d i s a b i l i t y a n d r e h a b i l i t a t i o n , 2000 ; v o l . 22, n o . 6, 275± 280

An assessment of gait analysis in therehabilitation of children with walkingdi� culties

DAVID HAILEY ‹ * and JO-ANNE TOMIE ‹ Œ

‹ Alberta Heritage Foundation for Medical Research, 10180± 101 Street, Edmonton, Alberta, T5J 3S4,

Canada

ΠCurrently at : Community Health Sciences, University of Calgary, Alberta, Canada

Abstract

Purpose : To assess the current status of computerized gaitanalysis techniques in the management of children with cerebralpalsy or spina bi® da who have signi® cant walking disorders.Method : Synthesis of available data from a review of theliterature, drawing on MEDLINE, EMBASE, PRE-MED-LINE, HealthStar and PsychInfo . Other information wasobtained from persons with expertise in computerized gaitanalysis. Cost data were obtained from Canadian rehabilitationcentres and the provincial health ministry.Results : This technology seems helpful in detecting gaitchanges. However, available evidence is insu� cient to drawconclusions about the in¯ uence of computerized gait analysison treatment outcomes. Part of the rationale for use of thetechnology is that costs of gait analysis (of the order of $ CAN2,000 per examination) would be oŒset by a decrease in follow-up surgical procedures and associated hospital care. Therecould also be a major in¯ uence on children’s independence andquality of life. However, there are as yet no convincing data tosupport these propositions.Conclusions : Computerized gait analysis is a potentially usefultechnology in the management of children with walkingdisabilities, but its e� cacy is not established. It should beregarded as a developing technology and its clinical applicationlinked to systematic collection and assessment of outcomesdata.

Introduction

Abnormal gait is a functional problem for many

children with cerebral palsy and spina bi® da. Treatments

for gait abnormalities in these children depend upon the

nature and severity of the de® cit and include lower

extremity surgical procedures, injection of botulinum

toxin and use of orthotic devices." ± $ Surgery may have

adverse eŒects, such as stiŒer walking and instability of

* Author for correspondence; e-mail : dhailey! ahfmr.ab.ca

the hip and lumbar spine that need further treatment." , % ± ’

Because of the di� culty in planning and predicting

outcomes of lower extremity surgery for such patients,

treatments often follow the `birthday operation synd-

rome ’ .’ , ( A series of procedures may be performed, with

intervals between surgery of about a year.

Gait analysis techniques have been proposed for use in

the management of some children with walking disa-

bilities. They are intended to provide health care pro-

fessionals with detailed data on the components of

walking so that better decisions can be taken when

planning surgery or other remedial treatment.

Modern gait analysis techniques, such as videotaping,

electromyography (EMG) and computerized kinematic

and kinetic analysis allow detection and analysis of fast

moving events and can have good reliability.) ± " ! Infor-

mation from gait analysis has the potential to enable

`single-event, multi-level ’ surgery to replace the `birth-

day syndrome ’ approaches that have often been used

when treating children with neuromotor disorders. While

there are various sources of error & , ) , " " , " # and no widely-

used standards, such techniques have attractions as a

means of improving outcomes for a group of patients

who present management di� culties.

In this paper, we report an assessment of the status of

gait analysis as a method in the rehabilitation of children

with walking disorders. The main focus was on the

evidence of the e� cacy or eŒectiveness of the technology,

as judged by its in¯ uence on patient outcomes. That is,

on the availability of evidence that data from gait

analysis in¯ uence management decisions so that health

outcomes are substantially better than those obtained

using approaches which incorporate older methods of

diagnosis and treatment planning. From the perspectives

of health technology assessment and those who fund

health services, a major question is whether expenditure

Disability and Rehabilitation ISSN 0963± 8288 print} ISSN 1464± 5165 online ’ 2000 Taylor & Francis Ltdhttp:} } www.tandf.co.uk } journals

Dis

abil

Reh

abil

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Yor

k U

nive

rsity

Lib

rari

es o

n 08

/12/

14Fo

r pe

rson

al u

se o

nly.

D. Hailey and J-A. Tomie

Table 1 Examples of studies using gait analysis techniques for pre-treatment decision and} or post-treatment assessment

Study and purpose Measurement

Gait Analysis

Techniques Used Main Findings

Adams et al. " ( Assess eŒect

of dorsal rhizotomy on

foot contact patterns of

spastic CP ; n = 14.

