The impact of physical therapy on functional outcomes after stroke: what's the evidence?

(1)

CtllnicalRehabilitation 2004; 18: 833-862

The impact of physical therapy on functional outcomes after stroke: what's the evidence?

RPS Van Peppen DepartmentofPhysical Therapy, VU University Medical Center, Amsterdam, G Kwakkel Departmentof Physical Therapy, VU University Medical Center, Amsterdam and Center of Excellence for Rehabilitation Medicine 'de

Hoogstraat'. Utrecht, TheNetherlands, S Wood-DauphineeSchool ofPhysical andOccupational Therapy, Department of Epidemiology andBiostatistics, McGill University, Montreal, Canada, HJM Hendriks Dutch Institute of Allied HealthCare(Npi), Amersfoort and Maastricht University, Department of Epidemiology, Maastricht, PhJ Van der Wees Royal DutchSocietyfor Physical Therapy(KNGF), Amersfoort and J DekkerInstituteforResearch in Extramural Medicine (EMGOInstitute), Departmentof Rehabilitation Medicine, VU University MedicalCenter, Amsterdam, The Netherlands

Received 23rd March 2004; returned for revisions 10th June 2004; revised manuscript accepted 25th July 2004.

Objective: To determine the evidence for physical therapy interventions aimed at improving functional outcome after stroke.

Methods MEDLINE, CINAHL, Cochrane Central Register of ControlledTrials, Cochrane Database ofSystematic Reviews, DARE, PEDro, EMBASE and DocOnline were searched for controlled studies. Physical therapy was divided into 10

intervention categories, which were analysed separately. If statistical pooling (weighted summary effect sizes) was not possible due to lack of comparability between interventions, patient characteristics and measures of outcome, a best- research synthesis was performed. This best-research synthesis was based on methodological quality (PEDro score).

Results: In total, 151 studies were included in this systematic review; 123 were randomized controlled trials (HCTs) and 28controlled clinical trials (CCTs).

Methodological quality of all RCTs had a median of 5 points on the 10-point PEDro scale (range 2-8points). Based on high-quality RCTs strong evidence was found in favourof task-oriented exercise training to restore balance and gait, and for strengthening the lower paretic limb. Summary effect sizes (SES) forfunctional outcomes

ranged

from 0.13 (95% Cl 0.03-0.23)for effects of

high

intensity of exercise training to 0.92 (95% Cl 0.54-1.29) for improving symmetry when moving from sitting to standing. Strong evidence was also found for therapies that were focused onfunctional training of the upper limb such as constraint-induced

movementtherapy(SES 0.46; 95% Cl 0.07-0.91), treadmill training with or without body weight support, respectively 0.70 (95% Cl 0.29-1.10) and 1.09 (95% Cl 0.56-- 1.61), aerobics (SES 0.39; 95% Cl 0.05-0.74), external auditory rhythms during gait (SES 0.91; 95% Cl 0.40-1.42) and neuromuscular stimulation for glenohumeral subluxation (SES 1.41; 95% Cl 0.76-2.06). No or insufficient evidence in terms of functional outcome was found for: traditional neurological treatment approaches;

exercises for the upper limb; biofeedback; functional and neuromuscular electrical stimulation aimed atimproving dexterityorgaitperformance; orthotics and assistive Address for correspondence: Gert Kwakkel, Department of

Physical Therapy,VU UniversityMedicalCenter,POBox7057, 1007 MBAmsterdam, The Netheirlands.

e-mail:g.kwakkel&vumc.nl

(('' Arnold 2004 10.I191/0269215504cr843oa

(2)

834 RPS Van

Peppen

et al.

devices;andphysical therapyinterventionsfor reducing hemiplegicshoulderpain and hand oedema.

Conclusions: This review showed small to largeeffect sizes for task-oriented exercisetraining, inparticular when applied intensivelyandearlyafterstroke onset. In almostall high-quality RCTs, effectsweremainly restricted totasks directly trained in the exercise programme.

