi,2 Roberto Gasparotti,4 Elisa Fazzi,2,3 and Jessica Galli2,3

1Department of Medical and Surgical Sciences, University of Magna Graecia, Catanzaro, Italy2Unit of Child Neurology and Psychiatry, ASST Spedali Civili, Brescia, Italy3Department of Clinical and Experimental Sciences, University of Brescia, Brescia, Italy4Department of Diagnostic Imaging, Neuroradiology Unit, University of Brescia, Brescia, Italy5Department of Paediatrics, Ospedale di Cremona, Cremona, Italy6Department of Diagnostic Imaging, Medical Physics Unit, ASST Spedali Civili, Brescia, Italy7Neuroscience Unit, Department of Medicine and Surgery, University of Parma, Parma, Italy

Correspondence should be addressed to Giovanni Buccino; buccino@unicz.it

Received 29 December 2017; Revised 27 March 2018; Accepted 26 April 2018; Published 4 July 2018

Academic Editor: Michela Bassolino

Copyright © 2018 Giovanni Buccino et al. This is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work isproperly cited.

The aim of the present study was to assess the role of action observation treatment (AOT) in the rehabilitation of upper limb motorfunctions in children with cerebral palsy. We carried out a two-group, parallel randomized controlled trial. Eighteen children (aged5–11 yr) entered the study: 11 were treated children, and 7 served as controls. Outcome measures were scores on two functionalscales: Melbourne Assessment of Unilateral Upper Limb Function Scale (MUUL) and the Assisting Hand Assessment (AHA).We collected functional scores before treatment (T1), at the end of treatment (T2), and at two months of follow-up (T3). Ascompared to controls, treated children improved significantly in both scales at T2 and this improvement persisted at T3. AOThas therefore the potential to become a routine rehabilitation practice in children with CP. Twelve out of 18 enrolled childrenalso underwent a functional magnetic resonance study at T1 and T2. As compared to controls, at T2, treated children showedstronger activation in a parieto-premotor circuit for hand-object interactions. These findings support the notion that AOTcontributes to reorganize brain circuits subserving the impaired function rather than activating supplementary or vicariating ones.

1. Introduction

There is an urgent need in neurorehabilitation of bothadults and children of approaches that take into accountthe progresses of our knowledge in basic neuroscience.These approaches should aim at transferring ideas and factsfrom basic neuroscience to clinical practice with the finalgoal to build up tools well-grounded in neurophysiologyand to provide a cure for several neurological (and non-neurological) diseases [1, 2]. Such a rehabilitation approach,

grounded in basic neuroscience, would also be a model oftranslational medicine.

The use of such approaches may help to overwhelm ageneral attitude in neurorehabilitation to focus on waysto circumvent functional deficits, thus leading to a compen-sation or a reeducation of functions rather than a cure forthem through remediation (for a more general discussionon the notion of compensation and remediation, see [3–5]).Although compensation sometimes works and helps patientsto recover in daily activities, it does not aim at repairing the

HindawiNeural PlasticityVolume 2018, Article ID 4843985, 11 pageshttps://doi.org/10.1155/2018/4843985

neural circuits underlying specific functions through a director indirect restoration. Moving to a translational model inneurorehabilitation would imply to plan specific rehabilita-tive tools aiming at restoring the neural structures whosedamage caused the impaired functions or activating sup-plementary or related pathways, which may perform theoriginal functions. Last, but not least, rehabilitation toolswell-grounded in basic neuroscience allow researchers toplan well-designed randomized controlled trials. This in turnallows clinicians and therapists to measure outcomes notonly in terms of functional and/or behavioural gains (as itcurrently happens by means of functional scales) but alsoin terms of changes in biological parameters, whichresearchers can test using neurophysiological and brainimaging techniques. There are indeed some approaches inthe neurorehabilitation of children that fit these criteria. Forexample, constraint-induced movement therapy (CIMT)has a well-established neurophysiological basis grounded onthe experimental evidence that monkeys can be induced touse a deafferented limb by restricting movements of the unaf-fected limb over a period of days [6]. CIMT has been widelyapplied in patients with acute and chronic stroke and in chil-dren with cerebral palsy [7]; similarly, HABIT (hand-armbimanual intensive training) is a highly structured form ofbimanual training, whose goal is to improve the quality andquantity of hand use in bimanual tasks in children withhemiplegic CP [7]. Another example is the mirror therapy[8]. In this treatment, a mirror is placed in the patient’s mid-sagittal plane so that he/she can see her unaffected arm/handas if it were the affected one. This strategy has been proven tobe effective to relieve phantom pain in arm amputees as wellas in the recovery of upper limb in chronic stroke patientsand in children with cerebral palsy [9, 10]. This approachgrounds on a neurophysiological mechanism known asmirror mechanism. Based on this mechanism, the observa-tion of actions performed by other individuals recruits inthe observer the same areas involved in the actual executionof those same actions [11]. In the case of mirror therapy,patients have the opportunity to look at their own actionsperformed with the unaffected arm/hand. More recently, weproposed a novel approach in neurorehabilitation known asaction observation treatment (for a review, see Buccino[12]). AOT exploits the mirror mechanism in an evenmore straightforward manner than mirror therapy, becausepatients observe daily actions performed by other healthyindividuals. During one typical session, patients observe adaily action and afterwards execute it in context. So far, thisapproach has been successfully applied in the rehabilitationof upper limb motor functions in chronic stroke patients, inmotor recovery of Parkinson’s disease patients, includingthose presenting with freezing of gait; interestingly, thisapproach also improved lower limb motor functions in post-surgical orthopaedic patients [13–16]. Pivotal studies wereconducted also in children with cerebral palsy [17–19].AOT is well-grounded in basic neuroscience, thus represent-ing a valid model of translational medicine in the field ofneurorehabilitation. Moreover, the results concerning itseffectiveness have been collected in randomized controlledstudies, thus being an example of evidence-based clinical