E Tested pre- and post-operatively;

E Compared data to normal age-

matched controls from studies by

others.

E Foot switches taped to

bare feet;

E Computerizd 2D

kinematics;

E Clinical exam Ð 1

observer

E Improvement in foot contact, patterns,

velocity and stride length;

E High incidence of post-operative side

eŒects;

E Improves gait of CP children by

reducing spasticity ; consider

modi® cations to procedure or use of

orthotic devices to deal with side-

eŒects.

Brunt and Scarborough

(18) Describe EMG

patterns in CP, prescribe

treatments based in part

on the EMG ® ndings;

n = 13.

E Compared data to expected

normal data (source not stated);

E Treatments prescibed in part on

basis of EMG± 8 children operated

on, 3 assessed post-operatively.

E Surface EMG;

E Foot ground contact

with switches on feet.

E Could divide the children into 3 groups

based on EMG patterns. Positive

results for 2 of 3 children, mild side

eŒect in the third;

E Concluded use of EMG before surgery

provides an indispensable contribution

to treatment decisions.

Chicoine et al. (19) Assess

eŒect of dorsal

rhizotomy on gait of

CP ; n = 90

E Compared pre- and post-operative

data.

E Observational;

E Videotapes.

E Variance in ability to walk post-

operatively accounted for largely by

the pre-operative gait scores.

Chung et al. # ! assess eŒect

of psoas surgery on hip

function, CP ; n = 34

had psoas surgery, n =

13 CP controls.

E Tested and pre and post-

operatively.

E 2D and 3D kinematics

and kinetics;

E Oxygen consumption.

E Improved pelvic tilt, hip ¯ exion, hip

extension moments and powers,

walking velocity, oxygen consumption

Ð in study group.

E Psoas lengthening improves hip

kinematics} kinetics without sacri ® cing

hip ¯ exor power generation.

Cosgrove et al. $ Assess

eŒect of intramuscular

botulinum toxin on lower

limb management, CP;

n = 26 with abnormally

high muscular activity

E Analysis before and after drug

injections

E Clinical examination,

kinematic analysis ;

E Subjective parent rating.

E Bene® ts, (reduced spasticity, increased

ambulatory status, decreased popliteal

angle of knees during walking)

marginal in 7, good but gradually

relapsed in 12, good, did not relapse

(18 months) in 7.

Damiano et al. # " Assess

eŒect of quadriceps

femoris strength training

on crouch gait in CP ;

n = 14

E Exercised using ankle weights

E Contraction of muscles measured

at diŒerent degrees of knee ¯ exion

E Gait analysis before and after

exercise program

E observational

E electrogoniometer

E computerized kinematics,

E force gains at all angles; 10 of the 14

children exhibited less knee ¯ exion at

initial ¯ oor contact, 7 less knee ¯ exion

at midstance, 5 developed mild

hyperextension

E increased stride lengths

DuŒy et al. # # Describe gait

patterns in SB, assess

eŒect of speci® c surgeries

on their gait ; n = 28 SB,

15 normals

E 10 SB had L4, 8 had L5, and 10

had S1 lesions

E Post-surgical gait analysis.

E Observational

E Videotaping

E Computerized kinematic

and kinetic data

E DiŒerences between SB and controls,

and between the 3 SB groups.

E Iliopsis and tibialis transfers did not

help gait of the SB children.

Gage et al." $ Demonstrate

bene® ts of computerized

gait analysis for pre-

operative decision making,

post-operative assessment ;

n = 20, spastic diplegia.

E Pre- and post-operative gait

analyses compared. Considered

improved or worse if change

greater than 10% (based on

control data, source not

discussed). Also compared with

parental opinion, clinical

assessments.

E Observational, clinical

assessments

E Kinematics, kinetics,

telemetry EMG

E Estimate of walking

e� ciency

E 13 patients improved, 6 were

unchanged, 1 was worse.

E All parents thought their children

improved; 19 children were considered

improved after observational clinical

assessments.

Gage and Ounpuu# $ To

demonstrate use of gait

analysis techniques for

decision making, post-

treatment assessment ;

n = 3.