Introduction

Systematic research has shown that organized multidisciplinary care and rehabilitation after stroke enhance patient survival and independence, as well ^as reducing the length of inpatient stay. - It remains unclear, however, why specia- lized stroke units are more effective than usual

care. Anumber ofcomponents have been identified

as contributing to the efficacious care delivered in such units. These include the comprehensive as-

sessment of medical problems, impairments and disabilities;activephysiologicalmanagement;early mobilization and avoidance of bedrest; skilled nursing care; early setting of rehabilitation plans involving carers; and early assessment and planning for discharge needs. A Several of these factors ^are closely related to physical therapy which is often perceived as one of the key disciplines in organized stroke care.5 In addition,

a recent Cochrane review of 14 trials (N= 1617) showed that outpatient services, including physical therapy, mayprevent deterioration in seven of 100 stroke patients residing in the community.6 The main foci of physical therapy after stroke are to restore motor control in gait and gait-related activities and to improve upper limb function, ^as well as to learn to cope with existing deficits in activities of daily living (ADL) and to enhance participation in general. Besides using physical exercises, physical therapists often apply assistive devices forgait, and employ other equipment such

as treadmills and electronic devices to support their treatments. In addition, advice and instruc- tions areprovided to the patient, family and other members of the stroke team regarding prevention ofcomplications such as falls and shoulder pain.

Today, theimportance of evidence-based medicine

as a guide for the clinical decision-making process

is increasingly being recognized by physical therapists.7,8 However, the efficacy of physical therapy interventions for stroke has not been summarized

in a systematic review. The objective of the present

systematic

review was to

establish

the

evidence

of

physical therapy

interventions related to

improving

functional outcomes after stroke.

Material and methods

Literature search

A

computerized

literature

search

was

conducted

in

MEDLINE, CINAHL, Cochrane Central

Register

of

Controlled

Trials,

Cochrane Database

of Systematic Reviews, DARE, PEDro,

EMBASE and

DocOnline (Database

of the

Dutch

Institute

of Allied Health Care).

Two

researchers (RPSvP

and JCFK)

independently searched these

electro-

nic databases

for

relevant

articles. The

search

strategy was

built

on

cerebrovascular disease

(patient

type) and physical therapy interventions

(treatment type).

Randomized controlled trials

(RCTs) as well as

controlled

clinical

trials

(CCTs) were

included

for

review. Excluded

were

noncontrolled pre-experimental studies

and con-

trolled studies that investigated robotics

or

the effects of physical therapy in combination with

acupuncture or

drug therapies. Studies

were

collected

up to January 2004. The

following

MeSH and

keywords

were

used for the electronic databases: cerebrovascular disorders, cerebrovas- cular accident, stroke, hemiplegia, physical therapy, occupational therapy, exercise therapy, and rehabilitation. Bibliographies of

review

articles, narrative

reviews

and

abstracts

published

in

conference proceedings

were

also evaluated

for relevant

publications. In addition, citation tracking

of all article references was conducted.

Only articles written

in

English,

German or Dutch were included for review. Inclusion of

articles

was

based

on agreement between

the

two

independent reviewers.

The

full

search strategy is

available

on request from the

corresponding

author.

(3)

Impact

of

physicld

therap^y after stroke 835

Subsequently,

the two

reviewers independently determined from

the

title

and

the abstract

if

the

papers

satisfied the following criteria: population

of adults (18 years or

older) diagnosed

with stroke

and studies evaluating effectiveness of physical therapy interventions.

Intervention

review, physical therapy

was

classified into

10

intervention categories

to

evalu-

atethe

effectiveness

of: (1)

traditional neurological

treatment

approaches; (2)

programmes

for training sensorimotor function

or

influencing

muscle tone;

(3)

cardiovascular fitness

and aerobic

programmes;

(4) methods for training mobility and mobility- related activities; (5) exercises for the

upper

limb;

(6) biofeedback therapy for the

upper

and lower limb;

(7)

functional

and

neuromuscular electrical stimulation for both limbs;

(8)

orthotics and assistive devices

for

both limbs; (9)

treatments

for hemiplegic shoulder pain and hand oedema; and (10) intensity of exercise therapy.