practice. The present study aimed at assessing whether thisnovel rehabilitation approach has the potential to improvethe functional recovery of children with CP aged 5–11(primary school cycle in Italy), within a comprehensive reha-bilitation program. The focus was on the recovery of upperlimb motor functions. We used the same protocol of an ear-lier pilot study from our group [17]. We also tested whetherthis approach may lead to neural changes in the brain bymeans of a functional magnetic resonance study, in whichwe asked some of the children that entered the study tomanipulate complex objects in the scanner. Control condi-tion was the manipulation of a small sphere.

2. Methods

2.1. Study Design.We used a two-group, parallel randomizedcontrolled trial. Recruitment criteria and methodologicalprocedures were approved by the Ethical Committee of theHospital of Brescia.

2.2. Participants. All children referred to the Centre of ChildNeurology and Psychiatry at the Hospital of Brescia with adiagnosis of cerebral palsy (CP) were eligible. Inclusioncriteria were the presence of CP confirmed by neuroimagingtechniques (MRI), Manual Ability Classification System(MACS)≤ 4 [20], verbal IQ> 70, age between 5 and 11(primary school cycle in Italy), absence of major visual and/orauditory deficits, and no antiepileptic treatment. We enrolleda group of 18 children that met the inclusion/exclusion cri-teria. Before entering the study, the parents of each child gavewritten informed consent.

2.3. Allocation and Assessment. Patients were enrolled byone of the authors (Elisa Fazzi); enrolled children were ran-domly allocated to the treatment (n = 11) or the control group(n = 7) by means of a dedicated software. Both children andtheir parents were blind to group allocation. After randomiza-tion, children were evaluated clinically with a neurologicalexamination carried out by two expert child neurologists(Elisa Fazzi, Anna Molinaro), while functional assessmentwas carried out by a physician blind to treatment allocation,using the Melbourne Assessment of Unilateral Upper LimbFunction Scale (MUUL) and the Assisting Hand Assessment(AHA).MUUL consists of 16 items involving reaching, grasp-ing, releasing, and manipulation, specifically developed tomeasure quality of upper limb motor functions in childrenwith CP aged 5 to 16 [21]. It has been shown to have a goodreliability on a sample of 20 children with different severitydegrees of CP. AHA is a hand function evaluation instrumentthat measures and describes how children with an upper limbdisability in one hand use his/her affected hand collabora-tively with the nonaffected hand in bimanual actions [22]. Inthe present study, children underwent functional evaluationwith MUUL and AHA at three different time points: at base-line (T1), at the end of the treatment (T2), and at two monthsof follow-up (T3).

2.4. Stimuli.We prepared fifteen video clips to be used duringAOT in the treatment group, each showing a specific dailyaction implying the use of the arms/hands, (i.e., grasping an

2 Neural Plasticity

object, using a pencil, and playing with Lego). All recordedactions were chosen among those which are familiar to chil-dren in primary school age. We used the same videos as in aprevious study from our group [17]. In that study, we reportalso a complete list of all seen actions. In the videos, theseeveryday actions, performed both by normal children andadults, were recorded from different perspectives, to makethe video clips more interesting and to sustain the attentionof children during the rehabilitation sessions. Each actionwas subdivided into 3 or 4 constituent motor segments. Forinstance, eating a candy, one of the shown actions, was sub-divided into taking the candy from the table, approaching itto the mouth, and giving back to the therapist. Each motoract was presented for 3 minutes so that the total duration ofeach video clip was 9–12 minutes. We also prepared the samenumber of video clips addressing various topics (geography,history, and science adapted for children) but with no motorcontent, for the control group. Video clips for the controlgroup were also divided into three-four parts, each lasting3 minutes.

2.5. Treatment Procedure. For 3 weeks, children in thetreatment group attended daily rehabilitation sessions fromMonday to Friday, during which they were asked to observeone movie showing an actor/an actress performing onespecific daily action with the hand. Actions were presentedin a fixed order according to their complexity, as judged bythe experimenter.

After observation of each motor segment (3-4 per eachvideo clip), children were required to execute for 2 minuteswhat observed to the best of their ability. They were advisedthat the quality of their imitation was not the goal of the reha-bilitation treatment. Children in the control group viewedshort video clips (for the same time as treated participants)showing scenes with no motor content (e.g., geographicaldocumentaries). After observing each part of a video clip(3-4 parts per each video clip), controls were also asked toexecute the same actions as treated participants for the sameduration. In this way, the total amount of visual stimulationand motor activity following observation was similar in thetwo groups. The only difference concerned the content ofvideos: treated participants observed videos with motor con-tent (everyday arm/hand actions), while controls observedvideos with no specific motor content. As a whole, eachrehabilitation session lasted about half an hour. The physio-therapist devoted up to 10 minutes to explain the task andencourage children to observe carefully the videos andperform the seen actions at their best. Twelve minutes wasdevoted to observation (motor acts for cases, documentariesfor controls) and 8 minutes to the execution of the observedactions (cases) or just execution of the same actions, butwithout a model (controls).