E Abnormalities described with

reference to kinematics and

kinetics collected from 66

normals.

E EMG

E Computerized kinematics

and kinetics

E Many improvements noted for the 3

patients e.g. decreased pelvic tilts,

improved knee motion and decreased

¯ exion.

276

Dis

abil

Reh

abil

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Yor

k U

nive

rsity

Lib

rari

es o

n 08

/12/

14Fo

r pe

rson

al u

se o

nly.

Gait analysis rehabilitation

Table 1 (cont.)

Study and purpose Measurement

Gait Analysis

Techniques Used Main Findings

E Lee et al. % Compare

outcomes of surgeries

based on clinical analysis

alone with outcomes of

surgeries based on

detailed gait analysis ; n

= 23, diplegics.

E Pre-op and post-op gait analysis,

surgery for 15 children based on

gait analysis, and for 8 based on

clinical analysis alone.

E Compared to normal data, source

not identi® ed.

E Observational

E Videotaping

E Computerized 3D

kinematics

E Surface EMG, ® ne-wire

EMG for deep muscles

E Based on the gait data, 7 did not

improve, 5 of these received surgery

based on clinical analysis alone.

E Conclude that this provides support for

use of gait analysis.

E Scott et al. " % Assess

eŒect of adductor

transfers on gait of CP

patients; n = 33.

E Tested pre-operatively and 1 y

after surgery without computers,

and 9.6 y after surgery with

computerized techniques.

E Observational

E Videotaping

E Photographs and

radiographs

E Goniometer

E Computerized 3D

kinematics

Most patients had a functional increase in

ambulatory status, but the gait analysis

shed light on complications (e.g., pelvic

obliquity). Many patients needed

further surgeries, so procedure was

discontinued at the institution.

E Stott et al. # % To assess

eŒect of tibialis anterior

transfers on gait of SB

children; n = 7, all

community ambulators

and ankle-foot orthoses

users.

E Gait analysis 24 to 60 months

(mean 40) post-operatively

E compared to age-matched

normative data collected by

others.

E Observational clinical

assessment

E Videotaping

E Radiographs

E Computerized kinematics

telemetry EMG

E Energy Expenditure

Index

Decreased calcaneal deformities in all

patients; increased stride lengths and

velocities; decreased EEI. Some

drawbacks (e.g., need for further

surgeries ; spastic lower extremity

muscles ; orthotic devices needed to

facilitate more normal appearance and

energy e� cient gait).

E Sutherland et al. # & Assess

eŒect of intra-

gastrocnemius muscle

botulinum toxin injection

on gait of CP; n = 26.

E Gait analysed before injection and

then at 3 month intervals for 1y.

E Additional injections given at

orthopedist’s discretion.

E Observational clinical

assessment

E Computerized kinematics

E EMG

11 subjects did not complete the study.

Improvements noted in the others.

Conclude results are promising but

that further studies are needed.

CP = Children with cerebral palsy. SB = Children with spina bi® da.

on this high cost technology can be justi® ed by the

bene® ts it produces to the health status of the individuals

for whom it is used.

A more speci® c purpose of assessment was to provide

advice to a rehabilitation hospital which was considering

using the technology in the clinical management of

selected children with cerebral palsy or spina bi® da.

There was interest in the use of gait analysis for pre-

treatment decision making and post-treatment assess-

ment. Methods of gait assessment which are standard

practice at many rehabilitation centres include clinical

examination and observation, using checklists of pati-

ents’ ability to engage in various behaviours, such as the

Gross motor function measure and analysis of video-

tapes. The hospital proposed to continue use of such

approaches and to incorporate computerized kinematics

and kinetics into treatment decisions. Computerized gait

analysis would be used only for severely disabled,

di� cult-to-assess ambulatory children. Annual caseload

of individuals whose walking disabilities were su� ciently

severe to justify such intervention was estimated at

100± 120 from approximately 2500 children with motor

disorders who were patients at the hospital.