This classification

was

based

on

the

Interna-

tional Classification of Functioning, Disability and Health

(ICF) of the World

Health Organiza- tion9 and the Amnerican Physicail Therapy Associa-

tion

guide

to

phsical therapist

practice

(2nd

edition).'0 A

group

of eight physical therapists and

two

reviewers (GK and RPSvP) reached

consensus

about

the

categories.

Methodological quality

The methodological quality of the RCTs

was

rated with the PEDro scale.

'l RCTs were

scored by

two

independent reviewers (RPSvP and GK).

Inter-rater reliabilities of individual items of the

PEDro

scale

were

calculated by Cohen's

kappa. In case of

disagreement,

consensus was

sought,

but

when disagreement persisted,

a

third independent reviewer (SWD) made the final decision.

PEDro scores

of

4

points

or

higher

were

classified

as

'high quality', whereas studies with

3

points

or

lower

were

'low quality'.

PEDro scores were not used as

inclusion/exclusion criteria, but rather

as a

basis for best-evidence synthesis and

to

discuss the strengths and weaknesses of studies.

Quantitative

analysis

Analysis of the results

was

performed separately for

each

intervention and restricted

to RCTs.

When

they

were comparable in terms of

interventions, patient characteristics

and outcome measures, statistical

pooling

was

performed. Randomized

studies

using

a cross-over

design

were

judged

as an RCT

by

calculating effects before the point of cross-over. The data were

reanalysed by pooling

the individual effect sizes

using fixed

effect sizes. ' Fixed effect sizes,

gu

(Hedges'

g),

were

calculated

for each study by finding the

difference between

mean

changes

in the experimental group and in the control group and dividing by the average population standard

deviation (SDi).

To

estimate

SDi for

gu,

baseline estimates and standard

deviations

of the

control

and experimental groups werepooled. The impact of

sample

size was addressed

by

estimating a

weighting

factor

(wt,i)

for each

study,

and

assigning

larger

effect-weights

in studies with

bigger samples. Subsequently, g'

values ofindividual studies were

averaged,

result- ing in a

weighted SES, whereas the weights

of each

study

were combined to estimate the variance of the SES.'4 If significant

between-study

variation existed

(statistical heterogeneity)

a

random

effects model was

applied.'

Based onthe

classification

of

Cohen,

effect

sizes below

0.2 were

classified

as

small,

from 0.2 to 0.5 as

medium

and above 0.5 as large.16

Best-evidence synthesis

If pooling of studies was not

possible due

to

differences

inoutcomes, intervention types, patient

characteristics

or lack of point

estimates

(means and

medians) and/or

measures of

variability (e.g., standard

deviations

and confidence intervals)

a

best research synthesis

was

applied.

For this purpose we

used the criteria

set out

by

Van

Tulder

eta.

17

based on the

methodological quality

score of the PEDro

scale. Subsequently, studies

were

categorized into five levels of evidence: (1)

strong

evidence, (2) moderate evidence, (3) limited

evi-

dence, (4) indicative findings, (5)

no or

insufficient evidence (Appendix 1).37

Results

Literature search using multiple databases yielded

8024

citations

on

29 January, 2004. After restrict-

ing these

tothe

publication

type

'clinical trial',

735

(4)

836 RPS Van

Peppen

et al.

remained. Following the exclusion of: (1) pre-

experimental studies and (2) controlled studies investigating the effects ofphysical therapy interventions that included people with nonstroke as a

diagnosis orinterventions withacupuncture, drugs

or robotics, 204 relevant studies ^were selected by title and abstract. Twenty-two of these articles were

systematic reviews,

38-38 and 20 were

critical or narrative reviews.39

58

Eleven of the remaining 162 studies had been published in ^more than one

article.59-8)

A total of 151 publications (123 RCTs and 28 CCTs) that focused on the effectiveness of physical therapy interventions in people with stroke were included for further analysis. Cohen's K, as an estimate of agreement between the two raters for methodological quality of the 123 RCTs,^was 0.81.

