Machine Translated by Google

This is an enhanced PDF from The Journal of Bone and Joint Surgery

The PDF of the article you requested follows this cover page.

Gait patterns in spastic hemiplegia in children and young adults

TF Winters, JR Gage and R Hicks
J Bone Joint Surg Am. 1987;69:437-441.

This information is current as of November 6, 2008

Reprints and Permissions
Publisher Information
Click here to order reprints or request permission to use material from this
article, or locate the article citation on jbjs.org and click on the [Reprints and
Permissions] link.
The Journal of Bone and Joint Surgery
20 Pickering Street, Needham, MA 02492-3157
www.jbjs.org
Machine Translated by Google

Copyright 987 by The Journal of Bone and Joint Surgery. Incorporated

Gait Patterns
in Spastic Hemiplegia

in Children
and Young Adults*

BY THOMAS F. WINTERS, JR., MDt, JAMES R. GAGE, MD1, AND RAMONA HICKS, M.A., RPT1,
NEWINGTON , CONNECTICUT

desde the Kinesiologv Laboratory, Newingion Children's Hospital, Newingion

ABSTRACT: Four homogeneous in patterns ofgait were of locomotive coordination, and the rates at which muscles

defined forty-six patients palsy who had spastic hemiplegia are stretched also may influence has the findingsmo. are The result

secondary brain cough or other neurological dis been that variable regimes who, from the of treatment prescribed
orders by analyzing kinematic data in the sagittal plan for many patients clinical standpoint, appear

and electromyographic data. InGroup I (twenty pa to have similar involvement of the central nervous system.

agents) the primary swing abnormality was a drop foot patients in Group in the We decided to study whether more pertinent information

phase. tight heel The thirteen phase as well as a drop foot II had a could be obtained by use of new laboratory tests: dynamic

cord in the stance in the swing phase. had electromyography and computerized rotation of measurements of the

more proximal tion of the knee) The five patients inGroup III also joints. These tests allow the physician to evaluate
involvement (that is, restricted mo the function of the muscles and the position of the joints

as well as an equinus IV, the deformity of the during gait and provide a more extensive and objective
ankle. In Group eight patients of the hip. had, in addition, assessment of the neural deficit in patients who have spastic
restricted motion disorders.

In 1977, Bekey et al. classified categories the spastic gait of thirty

The term spastic icit hemiplegia is used for the neural def patients using into four
electromyographs muscles that activate motion of the foot of the

that may occur in patients who has a neurological and the ankle. They
disorder such as cerebral palsy, traumatic injury to the brain, concluded that is accurate description of the various patterns
or cerebrovascular accident The word hemiplegia means of gait is possible lower if the patterns of movement of the entire

the neuromuscular disorders that involves one-half of the extremity Chang and also were evaluated. In a follow-up study,

body in the front plan while the other half is normal or Bekey discussed of patterns to the how they were able to apply
near normal. Cerebral palsy is generally taken to mean a expected prediction recognition gait that could be of the changes in

non-progressive that beginsdisorders

either in utero or within postoperatively. Knutsson and

the first two years of life, loss of muscle with a selective Richards electromyographic
combined light photography and studies and inter

control iO Many of the characteristics palsy that are seen in patients mitten found three predominant terns of gait in patients who had spastic pat
who has brain are similar to those in patients who hemiplegia secondary to cerebrovascular accidents; one pattern was that was

have injury to the brain or a cerebrovascular accident in

which the onset occurs later in life. similar to the pattern found in our Group-Il patients and two

Surgery, bracing, to or both, have been commonly who have used patterns were similar to that in our Group-Ill patients.