Methods

A review was undertaken of available literature on the

clinical use of computerized gait analysis. Data bases

searched included MEDLINE (1986 to April 1997),

EMBASE (1988 to May 1997), PRE-MEDLINE (April

to May 1997), HealthStar (1975 to May 1997) and

PsychInfo (1984 to April 1997). `Gait ’ , `movement ’ ,

`motor performance ’ , `walking’ , `cerebral palsy ’ ,

`neural tube defects’ , `spinal cord malformation’ and

`congenital disorders ’ were the subject words used.

Tables of contents for Gait and Posture were viewed on

the Elsevier Publishing Company’ s home page on the

World Wide Web. The search results were screened and

the literature considered to be the most relevant was

obtained. Reference lists in retrieved literature were also

screened. Other information was obtained by consul-

tation with persons with expertise in computerized gait

analysis. Cost data were obtained from Canadian

rehabilitation centres and the provincial health ministry.

Studies comparing outcomes of treatments decided by

modern gait analysis techniques with those decided by

clinical assessment alone were most wanted. Other

277

Dis

abil

Reh

abil

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Yor

k U

nive

rsity

Lib

rari

es o

n 08

/12/

14Fo

r pe

rson

al u

se o

nly.

D. Hailey and J-A. Tomie

literature of interest included reports on reliability and

validity of gait analysis and reports on studies in which

gait analysis was employed for pre-treatment decision or

post-operative assessment purposes. Further details of

the search methodology and retrieved literature are

available from the authors.

Results

e v i d e n c e o f e f f i c a c y

From the information located in the literature, it

appeared that evidence of clinical bene® t from use of

modern gait analysis techniques is limited. Only one

study, by Lee et al.% , was found that compared outcomes

of treatments based on clinical analysis alone with those

of treatments based on modern gait analysis techniques

(table 1). In that study, 15 diplegic children received

surgery based on gait analysis ® ndings and 8 received

surgery based on clinical analysis alone. The gait of 7

children, 5 of whom had received treatment based on

clinical analysis alone, did not improve after surgery. It

was concluded that this provided support for use of gait

analysis techniques in clinical practice.

Other reports on use of gait analysis techniques for

pre-treatment decision making or post-treatment asse-

ssment resembled the remainder of the studies shown in

table 1, which are representative of those identi® ed from

Table 2 Examples of studies using gait analysis for descriptive or categorizing purposes

Study and purpose Gait analysis techniques used Main ® ndings

Ho� nger et al. # ’ Evaluate role of the

hamstrings in CP patients with crouch

gait, n = 16.

E Computerized kinematics.

E Surface EMG.

E Hamstrings were not short, and were assisting

with hip extension in many individuals.

E Conclude that surgeons should be judicious

with hamstring lengthening for crouch gait.

Hullin et al. # ( Describe and classify gait

abnormalities in CP patients; n = 26.

E computerized 3D kinematics and kinetics Based on the gait analysis data, could divide the

children into 5 groups.

Kelly et al. # ) Compare observational with

kinematic gait analysis, derive formula for

quality of walking score ; n = 55, 5

normals.

E Observational functional walking test

E Video analysis

E Sagittal plane kinematic analysis

E Found that the features most discriminatory

in determining a normal from an abnormal

gait were balance, speed, range of motion and

apparent eŒort.

Sojka et al. # * Describe gait de® cits of CP

children; n = 6, 4 controls.

E Observational

E Videotaping

E Computerized kinematics footswitches

E Surface EMG

E Children could be divided into 2 groups,

based on gait data.

E Because of diŒerences between children,

similar treatments will not necessarily lead to

similar outcomes. Further study needed.

Tylkowski and Howell " & Describe and

classify CP children with crouch gait,

n = 46.

E Observational, with goniometers

E Computerized kinematics,

E Surface EMG

Based on the gait data, could divide patients into

3 groups; not able to do this with clinical

analysis alone. Results show that gait labs can

diŒerentiate causes more speci® cally than

clinical examination alone.

Vankoski et al. $ ! Describe and classify gait

abnormalities of SB children; n = 21

E Kinematics E Could divide the SB children into 2 groups,

based mainly on diŒerences between them at

the pelvis and knee, and muscle weakness.

CP = Children with cerebral palsy. SB = Children with spina bi® da.

review of the literature. Most had small sample sizes.