Foreach interventioncategory theresults of the studies that contributed to the meta-analysis or

best-evidence synthesis are presented in Tables 1

and 2. The methodological quality of the RCTs is reported in Table 3.

Evidencerelated tothe effectsofthe traditional neurological treatment approaches

Eight RCTs67'8' ⁸⁷ and two CCTs8 9 investigated the effects of using ^a specific neurological treatment approach. Numbers of patients, characteristics of the interventions, ^measures of out-

come and observed effects ^are shown in Table 1.

Different

neurological

_i81 treatment_entB approaches in-

un st 85,87,89

cludin Bobath,' ⁸ Brunnstrom,

Rood, 2,83 Johnstone,84

Proprioceptive

Neuromus- cular Facilitation (PNF),8 88 Motor Relearning Programme (MRP),67 Ayres82 or combinations of the above89 were investigated. With ^exception of two RCTs8485 all studies evaluated the effects of Bobathin oneof the treatment arms, whereas^one study used two experimental groups.86 Eight studies measured ADL with the Barthel Index (BI) 67,82,83,86,88 the Functional

Independence

Measure (FIM)87 or other ADL scales8-89 as an

outcome, and four studies evaluated strength,83'85

synergism84

or muscle tone.88 Three studies as-

sessed the effects of a neurological approach on

length of stay

(LOS)67,81589

and compared the effects of MRP and

Bobath,

67 PNF and Brunn-

strom85

or neuromuscular retraining techniques,89 whereas one CCT compared an impairment or-

ientated with adisability focused

approach.9")

Thequality scoreof the RCTs ranged from

3823.83

to

6.67

Due to

differences in both aims and

outcomes,

pooling of

the studieswas not

possible.

Best-evidence

synthesis

showed moderate evidence

for a

reduced

LOS in favour

of

MRP or

tradi-

tional care

compared

with an

impairment-focused neuromuscular

treatment

approach such

as

Bobath.67'89'90

Noevidence was found for

applying

a

specific neurological

treatment programmes in terms

of

muscle

strength' 8385 synergism,

84

muscle tone,88 walking ability,88 dexterity67^81,87

or

ADL.67'82,838

^-89

Programmes for training sensorimotor function or influencing muscle tone

'Sensory motor

training'

was

defined

as exer-

cises for

improving motor

performance, strength,

power

and endurance,I0(S72)

as

well

as sensory

integrity

(proprioception,

pallaesthesia,

stereogno- sis

and topognosis).'0(590)

Strengthening

paretic

muscles

Six RCTs9'

96

and

two

CCTs9798 investigated eccentric98 and concentric strengthening exercises for

the

lower9' 94,96

98

and

upper

limbs.92

97 Treatment sessions ranged from

3096

to

9095min

per

day, and

were

applied 29'

to 5 imes aweek for 2 to 6 weeks

(Table 1).

Methodo-

logical quality ranged from 49193

to 7.94 A

meta-analysis

was

possible

for

three RCTs9'9294 that assessed self-selected comfortable walking speed.

A

homogeneous nonsignificant SES

was

found

in

favour of strengthening muscles of the paretic

lower

limb

on gait

speed (Table 2). By weighting the quality of

the

studies,

a

best-evidence synthesis showed

strong

evidence for

the

value of increasing the muscle strength of the lower limb

in terms

of maximal voluntary efforts,92

mass motion

(on

an

Elgin-table)93

or

maximal isokinetic strength

(on a

Kin-Com).94 Limited evidence

was

found for increasing walking endurance91'99 and gait performance (average

torque

of muscle

groups

of paretic and nonparetic leg).9498

No

evidence

was

found for strengthening

exercises to improve

hand

grip

force,92'97

to support

strengthening muscles for climbing stairs,94 transferring,92 estab- lishing

symmetry

of weight distribution between hemiplegic and nonhemiplegic sides,98

dexter-

ity92'97

or

for physical and mental health.94

(5)