Both treated participants and controls received writteninstructions. The physiotherapist read them aloud twice.This was in order to avoid any influence of the physiothera-pist in giving instructions.

During the treatment, children continued to follow theirroutine conventional rehabilitation program that was the

same for cases and controls. All children (treated participantsand controls) completed the study.

2.6. Outcome Measures. Primary outcome measures werescore changes on the MUUL and AHA.

2.7. Statistical Analysis. A mixed linear model, with fixedeffects: time (T1, T2, and T3) and group (treatment, control),was carried out on MUUL and AHA scores. The best modelwas identified using the Akaike information criterion (AIC).The significance level was set at 0.05. Statistical analyses werecarried out using SPSS version 23.

2.8. fMRI Study. Twelve children (six treated participants)out of 18 enrolled children also entered an fMRI study toassess a reorganization of brain neural structures followingtreatment. While being scanned, children with CP from bothgroups manipulated complex objects with both hands, inorder to explore all the motor properties of the manipulatedobject. As a control condition, children manipulated a simpleobject, a sphere. All objects used in the scanner were differentfrom those used during the treatment. Figure 1(a) shows theexperimental paradigm. fMRI data were collected on a 1.5TSiemens Avanto scanner. The protocol included four EPIsequences (TR/TE 2500/50ms, 3.3× 3.3× 3.3mm isotropicvoxel) and a high resolution T1W 3D MP-RAGE sequencefor anatomical reference (TR/TE 2050/2.56ms, 1× 1× 1mmisotropic voxel). Imaging data were collected before startingtreatment (T1) and at the end of treatment (T2). ThefMRI paradigm consisted of 14 alternating task-rest blocks(8 volumes/block were acquired) repeated 4 times to increasestatistics. fMRI data underwent the following preprocessing.The mean EPI was first computed for each participant andvisually inspected to ensure that none showed artifacts. Thefirst 2 EPI volumes of each functional run were discardedto allow for T1 equilibration effects. For each subject, all vol-umes were spatially realigned to the mean volume of the fourruns. Next, the 3D structural data of each subject were nor-malized to the ANTS standard space, a T1 pediatric templatein a standardized MNI space [23]. The normalization matrixwas subsequently transferred to the fMRI images, resampledin 1mm× 1mm× 1mm voxels using trilinear interpolationin space and then the images were spatially smoothed witha 6mm full width at half maximum isotropic Gaussian kernelfor the group analysis. No participant showed head move-ments greater than 3mm; thus, none was excluded fromfurther analyses.

Data were analyzed using a random effects model [24],implemented in a two-level procedure. In the first level,single-subject fMRI data entered an independent generallinear model (GLM) by design matrixes modelling the onsetsand durations of two experimental factors, one related to theexperimental task and one related to its corresponding base-line. For each participant, we generated contrast imagesdisplaying the effect of the experimental task (manipulatingcomplex objects) contrasted with the respective baseline(manipulating a sphere). Next, each contrast entered a secondlevel GLM to obtain (i) SPM{T} maps (one sample t-test)related to each task at group level and (ii) to test for the

3Neural Plasticity

existence of brain areas specifically involved in manipulatingcomplex objects. Moreover, we were interested in assessingdifferences in brain area recruitment between treated childrenand controls. For all analyses, location of the activation fociwas determined in the stereotaxic space of the MNI coordi-nates system. A significance level of p < 0 001 uncorrectedand an extended threshold on cluster dimension of 10 voxelswas applied.

3. Results

Demographic data, clinical features, and brain imaging find-ings of children in the two groups are shown in Table 1.Mean scores and SD of AHA and MUUL in treated partici-pants and controls at T1, T2, and T3 are shown in Table 2.

Mixed linear model showed that for MUUL, the bestmodel included the random effects of intercept and the fixedeffect of group, time, and interaction. For the AHA, the bestmodel included the random effects of intercept and timeand the fixed effects of group and interaction. The mixed lin-ear model analysis disclosed a significant interaction betweentime and treatment, both for MUUL (b1 interaction= 2.71,t36 = 3 99, and p = 0 000) and for AHA (b1 interaction= 2.36,t18 = 3 61, p = 0 002). Score improvements, in both scales,were higher in the treated participants than in the controls;furthermore, in the treatment group, those improvementswere not only maintained but became even stronger at T3.

Post hoc analysis showed that for MUUL, results at T2were significantly different from results at T1 only in cases(p < 0 001), but not in controls. As for AHA, results at T2were significantly different from results at T1 (p < 0 001).Even more interestingly, results at T3 were different from

results at T2 (p < 0 001) for both scales, but again onlyin cases, but not in controls. Figure 2 shows the results.

4. fMRI Results

For the aim of the present study, we will present resultsconcerning the differences between treated children andcontrols. It is worth stressing that at baseline (T1), there wereno differential activations when comparing cases versus con-trols. In contrast, after treatment (T2), differential activationswere located in the left premotor cortex extending to the infe-rior frontal gyrus (−49; 19; 26), in the right premotor cortex(53; 14; 31), in the left supramarginal gyrus (−47; −51; 37),and finally a weaker activation in the left superior tem-poral gyrus (−52; −47; 23). Figures 1(a) and 1(b) showsfMRI findings.