Also, there was a focus on the magnitude of within-

subject pre-treatment versus post-treatment changes,

rather than on the eŒect of gait analysis data in

in¯ uencing these. Some studies used control data sets

that were not well de® ned. Few attempted to compare

decisions based on clinical analysis alone with decisions

based on modern gait analysis techniques and none

considered cost issues.

Two other studies noted in table 1 are of interest in

indicating potential bene® ts from use of gait analysis.

Gage et al." $ compared the post-operative gait data of 20

diplegic children to pre-operative and control data.

When gait was assessed clinically or by the children’ s

parents, 19 of the 20 individuals were considered to have

improved post-operatively. When gait was assessed with

modern techniques, 13 of the 20 children were considered

improved. Gage et al. conclude that modern gait analysis

techniques allow more objective planning and docu-

mentation of surgery, and as a result, surgical judgment

should improve and errors in treatment should be

reduced.

In the study by Scott et al." % the gait of 33 persons with

cerebral palsy was assessed pre-operatively, 1 year after

tendon transfer surgery and then at an average of 9.6years after surgery. All patients showed functional

improvement post-operatively, but the computerized

278

Dis

abil

Reh

abil

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Yor

k U

nive

rsity

Lib

rari

es o

n 08

/12/

14Fo

r pe

rson

al u

se o

nly.

Gait analysis rehabilitation

Table 3 Indicative annual costs for a gait analysis facility

Component Annual cost, $CAN

Capital items ‹ 50,000

PersonnelΠ100,000

Maintenance and upgrade 40,000

O� ce supplies 10,000

Space rental and

utilities

10,000

Total 210,000

‹ Computer equipment and software $150,000 . Cameras $70,000,

Forceplates $30,000. Assumed that equipment is depreciated over ® ve

years (undiscounted and ignoring interest foregone).

Œ Director $50,000, Technician $35,000 Part time clerical staŒ$15,000

Source : Canadian rehabilitationcentres and provincial health ministry.

gait analysis conducted at long term follow-up revealed

complicating factors in many individuals, which resulted

in abandonment of this surgical procedure.

Table 2 shows studies representative of those that used

computerized gait analysis for descriptive or categorizing

purposes. In several of these, gait analysis helped to

classify children with cerebral palsy or spina bi® da

within groups who were otherwise diagnostically indi-

stinguishable. In the study by Tylkowski and Howell" &

children with cerebral palsy could be divided into three

groups on the basis of computerized gait data. The

authors concluded that gait laboratories can diŒerentiate

causes more speci® cally than clinical examination alone.

Results reported in the literature seem to support

claims that computerized gait analysis is helpful in

detecting gait subtleties, within Ð subject gait changes

and between Ð subject gait diŒerences. However, the

available data do not allow conclusions to be drawn

about the eŒects of gait analysis on management

decisions or subsequent health status. In terms of a

widely-used classi® cation system " ’ that considers the

type of study design and conditions of scienti® c rigour,

the quality of evidence of clinical bene® t is poor.

Controlled, long term studies are still required.

c o s t c o n s i d e r a t i o n s

Indicative estimates of the costs of establishing and

operating a gait laboratory are shown in table 3. It is

assumed that a gait analysis system would incorporate

six cameras and two forceplates. Overall annual expe-

nditure could be of the order of $200000. For a caseload

of 100± 120, cost per examination could be $1800 Ð

$2000. A substantial proportion of patients might be re-

examined in the gait laboratory following treatment so

that overall cost per patient would be substantially

higher than the cost per examination. The eventual

impact of re-examinations on the availability, scheduling

and cost of gait analysis is uncertain and would require

close consideration. The cost of a gait laboratory would

also be aŒected by acquisition of other gait analysis

equipment, such as EMG.

This level of cost resembles that for various advanced

diagnostic imaging methods. Costs are high, but might

be justi® ed by the impact of the additional diagnostic

data on the management decision and subsequent patient

outcome.

Discussion

Better information on the gait characteristics of

children with walking disabilities can be expected to

improve decision making on their subsequent treatment.

In principle, gait analysis appears a useful method of

increasing the eŒectiveness of surgical and other inte-

rventions. Avoidance of additional surgery or long-term

consequences of failed procedures which cannot be

corrected would be major gains. There would also be

expectation of greater independence and improved

quality of life for children with cerebral palsy or spina

bi® da who had severe walking problems. It is, however,

unclear how many children would be aŒected in this way

through the availability of additional data provided.