Impact of/phy sical therapy after stroke

837

00cro

u _[r _v

00 CO-

.' C0 Ca)

Cr) .a0 Q-

n-

C0

C z a)X o0 °

a) Q-oo

x)

E

(a UCD- C a)Ca ,-0

ao a,,,

a) a-)

'c (UC)

LC UD CD X 0D oCCU

C)

TC

a)

>- a)2, C) CD X C)

C)LO C-

CO

CDI :

'Or LU:U-

C,)U

H)

ccoo

0 E co

U)

a) \(

v a)

a) a)

C , ) :>, X 30 XQ

CC4

^C

C C L C a)

co cll

V

a:Ca) an

coou

09

co

C-

0

a) a)(

co

0 UD

c C) n

-0(E ma0

v a)

C a0C c

(a0

Ma) a

_(aa)

LO

Ca)

00

aD

a)U)

Ca

C '

U)if

a)

co 0o E

0._

C5(o 0g

o C C "---

o a) 0

> -0 -o 0

2 'D aoE o 0

00

,_ oc)u)c 0 or )(?,

U) 0 o<~0

~g aa0a) a -

z- a)~a

c a)°

-o a)

o LD C:m

O (D

a)CC

CU0~~~C-

o)m

c>

0 )

A

E Cu)

A .)U Eco).c

a) L)

CD

00

coo

-o C-a

nnc m> ^o

I, m

co o00m

r- GO QD co 9cD m:u 00cl n

a)

0 0t

a)a)

a)

- a )a))) U 0 O0)j

0LLJ tn 5 O LU c

C a).>

70 _0 a

co

m ,\

0 m

cr

CD0

a)

(D

R0

a, VU

QL

coa)

a)

0 a

0 0

C >

a) C (D

a)00

+-

O!

^C-

Q-}a) oa)2

COoC-

O Q)

a)a>

QfaD

CLC

a a CD-

a) o> aM

CV

(6)

838 RPS Vain

Peppen

et al.

0) -~~~~~~~~~~~

C- cc coccc

0CY

00-cc C: C C

0 0 0 0 0

a7onoc6 m) (0aLI)

0 COC~~~~~~~~-0)C00

-o 0

0~00

^cc ⁰⁰ ⁰⁾ ^-

0 cc~~>0V)C,~0)o) 0 0-

~~~C~~~00a~~~~~~~ 00050 +-c- 0

0)> 2- 0cr 01) 270 000 C> ccC> 0

-o ()0 ocC c o0c0 C~) -0 70~o 0 0-OC C

> _C- :3 >0 0 (L) ~ ~ ~ C )0)CC 0)1cc 0)>-

n 0- x0 00)0)cooi C 0 -c c CHOC (L

CD EDa) > 0 0)0Cn C 0) 0 ) > aC

C cc C- 02 0 0C0C 0 0

0) ccc~~~~~~COQ

co ccc -o'C) c Q

Ocz~~~~~~~~~~~0iciC)co0

co >-~ Co 0 0

C)LD E 0 >-0 >c 0-) -> CY

(0 C~~~~~~~~~~~~~~~cca £ (~~~~~~~~~~~~~~~0 -00:)

0)0 0) E (D 0 10 0 E 0)a) 0

0)~~ ~ ~ ~ ~~--

OC 0)~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~D0

Q) 0~~~~~~~~

0)E00-6m~

200)0 0) (0 0) CY 205co)

(0a000)~~~~~~~~~~~~~~~~- 0)~~~~~~~0)0)E0Q) --

(N (0 (0 o )- Q(0

0) 0)~~~~~~~> Q (0 m (0

0) 0)0) 0) 0)~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~)Q

0) 0) 0)

0 00 C

C: E CD x ccc C 0

q)CDc 0) 0)C xL ^X0

C 0)_00D 00)000

:(1 O OCCoCO

cOcca L0 LC-0- 0Y00ocl~cr)-

U)~~ ~ ~ ~ ~ ~~~0

E~~~~~~~~

O 0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 00

C cc-C

cc)

0 -0 (D.> c

.0 ^a-E 0

C0~~~~~~~~~~C