improve the gait in patients but until recently spastic hemiplegia,

Materials and Methods
the treatment was entirely empirical. static

muscle tests and observation of the patient's gait are done Forty-six patients who had spastic hemiplegia went to computerized under

to determine which muscles are overactive and in need of analysis of gait at the Newington Chile

lengthening. estos tests, however, do not yield critical in drain's Hospital Kinesiology Laboratory. The thirty-three
training from patients for who has injuries of the brain be- male and thirteen female patients had an average 1 1 .2 years age of
cause manual ability to testing techniques muscle depend on the (range, 3.4 to 30.5 years). Thirty-eight had ce

patient's selectively activate simultaneously releasing a particular muscle rebral palsy; six, traumatic injury to the brain; juvenile cerebrovascular and two,
while patients who have the conditions the antagonist, and often accident. The patients with the two

unable to coordinate these two muscular that are under study are The latter diagnoses were tested at least two years after the onset
functions. It varies andthe
of the hemiplegia to have reached were believed

bles such as posture of the limbs, vestibular tone, patterns limit of neurological recovery. None of the patients had had

orthopedic surgery before the analysis and all could walk

*No benefits in any form have been received or will be received from without orthoses or other aids.
independently
to commercial party related directly or indirectly to the subject of this study. article.
No funds were received
In the Kinesiology Laboratory, we used four systems
in support
Orlando Regional Medical Centre, 1315 South Orange Avenue, of measurement that were connected to a supervising Maynard, corn-
Suite D, Orlando, Florida 32806.
puter (Digital Equipment Corporation, Massa
Kinesiology Laboratory, Newington Children's Hospital, Newing to
ton. Connecticut 061 1 1 . por favor address requests for reprints Dr. Gage. chusetts): a three-dimensional movement camera (United

VOL. 69-A, NO. 3. MARCH 1987 437

Machine Translated by Google

438 TF WINTERS, JR., JR GAGE, AND RAMONA HICKS

Three-channel Research Center, East Hartford , Connecticut), visual camera compares the speed of walking in the and rotations of the joints
technologies (Panasonic, Secaucus, New sagittal plane of the four groups of normal gait. with the parameters

Jersey), a force measurement with two force plat system

forms (Advanced Mechanical Technology, Newton, Mas Group I consisted of twelve boys and eight girls, whose

sachusetts), and an eight-channel electromyographic average was 9 . S years (range , teen patients 5 . 2 to I 5 . 8 years). Nine age

telemetry system (Biosentry, Torrance, California). The had cerebral injury palsy and one had trauma

electromyographic telemetry system monitored electrodes signals from to the brain. The patients in this group had the mildest
surface amplified City, (Motion Control, salt lake deviations from normal gait. The average speed of walking per second; the normal

Utah) or implanted All patients fine-wire electrodes, or both45. was 97.3 centimeters speed for child drain who are between the ages of
had an orthopedic the analysis they walked evaluation of eight and fourteen is 1 16.6 centimeters years
as part
in which
at self-selected during testing. Two runs of kinematic each subject and speeds per second7. The most significant normality ab

were compared have spastic hemiplegia the differences data were done for of gait was a drop foot in the swing phase (Fig.

run are usually gross differences study. In visually. In patients who 1), but equally important was the finding that they had an
patients who had minor differences, more closely in data from run to adequate range of dorsiflexion (average, 12 degrees) (Table
approximated minor. Any patients whose analysis showed I) of the ankle during the stance phase of gait. The maximum

Although analysis between runs were eliminated from the amount of dorsiflexion in the stance phase was more than

was done in all patients, patients into the four groups the the data that zero degrees in all of the Group-I patients. from normal gait Other deviations

recordings the plane of progression normal were selected. that were recorded in this group increased of the knee at terminal swing included

lower extremity) sify of three-dimensional motion mea- phase, at flexion

surements patients who could not be grouped classification of the initial contact, and in loading hyperfiexion of
response;
the
to be described was based on hip during the swing phase; and increased pelvis throughout lordosis of the

that were made in the sagittal plan (that is, the gait cycle (Fig. 1).

for the three major joints . of the Group II consisted of of thirteen patients, nine boys and
Electromyography data were used to class four girls, whose average age was 10.2 years (range, to 15.8 years). 4.3

by kinematic data Eleven patients had cerebral injury The palsy; one pa
alone. tent, traumatic to the brain; and one, a cerebrovas
cular accident average speed of walking in Group II for the patients
Results
was 90.0 centimeters per second. The gait in
Four distinct patterns of gait were identified. Table I inflexion this group was characterized by plantar that persisted