While there is a considerable literature on use of

computerized gait analysis techniques in research, evid-

ence about their clinical usefulness appears to be limited.

There is an absence of data on the incremental bene® ts of

computerized gait analysis in terms of patient outcomes

following treatment. Also, information on comparative

costs is lacking.

A possible issue for further resolution is the extent to

which all the information obtained from multivariate

measurements can be condensed to provide a reliable,

global evaluation technique for those providing trea-

tment. There is also the question of the extent to which

the results of gait analysis at re-assessment after

treatment should be used as an outcome measure, as

compared to clinical assessment and the opinions of

patients and their families.

On the basis of information available for the asse-

ssment, computerized gait analysis in the rehabilitation

of children with walking disorders is a promising but

developmental technology. Given the absence of info-

rmation on its clinical bene® ts, if a centre were to

introduce gait analysis, it would be necessary to validate

the eŒectiveness of the technology. As a ® rst step, it

would be important to document the eŒectiveness of

current management options for children with severe gait

di� culties. It would also be necessary to put in place

279

Dis

abil

Reh

abil

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Yor

k U

nive

rsity

Lib

rari

es o

n 08

/12/

14Fo

r pe

rson

al u

se o

nly.

D. Hailey and J-A. Tomie

mechanisms for systematic, long-term collection of

outcomes data for children who had been assessed using

computerized gait analysis. Elements to be considered

would include patient selection, eŒects on management

decisions, eŒects on patient outcomes and cost impact.

These conclusions, we suggest, seem generally appl-

icable to the use of gait analysis in the rehabilitation of

those with walking disabilities. The patients whom it is

intended to help, their families, health care professionals

and those that provide funding for health services deserve

assurance that use of this method will be associated with

appropriate clinical bene® ts. Application of this rela-

tively high cost diagnostic method may represent good

value for money. However, until good quality evidence

of bene® t emerges, gait analysis in the management of

those with walking disabilities should be regarded as

investigational and operated on that basis.

References

1 DeLuca PA. The musculoskeletal management of children withcerebral palsy. Pediatric Clinics of North America 1996 ; 43 :1135± 50.

2 Cosgrove AP, Cory IS, Graham HK. Botulinum toxin in themanagement of lower limb cerebral palsy. Developmental Medicineand Child Neurology 1994 ; 36 : 5386± 96.

3 Bloswick DS, King EM, Brown D et al. Evaluation of a device toexercise hip extensor muscles in children with cerebral palsy: aclinical and ® eld study. Assistive Technology 1994 ; 6 : 147± 51.

4 Lee EH, Goh JC, Bose K. Value of gait analysis in the assessmentof surgery in cerebral palsy. Archives of Physical Medicine andRehabilitation 1992 ; 73 : 642± 6.

5 Rose SA, O4 unpuu S, DeLuca PA. Strategies for the assessment ofpediatric gait in the clinical setting. Physical Therapy 1991 ; 71 :961± 80.

6 Gage JR. Gait Analysis. An essential tool in the treatment ofcerebral palsy. Clinical Orthopaedics and Related Research 1993 ;228 : 126± 34.

7 Patrick J. Gait laboratory investigations to assist decision making.British Journal of Hospital Medicine 1991 ; 45 : 35± 7.

8 Drouin LM, Malouin F, Richards CL, Marcoux S. Correlationbetween the gross motor function measure scores and gait spatio-temporal measures in children with neurological impairments.Developmental Medicine and Child Neurology 1996 ; 38 : 1007± 19.

9 Kirkpatrick M, Wytch R, Cole G, Helms P. Is the objectiveassessment of cerebral palsy gait reproducible? Journal of PediatricOrthopaedics 1994 ; 14 : 705± 8.

10 Stillman B, McMeeken J. Use of a video time display in determininggeneral gait measures. Australian Journal of Physiotherapy 1996 ;42 : 213± 7.

11 McCoy A, Rodda J. The gateway to gait analysis for children inAustralia. Australian Journal of Physiotherapy 1996 ; 42 : 331± 6.

12 Watts HG. Gait laboratory analysis for preoperative decisionmaking in spastic cerebral palsy: Is it all it’ s cracked up to be?Journal of Pediatric Orthopaedics 1994 ; 14 : 703± 4.