5. Discussion

The results of the present study are relevant within the liter-ature devoted to rehabilitation of children with cerebralpalsy. Treated children improved significantly as comparedto controls in both MUUL and AHA. These results are inkeeping with earlier, pilot studies using AOT as a rehabilita-tion tool [17, 18]. It is worth stressing that our sample con-sisted of hemiplegic (both right and left) and tetraplegicchildren, thus suggesting that AOT may be useful for differ-ent clinical presentations of CP. As reported above, AOTexploits a neurophysiological mechanism known as mirrormechanism. The observation of actions performed by otherindividuals recruits in the observer the same areas involvedin the actual execution of those same actions [11], thiswhatever the biological effector involved in the observed

20 s20 s

Sphere

Complexobject

Sphere Sphere

20 s 20 s14 blocks

Complexobject

Complexobject

(a)

Cases + controls

R

(b)

Cases > controls

R L

(c)

Figure 1: (a) Graphic representation of the fMRI experimental paradigm, alternating manipulation of a simple object (a sphere), andmanipulation of complex objects. (b) Clusters of activations transposed on sections from standard pediatric brain (ANTS) beforetreatment (T1), when comparing manipulation of complex objects versus manipulation of a sphere. Cases and controls are taken as awhole group, p < 0 001. Note that at T1, no activation was present when directly comparing cases versus controls. (c) After treatment(T2), direct comparison between cases and controls shows increased activations in frontal and parietal areas known to be involved inhand-object interactions, p < 0 001. Clusters of activations transposed on sections from standard pediatric brain (ANTS), as in (b).

4 Neural Plasticity

Table1:Dem

ograph

icdata,clin

icalfeatures,and

radiologicalfind

ings

intreatedparticipantsandcontrols.

Pt.nu

mber

Case/

control

Sex

(M,F

)GA

(wk)

Age

(yr,m)

CPtype

Hagberg

Motor

abno

rmalities

GMFC

SMACS

CFC

SAssociated

impairments

TotalIQ

Verbal

IQPerform

ance

IQRadiologicalfi

ndings

(brain

MRI)

1Case

M33

9yr,

5m

Right

hemiplegia

Unilateral

spastic

hyperton

ia2

21

V:C

VI;H:n

o;M/A

:no;LD

:no;

E:n

o85

9282

Right

tempo

rooccipital,left

occipitoparietal,bilateral

periventricular,andleft

fron

totempo

ralsub

dural

hematom

as

2Case

F27

8yr,

2m

Right

hemiplegia

Unilateral

spastic

hyperton

ia1

21

V:R

OP;H

:no;

M/A

:no;LD

:no;

E:n

o100

106

107

Mild

bilateral

periventricular

leuk

omalacia,m

ildventriculardilatation

3Con

trol

M40

7yr,

10m

Right

hemiplegia

Unilateral

spastic

hyperton

ia2

21

V:n

o;H:n

o;M/A

:no;LD

:no;

E:n

o99

101

97

Leftsubd

ural

occipitotempo

ral

hematom

aandepidural

parietotem

poralh

ematom

a;hypo

xicischem

icenceph

alop

athy

characterizedby

signal

alteration

sin

both

putamen

tailandanterior

thalam

us

4Con

trol

M34

8yr,

3m

Tetraplegia

Bilateral

spastic

hyperton

ia,

leftside

more

affected

43

2V:C

VI;H:n

o;M/A

:yes;LD:no;

E:n

o73

9750

Hypoxicischem

icinjury

withthinning

ofthecorpus

callosum,enlargementof

CSF

spaces,w

idespread

hyperintensity

ofcentrum

semiovale,coron

aradiata,

andperiventricularwhite

matter,dilation

ofthe

ventricles

5Con

trol

F40

6yr,

8m

Tetraplegia

Bilateral

spastic

hyperton

ia,

leftside

more

affected

42

3V:C

VI;H:n

o;M/A

:yes;LD:no;

E:n

o114

139

77

Diffuseperiventricular

hyperintensity

withparietal

bilateralw

hitematter

involvem

ent;mild

dilatation

ofbilateral

ventriculartrigon

e

6Case

F30

11yr,

9m

Tetraplegia

Bilateral

spastic

hyperton

ia,

leftside

more

affected

43

2V:C

VI;H:n

o;M/A

:yes;

LD:yes;E

:yes

5689

50

Periventricular

leuk

omalacia,fronto-

parieto-occipitalw

hite

matterredu

ction,

exvacuo

enlargem

entof

bilateral

ventricles

5Neural Plasticity

Table1:Con

tinu

ed.