GROUP I GROUP II GROUP Ill GROUP IV NORMAL

Light

LU
Zr

I_____Gilt CVC1S
Csat Ccis Gilt CcIs CeitCcLi c.i'.cci.
FOOT DROP PATTERN I PLUS PATTERN II PLUS PATTERN III PLUS
IC KNEE FLEX.1 ST PLANTARFLEX. KNEE MOTION HIP MOTION
j
H.I.P. FLEXION KNEE HYPEREXT. TST LORDOSIS
t
LORDOSIS

FIG. Yo

The charts indicate the mean (solid line) and standard lines) of rotation of thedeviations
joints in the sagittal
(dotted normal values. plan for each group and for
Features of each group are summarized at the bottom of the chart using the following abbreviations: IC = and TST - terminal stance. initial contact, ST = stance,
Machine Translated by Google

GAIT PATTERNS IN SPASTIC HEMIPLEGIA 439

TABLE I

DATA ON AGE AND PARAMETERS OF GAIT IN NORMAL SUBJECTS AND HEMIPLEGIC GR0uPS*

Normalt Grou p I Grou p II Group III Group IV

Rotation of the joints

during gait (degrees)
Maximum dorsiflexion of the ankle 4±6 - 1±6 - 16 ± I 1 - 21 ± 8 - 14 ± 10
(swing phase)
Maximum dorsiflexion of the ankle 10 ± 4 12 ± 6 - 7 ± 10 - 6 ± 12 - 8 ± 14
(stance phase)

Total flexion- of 56 ± 4 51 ± 8 60 ± 10 37 ± 8 33 ± 12
extension the knee
Total flexion- 42 ± 3 44 ± 7 46± 6 46 ± 10 29 ± 4
extension of the hip
Age(vrs.) 1I 10 ± 4 10 ± 4 15 ± 12 12 ± 6

Speed of walking (cm/ sec.) I 17 ± 14 97 ± 19 90 ± 20 80 ±24 84±22

* Mean and standard deviation.

t Values obtained from Newington Children's Hospital.

throughout the gait cycle (Table I). The patients in this group showed a progression of involvement proximally, and had

also demonstrated full extension or hyperextension of the more limited flexion of the knee during the swing phase
knee in the stance phase (Fig. the hip and 1) as well as hyperfiexion of than Group-lI patients. during Total flexion-extension of the knee

increased lumbar lordosis throughout the gait gait was less than 45 degrees subjects patients to in Group-Ill patients;

cycle. the value for normal patients is 56.3 degrees. in Group-Ill

Group III consisted of five male patients, 4.9 to whose av- they also hadwere similar of the hip and
hyperfiexion Groups I and II in that

age age was 14.5 years (range, patients palsy; 30.5 years). Two increased lumbar

two, brain; andcerebral had

one, a cerebrovascular velocity of traumatic injury The to the lordosis.

walking patients in this group accident average Group IV consisted of eight patients, one female. seven male and
was 79.7 centimeters per second. The The average age was 12.2 years (range, 21 .5 years). 3.4 to

had planted flexion at the ankle, Six patients cerebral had palsy and two, trau

NORMAL
7(7 / / /LL
NORMAL

NORMAL

TOE OFF

* GAIT CYCLE

FIG. 2

The stick figures of the lower swing extremity and the electromyographs are representative of the patients in Group III. Flexion of the knee during that the
phase is limited because they are of the concomitant activation of the hamstring in and rectus femoris muscles. The bar graphs indicate normal values
based on data that were collected at the Shriners Hospital San Francisco. California2.