13 Gage JR, Fabian D, Hicks R, Tashman S. Pre- and postoperativegait analysis in patients with spastic diplegia. A preliminary report.Journal of Pediatric Orthopaedics 1984 ; 4 : 715± 25.

14 Scott AC, Chambers C, Cain TE. Adductor transfers in cerebralpalsy: long-term results studied by gait analysis. Journal of PediatricOrthopedics 1996 ; 16 : 741± 6.

15 Tylkowski CM, Howell VL. Crouch gait in cerebral palsy.International Pediatrics 1991 ; 6 : 153± 60.

16 Jovell AJ, Navarro-Rubio MD. Evaluacion de la evidenciacienti® ca. Medicina Clinica 1995 ; 105 : 740± 3.

17 Adams J, Cahan LD, Perry J, Beeler LM. Foot contact patternfollowing selective dorsal rhizotomy. Paediatric Neurosurgery 1995 ;23 : 76± 81.

18 Brunt D, Scarborough N. Ankle muscle activity during gait inchildren with cerebral palsy and equinovarus deformity. Archives ofPhysical Medicine and Rehabilitation 1988 ; 69 : 115± 7.

19 Chicoine MR, Tae Sung Park, Vogler GP, Kaufman BA. Predictorsof ability to walk after selective dorsal rhizotomy in children withcerebral palsy. Neurosurgery 1996 ; 38 : 711± 4.

20 Chung CY, Novacheck TF, Gage JR. Hip function in cerebralpalsy± the kinematic and kinetic eŒects of psoas surgery. Gaitand Posture 1994 ; 2 : 61.

21 Damiano DL, Kelly LE, Vaughan CL. EŒects of quadricepsfemoris muscle strengthening on crouch gait in children with spasticdiplegia. Physical Therapy 1995 ; 75, 658± 67.

22 DuŒy CM, Hill AE, Cosgrove AP, Corry IS, Mollan RAB, GrahamHK. Three-dimensional gait analysis in spina bi® da. Journal ofPediatric Orthopedics 1996 ; 16 : 786± 91.

23 Gage JR, O4unpuu, S. Surgical intervention in the correction of

primary and secondary gait abnormalities. In : Aftab E and Patla,(eds) Adaptability of human gait: Implications for the control oflocomotion. Advances in psychology, No. 78. Amsterdam, North-Holland, 1991; 78 : 359± 85.

24 Stott NS, Zionts LE, Gronley JK, Perry J. Tibialis anterior transferfor calcaneal deformity: a postoperative gait analysis. Journal ofPediatric Orthopedics 1996 ; 16 : 792± 8.

25 Sutherland DH, Kaufman KR, Wyatt MP, Chambers HG. Injectionof botulinum A toxin into the gastrocnemius muscle of patientswith cerebral palsy: a 3-dimensional motion analysis study. Gaitand Posture 1996 ; 4 : 269± 79.

26 Ho� nger SA, Rab GT, Abou-Ghaida H. Hamstrings in cerebralpalsy crouch gait. Journal of Pediatric Orthopedics 1993 ; 13 : 722± 6.

27 Hullin MG, Robb JE, Loudon IR. Gait patterns in children withhemiplegic spastic cerebral palsy. Journal of Pediatric Orthopaedics.Part B 1996 ; 5 : 247± 51.

28 Kelly IP, O’Regan M, Jenkinson A, O’Brien T. The qualityassessment of walking in cerebral palsy. Gait and Posture 1997 ; 5 :70± 4.

29 Sojka AM, Stuberg WA, Knutson LM, Karst GM. Kinematic andelectromyographic characteristics of children with cerebral palsywho exhibit genu recurvatum. Archives of Physical Medicine andRehabilitation 1995 ; 76 : 558± 65.

30 Vankoski SJ, Sarwark JF, Moore C, Dias L. Characteristic pelvic,hip and knee kinematic patterns in children with lumbo-sacralmyelomeningocele. Gait and Posture 1995 ; 3 : 51± 7.

280

Dis

abil

Reh

abil

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Yor

k U

nive

rsity

Lib

rari

es o

n 08

/12/

14Fo

r pe

rson

al u

se o

nly.