~~~~~~~~~~~~~~~~~~~a~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~c- Nc

(7)

Impact of physical

therapy

aJterstroke 839

r,L-H

00 0

0:0 0:r

(0

0 Lf)

(NQ0

-0 0

0-

-F 0

~~~

^E ^F

0 U)

-0

Fo

⁰

a)o

^-F

CUt0 )~ U £~ o Q) -

0 C)7

F0 U)r.U

)0

^U)⁽⁰

0 CU 0 < C 0 0 C 3 -

0 0C- CD 5;> 0 C

U) (D --F U)D

-Ez >FQ(0Z0 0 E< U)U 03), 0 ZU

(D v

U) UV

(0 2UU a)U)~

0N 0D

^C ^:

C >: >- F

c)X

°C)D 0)

E _( - Q (0 -

(E Q) Q(0N3LD(N£3

.D (D

00

U) m0 03\, a

Fc ^U)(

U)- CD X C

coQ0-

st

00

(0^E CD

L(

,,, U)

(N (N

00:

u 0

F

m

XT ._C

.- ^C m

U)0

co~

U)- Q)

750 0H

O

0Q-

(0 00U)

o

c~- U)U

E

Z- 0 oU)

00

.E 0O

LE Q-U) co U) -D

LO coU)(

5 o

co

CY)

U0

co

,M

H U) {OH0E: 0

(N)

0 Q0 (N

0

rr CD a) ,U)).,

0>

(00 0C C

_C

L

C-a

C)

;0°-c

(0

C-iD

U )~U)

-O U) U)

CD0)vx co CD

LO;

CN(N

(o

r-

CN (N

(0(CN (0CD

(NOC) (D(

00 00

U)) Q)U

0 0

Ql

U)( U)(0

U-0(0 a)

Q)

U) 0 (0 Ho

CO [rr

H 0

0:(N- ^0:

(0 0

*0

(0 Ld

r,

U

[E U U(

1- U)U.

cro c0n

o: 0

(0)

(N NO

n n

U)

0-(C

co(0

(N

Lr.,

0

(0 Cc rO

Q

U)

\1

U)

0) CD

C -

CD c

0-) U-

C)09 z U)

U)

a)

03

00

?

_-

-or-.

0:0:

_u_CD

H -

CO

CC E UC)⁰⁾

C)

COm

-axE U(

(.0 -oA CD

CY)

£3

(N

(0 -00:

E ⁵ (0

U)

> a)U)

m a)cE, Fo5:U)(cl

E crX )

C O m

CO (N(N~0

0

E CN

c0

(o

co

0

0 0:6 -bU)-

c E

a) U- LLJxLU -

0^\, a^I/

(DQ

C0 U)

C>-C- C) U)

0 (0 A

(0 0:0

m

(0 0 EJ

0:

r-

00 (0

(N H O 00 0

0 0:

u u a:

(0

C-

0 C

:> QO

0)

U)E U) --U) XU LU (0

- a) m -0 C- >, 000£3 0

F-F)

U 0)

0c-

_C

-U)^U) E

a 3<(

LU

UQ

X(Q

(8)

840 RPS

Vcan Peppen

et at.

co

0)

0

ua ( 0) Cc

m0)0

- Q) a) o

2l

O u r 5c

_ CL _ _c

O >,F- 1)O

(- .-U 3: c:

a): a.().>

cCO D 0) N

CE

.0 CO

- 00 - 0 CC

X -0 0) -0 0 Q°s0L0CC O

z > °° C 4C(1) (°> 0 C0 > C

o)°3 ( C°D"m °a _aac-a,ao

o o>> 0 9H__E0)-o C0+-

0). 0Q )>._°9 5 ._O aO>'~-00)) > S 0 D

U) E +O-~~~~~~