VOL. 69-A, NO. 3. MARCH 1987

Machine Translated by Google

440 TF WINTERS, JR., JR GAGE, AND RAMONA HICKS

matic injury to the brain. This group had the most neuro The average tremity was more involved in Groups in the patients in Group III than

logical involvement. velocity for walking per second. The ankle was was in those gait of I and II. Waters et believed that the

84.4 centimeters plantar of the knee was restricted, and the hip flexed, stiff-legged some hemiplegic patients is a regression
movement showed limb of the hip to primitive in locomotor patterns. These patterns are present
ited flexion-extension. Total flexion-extension quadrupeds that depends on the extensor reflex for stability

during gait was less than 35 degrees in Group-IV patients; in stance and activation ofthe lowerextremity. This extender

the value in normal subjects is 42. 1 degrees. Group-IV reflex occurs during terminal swing phase. There is a strong

patients compensated for limited motion of the hip by in stance of gait reflective tie between contraction of the quadriceps muscles,

creasing pelvic lordosis during terminal extension of the hip, and plantar fore, the flexion of the ankle; there during gait in

(Fig. 2). equinus position is normal ruped. quad reflex at the ankle has

In bipeds, the extensor blocked been

Discussion
so that humans do not normally initiate the stance
Group I phase with the foot plantar patients flexed.

The pattern of gait in the patients in Group I was best In the Group-Ill the extender reflex remained

characterized by plantar in an equinus flexion of the ankle in the swing at the hip and knee to resist the flexor thrust; in fact, extender

phase, resulting The deformity at initial contact. tone is enhanced when one rises to the upright position. TO

heel cord was not tight since there was adequate siflexion phase. dor central injury in Group-Ill patients released the plantar at the ankle from its reflex

in the stance To compensate for the drop inhibitory stiff gait with short steps that the block'2. The result was the

foot, the knee hyperflexed increased at foot strike, forcing the hip into Group-Ill onstrated. patients dem

flexion in order to maintain the body in a position

centered over the foot and to help clear the swinging limb In subjects who have a normal gait, the knee flexes to

with the drop foot. The pattern created lordosis of pelvic tilt showed in 35 degrees in the late phase of stance, just before toe-off.
throughout the gait cycle. Flexion increases to 65 degrees by the middle of the swing
classic treatment for these patients has been to after the activity of the quadriceps the Group-Ill muscles you have ceased'2. in

lengthen the heel cord. Such treatment, the however, dog patients, however, string muscles remained the quadriceps and ham

worsen the gait since pathological injury is not a short activate. This simultaneously contraction

ening of the gastrocnemius by and soleus muscles as evidenced tion of flexor and extensor muscles limited flexion of the

dorsiflexion of more than zero degrees during the stance knee during swing. Electromyography using surface A.M-

phase. It is more likely that the problem underactivity activity is weakness or plified electrodes confirmed this by showing during the abnormal

of the previous tibial muscle relative to over- activity of these muscles that the swing phase (Fig. 2).

of the gastrocnemius and soleus muscles. We believe loss of coordinated contraction of the quad of the

riceps and hamstring muscles is the primary cause

group II
decreased flexion of the knee that occurs during the swing

The patients in Group of the II had a static or dynamic and soleus with phase6.
tracture gastrocnemius in persistent muscles that re- Knutsson and Richards also found hyperextension of

sulted plantar stance and swing flexion of the ankle during the the knee in the stance phase and decreased flexion of the

phases. Perry9 stated that 15 degrees of of found

knee in the swing phase in one-third patients. ever, they also their decreased How

plantar flexion of the ankle places the trunk behind the foot electromyographic of two or more muscle groups and believed activity
unless the knee is hyperextended, the hip is flexed, or the the hyperex
heel is elevated by external support. flexion, the tibia When the ankle is in tension of the knee compensated rather than for weakness of the muscles

fixed plantar to and foot function together ace was secondary to plantar creased electromyographic flexor spasticity. Of

long lever that will not allow the usual rocker motion of activity Group-Ill patients. was not found in our

the tibia on the foot. Este forces the knee into hyperextened stages In one-sixth of the patients Knutsson in their series,
Zion in the middle and terminal of stance. In addition, and Richards concomitant found activation (ace

advancement of the trunk is curtailed and the length of the shown by electromyography) in four to six muscles groups

opposite step is decreased. over the To maintain the center of gravity and decreased flexion of the knee during were the swing phase.

foot, flexion of the hip and pelvic lordosis were These findings true of all of our patients who were in

increased, as in the patients who were in Group II. group III.