Pt.nu

mber

Case/

control

Sex

(M,F

)GA

(wk)

Age

(yr,m)

CPtype

Hagberg

Motor

abno

rmalities

GMFC

SMACS

CFC

SAssociated

impairments

TotalIQ

Verbal

IQPerform

ance

IQRadiologicalfi

ndings

(brain

MRI)

7Con

trol

F37

9yr,

1m

Right

hemiplegia

Unilateral

spastic

hyperton

ia1

21

V:C

VI;H:n

o;M/A

:yes;LD:no;

E:n

o87

84100

Leftperiventricularmalacic

area

withgliosis,extend

edinto

thecorona

radiata;

left

corticospinalprojection

hyperintensistywithmild

cerebellarpedu

ncle

hypo

trop

hy(W

allerian

degeneration

)

8Con

trol

F31

8yr,

9m

Right

hemiplegia

Unilateral

spastic

hyperton

ia2

11

V:C

VI;H:n

o;M/A

:no;LD

:no;

E:n

o87

9977

Bilateralp

arietalcystic

periventricular

leuk

omalacia,w

ithcentrum

semiovalewhitematter

involvem

ent,shortdistance

betweencortex

and

ventricularwallsin

tempo

roparietalareas,

thinning

ofthecorpus

callosum

9Case

FNot

know

n11

yr,

9m

Left

hemiplegia

Unilateral

spastic

hyperton

ia2

21

V:n

o;H:n

o;M/A

:no;LD

:no;

E:yes

Leiter

-R82

Right

fron

to-parieto-

tempo

ralm

alacicarea,ex

vacuoenlargem

entof

the

ventricleandWallerian

degeneration

ofthe

corticospinaltract

10Case

M32

6yr,

10m

Tetraplegia

Bilateral

spastic

hyperton

ia,

rightside

moreaffected

32

2V:C

VI;H:n

o;M/A

:yes;LD:no;

E:n

o89

120

70

Periventricular

leuk

omalacia,corpu

scallosum

hypo

plasia,

hipp

ocam

palcom

missure

agenesis

11Con

trol

M38

5yr,

2m

Right

hemiplegia

Unilateral

spastic

hyperton

ia2

32

V:C

VI;H:n

o;M/A

:no;LD

:no;

E:n

o98

118

87

Corticallam

inar

necrosis

(leftinsularcortex,left

fron

toparietalareas,andleft

tempo

rallobe).SignalT

2andFL

AIR

hyperintensity

intheleftcaud

atenu

cleus

andin

theleftcorona

radiata(ischemicevent)

6 Neural Plasticity

Table1:Con

tinu

ed.

Pt.nu

mber

Case/

control

Sex

(M,F

)GA

(wk)

Age

(yr,m)

CPtype

Hagberg

Motor

abno

rmalities

GMFC

SMACS

CFC

SAssociated

impairments

TotalIQ

Verbal

IQPerform

ance

IQRadiologicalfi

ndings

(brain

MRI)

12Case

F31

10yr,

1m

Tetraplegia

Bilateral

spastic

hyperton

ia,

leftside

more

affected

43

3V:C

VI;H:n

o;M/A

:yes;LD:no;

E:n

o85

9282

Severe

periventricular

leuk

omalaciawithmajor

involvem

entof

the

posteriorarea,associated

withsupra-

and

subtentorialventricular

dilatation

andsubarachno

idspaces

enlargem

ent,

thinning

ofthecorpus

callosum

13Case

F41

8yr,

2m

Right

hemiplegia

Unilateral

spastic

hyperton

ia3

21

V:C

VI;H:n

o;M/A

:no;LD

:no;

E:yes

8799

77

Lefthemisph

ericatroph

y(previou

sextensiveleft

fron

toparietal

intraparenchym

alhemorrhage,wideleft

parietalsubd

ural

hematom

a),exvacuo

dilatation

oftheipsilateral

ventricles

andmidlin

ebrain

rightto

leftshift,W

allerian

degeneration

ofthe

corticospinaltractand

ipsilateralcerebellar

pedu

ncleatroph

y

14Case

M33

5yr,

10m

Right

hemiplegia

Unilateral

spastic

hyperton

ia2

32

V:strabismus;

H:n

o;M/A

:no;

LD:n

o;E:n

o85

9282

Leftfron

to-parieto-

tempo

ro-insular

polymicrogyria(perisylvian

andperiroland

icwith

corticalinfolding),m

ildleft

tempo

ralatrop

hywith

subarachno

idspaces

enlargem

ent

15Case

F6yr,

8m

Left

hemiplegia

Unilateral

spastic

hyperton

ia4

31

V: n

o;H:n

o;M/A

:yes;LD:no;

E:yes

100

106

107

Ischem

icright

fron

toparietalmalacicarea

withfocalcorticalatrop

hy,

gliosis,subarachno

idspace

enlargem

ent,andmid

ipsilateralventricular

dilatation

.Mild

controlateral

periventricularwhitematter

hyperintensity

7Neural Plasticity

Table1:Con

tinu

ed.

Pt.nu

mber

Case/

control

Sex

(M,F

)GA

(wk)

Age

(yr,m)

CPtype

Hagberg

Motor

abno

rmalities

GMFC

SMACS

CFC

SAssociated

impairments

TotalIQ

Verbal

IQPerform

ance

IQRadiologicalfi

ndings

(brain

MRI)

16Case

M38

6yr,

3m

Left

hemiplegia

Unilateral

spastic

hyperton

ia2

32

V:n

o;H:n

o;M/A

:no;LD

:no;

E:n

o99

101

97

Malacicareasaffecting

the

rightmiddlecerebralartery

territorywithWallerian

degeneration

ofthe

corticospinaltractandof

thethalam

us,left

hemisph

erehypo

trop

hy

18Case

M40

5yr,

3m

Left

hemiplegia

Unilateral

spastic

hyperton

ia2

31

V:strabismus;

H:n

o;M/A

:no;