Q) Q-0)0)0>,0 0o>

z-Zoc E

0)

v

>- a)0)

C)(D

CDXC09 (O

CDLOz

0- (0 CO co

C-

o 0)-o

-0 T

(I)

0a)

L-')

0)

0- a) C O:co E 00

C) X 0)

0) C

COCf) ^- D0

0)

v

0)

C 20 C)CN CD X Oa

O Z

0C<)9 D

0

E

co 09 co

LO

co9

09

4;

HD 0)cC

c C- 0) ) 0)

c-c o v D+':

o

zna2 Lw

09CO 0 0- c u

E

a) D

~ O C0)

0 009 0) E

z ^C

09 ~~~~~--

0 0

> C- 0 C

~ -

E 0000C 00

E) 3 0 a0- 0

C Ca

o-0)0 >C 0. >

Q)9₀₎ Co_>) ₂₀₎_C- 0)0)X_ca

CN~~~~

C0 0)CO~C)-0 :003

0> -0 707

03ac- D0

o 9: 0) 0-U

-0

C0(D90V -0 0)

C Ca) E -aC

c OT

) (D 4

0 CDC CD _FDco_

_C V C-

V1 D 0 D 00)

', a) -

C Q1 C0

Fs

0)09

Un 0a_

09-

LOH0)0 09042Q

O 0 QL

a

co

H-H

0 0~

LdCO 4l^c

0)

o0-

a

'U))

U2)oa)

a) 0- -0

0) 0(D

>0)t.

C0C U)a 0>

00) (0)

a 0)

00) -0)-n)0

o0)0

c0) Q)

Q 0C .010

(9)

Impact

of physical therapy alfter

stroke 841

H-H 0

030 CN O

uC) mC)

- cl

0a_

0 0

I (D o o(N (N

U< U

m, C)

0 LO 0)

03H

0 N

- nr

0) 0 C 0

O (1) c-a ° aa q

C E 0 C

E, o ) - E oD

CaCaC)-

2T):3030 E 03C:

(D2)zU C )U)c °

Z) 0U)b 0)ZQ Z . D

0o

^C: ^0)CUo ^cc_

ao E 0>. -0 0)> E E o

CCa-C)0 00 0) C

3- 0 )C- 0 OC 0 C

U)C ) C7)C

0)LOI/C '_C ,fD

C-0CC_

(~N

co(

0 03

Lo

LOD

cil

aS) 0003

0

>_ E

0)

0o ^x 0) 0r) X 0) Li C L20)

0E

co LOD

03 (0 fn r-

0) (N

HH00 CC30

> U) m a -7 C) U) )

00)t (NJ CY)C:

x(T

,(N

Cr)

H

Cr_

H-

FL

E

> Co Q)-a)

0) CD 0C- 0)

C-

0 ) 0C0)C 0E

C Q >) U)X) 0 1)CO) H C

0F) 73 2oC 0 CD

C 03C

C)C

0) o)73 C

2 E

C-

1 H -a 0 a)

H-U U

OD 0

l- N C)

LO^m

a) 0) fOO

(-3

C)

0-Q_

0

., -o C-

O C)

CD6

0,)

C-O

*_

O)

C)>

Z,E ^m

00 4-C)

>

0) -3 W 0 X a C C

-C O

OE

g0,_o

00u 0) 0)w0)J

= Lm0)

20' U) -00-

nn' _F1-D 1

0^(a)oZE

4-4

0 EL) 0 Q 0/) C

t a o

-o.o Ea

o 3 0

._

mC1Co E^,E^H ⁰ E

0 >

90 0)0

0 0)

0~ ⁰

E E

0

> ) C co C > 0)

C-) o-

6L

o

oC U

0 U) Co

0 0

E °0O

O .0 Co

c) 0

*_C-LN

CoUE

o 0-

._ .

0 co

-o

c) 0

E

co