LD:n

o;E:yes

101

106

100

Right

periventricular

porencephalic

lesion

(hem

orrhagicveno

usinfarct)withhemosiderin

depo

sition

andWallerian

degeneration

ofthe

ipsilateralcorticospinal

tract

18Con

trol

M36

6yr,

4m

Left

hemiplegia

Unilateral

spastic

hyperton

ia1

21

V:n

o;H:n

o;M/A

:no;LD

:no;

E:n

o90

9493

Supratentorialrightm

alacic

areaswithrightlateral

ventriculardilatation

;hemosiderin

depo

sition

second

aryto

germ

inal

matrixhemorrhage

M:m

ale;F:

female;GA:gestation

alage;CP:cerebralp

alsy;G

MFC

S:Gross

Motor

Function

Classification

System

;MACS:ManualAbilityClassification

System

;CFC

S:Com

mun

icationFu

nction

Classification

System

;V:vision;

CVI:cerebralvisualim

pairment;H:h

earing;M

/A:m

emoryandattention;

LD:learningdisabilities(N

orth

American

usage;mentalretardation

);E:epilepsy;MRI:magneticresonanceim

aging.

8 Neural Plasticity

action. This mechanism may underlie the capacity to under-stand and imitate others’ actions even at an early stage oflife [25, 26] and contribute to interact with other people inan empathic manner (for review see, Hari and Kujala)[27]. This same mechanism may be helpful during learn-ing motor tasks or relearning daily actions following brain

lesions and therefore during rehabilitation [28–30]. AOT hasthe potential to become a routine approach in the rehabilita-tion of children with CP and could be easily applied by phys-iotherapists working with children. During the rehabilitationsession, physiotherapists have the role to motivate littlepatients to observe carefully every detail of the observedactions and to push children to use the objects provided athand, as in the videos, but also to reassure if children fail inperforming the observed actions. Patients, even children asin the present study, may follow the rehabilitation programwithout difficulties. It is worth stressing that AOT may beapplied in a very flexible manner: in fact, the trained actions,presented through videos, may vary depending on the realneed of patients. For example, children that have more diffi-culties in performing distal hand/arm actions (i.e., grasping,manipulating) should focus their training on these motortasks, while children that present with impairment of proxi-mal arm actions (i.e., reaching objects, coding objects inspace) should train this kind ofmotor tasks. Last, but not least,AOT may be used also at home where children may get theirrehabilitation session with the help of their parents or evenin telerehabilitation with a physiotherapist monitoring froma dedicated position what children perform at home.

In the present study, we collected functional scores onMUUL and AHA also at two months of follow-up. Interest-ingly, treated children, as compared to controls, maintainedand even improved their functional gain at follow-up. Inour opinion, these findings may be explained by the fact thatduring AOT, children learn novel strategies to interact withother people and common objects. They learn to look verycarefully at all details present in the scene, they pay attentionat the different motor segments of an action, and they spon-taneously prepare themselves to imitate a seen action or tointeract upon objects available in the environment. Eventu-ally, they transfer these strategies in everyday life situations;thus at the very end, they accomplish the goal of gainingbetter motor performances.

A main point of interest in the present study is that someof the treated children also underwent an fMRI study aimedat assessing whether AOTmay recruit areas within the motorsystem and eventually contribute to their reorganization. It isworth noting that, while being scanned, children performedan independent task, namely, manipulation of complexobjects that were not included in the set of actions trainedduring the treatment. When comparing treated participantsand controls, differential activation was present in a sectorof the premotor cortex and parietal cortex also involved inobject manipulation in both healthy adults and children[31, 32] and known to be endowed with a motor representa-tion of distal upper limb movements. This premotor sector isstrictly connected with a parietal area with which it builds upa sensorimotor circuit allowing individuals to code for themotor properties of objects and the implementation of themost appropriate actions to act upon objects [33]. Thesefindings suggest that the brain target of AOT is exactly ahand motor area possibly involved in executing actions aswell as in their processing. It therefore appears that thereare not vicariating areas emerging from AOT treatment,but rather a recovery of areas normally involved in a specific

Table 2: Mean scores (and SD) of AHA and MUUL in controls andtreated participants at different time points.

Group ScoreTime point

T1 T2 T3

ControlAHA 65.71 (7.23) 66.86 (7.31) 66.71 (7.52)

MUUL 96.00 (16.73) 98.00 (16.69) 98.14 (16.52)

TreatmentAHA 57.45 (12.18) 61.09 (10.79) 63.18 (11.06)

MUUL 81.73 (22.38) 87.27 (22.36) 89.27 (22.41)

45

50

55

60

65

70

75

80

T1 T2 T3

AH

A sc

ore

Time

ControlsCases

⁎⁎

⁎⁎

(a)

T1 T2 T36065707580859095

100105110115120

MU

UL 

scor

e

Time

ControlsCases

⁎⁎

⁎⁎

(b)

Figure 2: Scores obtained by cases (red line) and controls (blueline) at T1, T2, and T3 in two different functional scales (AHA,MUUL). Statistical analysis (see text for details) showed that onlyin case scores obtained at T2 differed significantly from scores atT1 in both scales. This was true also when comparing T3 withT2 in both scales (error bars: 95% CI). ∗∗refers to statisticalsignificant effects.

9Neural Plasticity

hand motor task. Further studies should assess to whatextent this concerns also other biological effectors (e.g., thefoot) and contributes to rebuild physiological sensorimotorcircuits. Another issue that future studies should help toascertain is whether there are specific subgroups of childrenwith cerebral palsy that may mostly benefit from AOT, orrather this approach may help clinical conditions in allchildren affected.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

The authors are grateful to Anna Alessandrini, NicoleD’Adda, Maria Fezzardi, and Federica Tansini for their valu-able help in performing the treatment of the childrenincluded in this study and to Federica Pagani for her helpin collecting data. The authors are also grateful to SerenaMicheletti for her precious assistance in evaluating cognitiveprofile of all the children and to Annalisa Pelosi for her greatsupport on statistical analysis.

References

[1] S. L. Small, G. Buccino, and A. Solodkin, “Brain repair afterstroke-a novel neurological model,” Nature Reviews Neurol-ogy, vol. 9, no. 12, pp. 698–707, 2013.

[2] E. Taub, G. Uswatte, and T. Elbert, “New treatments in neuror-ehabilitation founded on basic research,” Nature ReviewsNeuroscience, vol. 3, no. 3, pp. 228–236, 2002.

[3] A. Roby-Brami, A. Feydy, M. Combeaud, E. V. Biryukova,B. Bussel, and M. F. Levin, “Motor compensation and recoveryfor reaching in stroke patients,” Acta Neurologica Scandina-vica, vol. 107, no. 5, pp. 369–381, 2003.

[4] M. F. Levin, J. A. Kleim, and S. L. Wolf, “What do motor“recovery” and “compensation” mean in patients followingstroke?,” Neurorehabilitation and Neural Repair, vol. 23,no. 4, pp. 313–319, 2009.

[5] P. S. Lum, S. Mulroy, R. L. Amdur, P. Requejo, B. I. Prilutsky,and A. W. Dromerick, “Gains in upper extremity functionafter stroke via recovery or compensation: potential differen-tial effects on amount of real-world limb use,” Topics in StrokeRehabilitation, vol. 16, no. 4, pp. 237–253, 2009.

[6] R. J. Nudo, “Mechanisms for recovery of motor functionfollowing cortical damage,” Current Opinion in Neurobiology,vol. 16, no. 6, pp. 638–644, 2006.

[7] L. Sakzewski, A. Gordon, and A. C. Eliasson, “The state of theevidence for intensive upper limb therapy approaches forchildren with unilateral cerebral palsy,” Journal of ChildNeurology, vol. 29, no. 8, pp. 1077–1090, 2014.

[8] V. S. Ramachandran and E. L. Altschuler, “The use of visualfeedback, in particular mirror visual feedback, in restoringbrain function,” Brain, vol. 132, no. 7, pp. 1693–1710, 2009.

[9] H. Thieme, J. Mehrholz, M. Pohl, J. Behrens, and C. Dohle,“Mirror therapy for improving motor function after stroke,”Stroke, vol. 44, no. 1, pp. e1–e2, 2013.

[10] R. Bruchez, M. Jequier Gygax, S. Roches et al., “Mirror therapyin children with hemiparesis: a randomized observer-blinded

trial,” Developmental Medicine & Child Neurology, vol. 58,no. 9, pp. 970–978, 2016.

[11] G. Rizzolatti and L. Craighero, “The mirror-neuron system,”Annual Review of Neuroscience, vol. 27, no. 1, pp. 169–192,2004.

[12] G. Buccino, “Action observation treatment: a novel tool inneurorehabilitation,” Philosophical Transactions of the RoyalSociety B: Biological Sciences, vol. 369, no. 1644, article20130185, 2014.

[13] G. Bellelli, G. Buccino, B. Bernardini, A. Padovani, andM. Trabucchi, “Action observation treatment improves recov-ery of postsurgical orthopedic patients: evidence for a topdowneffect?,” Archives of Physical Medicine and Rehabilitation,vol. 91, no. 10, pp. 1489–1494, 2010.

[14] D. Ertelt, S. Small, A. Solodkin et al., “Action observationhas a positive impact on rehabilitation of motor deficits afterstroke,” NeuroImage, vol. 36, Supplement 2, pp. T164–T173,2007.

[15] E. Pelosin, L. Avanzino, M. Bove, P. Stramesi, A. Nieuwboer,and G. Abbruzzese, “Action observation improves freezing ofgait in patients with Parkinson’s disease,” Neurorehabilitationand Neural Repair, vol. 24, no. 8, pp. 746–752, 2010.

[16] G. Buccino, R. Gatti, M. C. Giusti et al., “Action observationtreatment improves autonomy in daily activities in Parkinson’sdisease patients: results from a pilot study,” Movement Disor-ders, vol. 26, no. 10, pp. 1963-1964, 2011.

[17] G. Buccino, D. Arisi, P. Gough et al., “Improving upper limbmotor functions through action observation treatment: a pilotstudy in children with cerebral palsy,”Developmental Medicine& Child Neurology, vol. 54, no. 9, pp. 822–828, 2012.

[18] G. Sgandurra, A. Ferrari, G. Cossu, A. Guzzetta, L. Fogassi, andG. Cioni, “Randomized trial of observation and execution ofupper extremity actions versus action alone in children withunilateral cerebral palsy,” Neurorehabilitation and NeuralRepair, vol. 27, no. 9, pp. 808–815, 2013.

[19] J. Y. Kim, J. M. Kim, and E. Y. Ko, “The effect of the actionobservation physical training on the upper extremity functionin children with cerebral palsy,” Journal of Exercise Rehabilita-tion, vol. 10, no. 3, pp. 176–183, 2014.