CD(D

0--

03

E

03

(0

0)i

(o

co 0) c 00V co0

9_ fV

a

- ()

0 C) U)U)U C20

Q)Q

c X a _0f)0

03

Q C-

< O u)

C QD - a

C >Uc)D ,C 0 V C

-0 >

D+-

0- C a °)

<o-0a)

(10)

842 RPS Vcin Peppen et at.

a)

-D0

CD uo

QDc) o

C.O

U-)

a a

-'.2 E Co Co C) Co CD CoCU

CoU) X CZ.

*

CoCo

Co

Co14 Co.

CoU)

X1

(D t-CCC Cs)AoCOCoCo

CUCD U CoCCO )CUCooQ Co

a >

rnCD

o U

CO

C)o c co

>L

Co C

J U)QOU) : U) U)c CU CoCU)Co

CUCo Co-

Lo Lo CoCoCoCo

CD

CUCo CUCCC

Co

COCC)CU U C

'OC m

CUoC CUCfGoC)Co

I= >-

C,-

C a

(D Q~ Uo

C)U) C

C0 --Coa

0 v D CU

c- } a)

:-B

^C ° U) <V JD ) ( CUCoC)C²^{C O}

V

^~3^Cf U)CU~

H C

-o

>- c 0

Co0

0 c}

CUC CU

*_ -C-~

._^l(f ⁼

U)

a)

>

0 0

Co Co CU

C-a) Q) 3:

-o

⁹ o

0C

3 (Da) QD C

c0 E 0E) C -

o coC-F-) CoCF-oCo CU

: aCCD V)CUCU

a)

C Co aC ^c(

E V C)C CO

E <~g0~Cg rUztl E

>°:]I0EC:

Co )C C C)Co..

- 0 C

Hr C Co 1U) Q0

-Co

a) a) --

U- aCoS

0 U)- co CO5CC-C)

COCo0C EU)Co C>

0oCo C->

Co CoCUC)

2~C

-C

-V

ICoCo

co Co

CU 00 C Co

C-- LO3

CUD C CU Lo C-Co U)

cs^0' Lf)

LO * 0

Co C CoCo

co oCo .-Co_ O Co Co CD CL Co o o o o CU Co C) CS U)

U) U)Q)UQ)

X Co X X X

-0 0

co

CDI4C

Co

Co~

C CO CoC

COCo C Co -6 E0)

O

Cox

Q)

-)Co coC LO LmCs}c

Co Co Co CD 0)Co- CO M "

Co Co Co Co

0

Co Co Co Co

U).)_)_

o Co C) Co tl; Q V :

co C C C)Co

CD CO

CO C° CO

Co, CDCO Co rv4 co

CoCo

Co 1- Co)

CO CO CO CD (C Co C) Co

U Co C) Co CO Co

CD ' 0- co

Co)co co Co Co ,t COCo C'

>-

Co Co Co

COCot

Co Co Co

C)

Q) Q) (

-aC-) Cot

0 0

4l-U)

EU)

CU U)o

U) U)

cU) CcQ)

Co

4:1

a) 0

U)C)

C)CoU)a) Q)

t.CO C) ^Co o

CoOCfna

0 c-

oU)

a U)

'OC

C 0~

. A5

_ a 0CU)

>C3) CCC

C-)

0

CU

Q)

CU

N C) I-

U)

(D eC

E.-

0)

U)

O

Th) E

o ~ ~ C

U) D,

a1) (f)

C)0U() U1)

.c020 C) CCU)C-_C) ₀ 030-

2O :3CU Z_ o3 C

'C E

Cou w- 2-

⁰

U) o 0 U)

E

tQ

UCo E

C ) g E

Q0^o '--0

> 0 O 0) CoO- C -U C)_c t_cm

0 m

C-1

ED-