[20] A. C. Eliasson, L. Krumlinde-Sundholm, B. Rösblad et al., “TheManual Ability Classification System (MACS) for childrenwith cerebral palsy: scale development and evidence of validityand reliability,” Developmental Medicine & Child Neurology,vol. 48, no. 7, pp. 549–554, 2006.

[21] M. Randall, J. B. Carlin, P. Chondros, and D. Reddihough,“Reliability of the Melbourne Assessment of Unilateral UpperLimb Function,” Developmental Medicine & Child Neurology,vol. 43, no. 11, pp. 761–767, 2001.

[22] L. Krumlinde-Sundholm, M. Holmefur, A. Kottorp, and A. C.Eliasson, “The Assisting Hand Assessment: current evidenceof validity, reliability, and responsiveness to change,” Develop-mental Medicine & Child Neurology, vol. 49, no. 4, pp. 259–264, 2007.

[23] S. S. Ghosh, S. Kakunoori, J. Augustinack et al., “Evaluating thevalidity of volume-based and surface-based brain image regis-tration for developmental cognitive neuroscience studies inchildren 4 to 11 years of age,” NeuroImage, vol. 53, no. 1,pp. 85–93, 2010.

[24] K. J. Friston, A. P. Holmes, C. J. Price, C. Buchel, and K. J.Worsley, “Multisubject fMRI studies and conjunction analy-ses,” NeuroImage, vol. 10, no. 4, pp. 385–396, 1999.

10 Neural Plasticity

[25] A. N. Meltzoff and M. K. Moore, “Imitation of facial andmanual gestures by human neonates,” Science, vol. 198,no. 4312, pp. 75–78, 1977.

[26] T. Falck-Ytter, G. Gredebäck, and C. von Hofsten, “Infantspredict other people’s action goals,” Nature Neuroscience,vol. 9, no. 7, pp. 878-879, 2006.

[27] R. Hari andM. V. Kujala, “Brain basis of human social interac-tion: from concepts to brain imaging,” Physiological Reviews,vol. 89, no. 2, pp. 453–479, 2009.

[28] P. Celnik, B. Webster, D. M. Glasser, and L. G. Cohen, “Effectsof action observation on physical training after stroke,” Stroke,vol. 39, no. 6, pp. 1814–1820, 2008.

[29] K. Stefan, L. G. Cohen, J. Duque et al., “Formation of a motormemory by action observation,” The Journal of Neuroscience,vol. 25, no. 41, pp. 9339–9346, 2005.

[30] R. Gatti, A. Tettamanti, P. M. Gough, E. Riboldi, L. Marinoni,and G. Buccino, “Action observation versus motor imagery inlearning a complex motor task: a short review of literature anda kinematics study,” Neuroscience Letters, vol. 540, pp. 37–42,2013.

[31] F. Binkofski, G. Buccino, S. Posse, R. J. Seitz, G. Rizzolatti, andH. J. Freund, “A fronto-parietal circuit for object manipulationin man: evidence from an fMRI-study,” The European Journalof Neuroscience, vol. 11, no. 9, pp. 3276–3286, 1999.

[32] L. Biagi, G. Cioni, L. Fogassi, A. Guzzetta, G. Sgandurra, andM. Tosetti, “Action observation network in childhood: a com-parative fMRI study with adults,” Developmental Science,vol. 19, no. 6, pp. 1075–1086, 2016.

[33] M. Jeannerod, M. A. Arbib, G. Rizzolatti, and H. Sakata,“Grasping objects: the cortical mechanisms of visuomotortransformation,” Trends in Neurosciences, vol. 18, no. 7,pp. 314–320, 1995.

11Neural Plasticity

Hindawiwww.hindawi.com Volume 2018

Research and TreatmentAutismDepression Research

and TreatmentHindawiwww.hindawi.com Volume 2018

Neurology Research International

Hindawiwww.hindawi.com Volume 2018

Alzheimer’s DiseaseHindawiwww.hindawi.com Volume 2018

International Journal of

Hindawiwww.hindawi.com Volume 2018

BioMed Research International

Hindawiwww.hindawi.com Volume 2018

Research and TreatmentSchizophrenia

Hindawi Publishing Corporation http://www.hindawi.com Volume 2013Hindawiwww.hindawi.com

The Scientific World Journal

Volume 2018Hindawiwww.hindawi.com Volume 2018

Neural PlasticityScienti�caHindawiwww.hindawi.com Volume 2018

Hindawiwww.hindawi.com Volume 2018

Parkinson’s Disease

Sleep DisordersHindawiwww.hindawi.com Volume 2018

Hindawiwww.hindawi.com Volume 2018

Neuroscience Journal

MedicineAdvances in

Hindawiwww.hindawi.com Volume 2018

Hindawiwww.hindawi.com Volume 2018

Psychiatry Journal

Hindawiwww.hindawi.com Volume 2018

Computational and Mathematical Methods in Medicine

Multiple Sclerosis InternationalHindawiwww.hindawi.com Volume 2018

StrokeResearch and TreatmentHindawiwww.hindawi.com Volume 2018

Hindawiwww.hindawi.com Volume 2018

Behavioural Neurology

Hindawiwww.hindawi.com Volume 2018

Case Reports in Neurological Medicine

Submit your manuscripts atwww.hindawi.com