In the study of Knutsson and Richards one-third of the

to cere- Group JV
twenty-six patients who had secondary hemiplegia demonstrated
brovascular accident a pattern that was similar In Group-IV patients, as in Group-Ill patients, reflex was implicated the ex

to that in our Group-Il plantar patients. The main disturbance in was tensor because they had decreased
flexion of the ankle that resulted hyperextension movement at the hip and knee and plantar of motion in flexion at the ankle.

of the knee. The reduction the sagittal difference plan at the hip
constituted the crucial between Groups III and
Group III IV. Increased of the adductors
activity iliopsoas and hip
The musculature in the proximal part of the lower ex prevented the hip from reaching full extension at terminal

THE JOURNAL OF BONE AND JOINT SURGERY

Machine Translated by Google

GAIT PATTERNS IN SPASTIC HEMIPLEGIA 441

stance phase. The length of the stride would have been ofthe condition should be specific For to the neurological of deficit.

severely shortened without the compensatory increase in example, patients who have a pattern our Group I do not gait similar to

former pelvic tilt. require a lengthening tendon since the gastrocnemius shortened. of the Achilles

In general, there was a progression The of involvement in and soleus muscles are not

the four groups. average speed of walking was approximately for Groups A leaf-spring orthosis foot (a flexible plastic ankle control the
I and II combined ninety-five per second compared with centimeters orthosis) should adequately Patients who have drop foot.

approximately eighty-three timers per second for Groups III and IV (p < 0. 1 ). cen a similar gait to our patients in Group
Patients II usually will require lengthening of the plantar flexors of

in Groups I and II appeared to have the least residual damage the involved ankle as well as a leaf-spring the orthosis. In order

to the center nervous system as compared with patients Groups III in to develop patient's ability to walk to the maximum,

and IV . It is of interest that the patients in Groups treatment will probably have to extend to the knee for patients who have a gait similar to
I and II combined were significantly younger than those in our Group-Ill patients and

Groups III and IV combined (approximately ten and thirteen to the knee and the hip for those who have a similar gait

years, respectively, 0.05). p< to our Group-IV patients.

Since spastic hemiplegia of deficits, in young patients encompas- logically

Noit: The authors thank Marianna Hospital. for Nelson. Research Coordinator at Ncwington Chtldrcn..
ses a spectrum it follows that treatment her editorial assistance.

References

1 . BEKEY, GA; CHANG, Proc. CW; PERRY, J.; and HOFFER, MM: Pattern-Recognition 65: 674-681 of Multiple EMG Signals Applied to Description of Human
Gait. Instit. Electr. and Electron. Eng. , , 1977.
2. BLECK, EE: Orthopedic CW . of Cerebral
Management WB Saunders, Palsy, p. Prediction
62. Philadelphia,
of Post-Operative
GA: Pattern Recognition Applied to the 1979.
3. CHANG, and BEKEY, the I 7th Symposium and Control, Including on Adaptive Institute pp. 674-681 . Piscataway, Institute of Gait in Man. In Proceedings of the Ninth
Conference onDecision Engineers, 1979. Electrical and Electronic Processes. Electrical and Electronic Engineers Control
Systems
Society, 4. GAGE, JR: Gait Analysis for Decision-Making in Cerebral Palsy. Bull. Hosp. Joint Dis., 43: 147-163, 5. GAGE, 1983.
JR; FABIAN, HICKS, RAMONA; and TASHMAN,
David; SCOTT: Pre- and Postoperative Gait Analysis Preliminary Report. J. Pediat. Orthop. , 6. GAGE. JR ; PERRY. in Patients with Spastic Diplegia. TO
4: 715-725. 1984.
J. ; HICKS, R. ; Koop. S. ; and WERNTZ. J.: Rectus Femoris Transfer as a Means of Improving Knee Function in Brain

Palsy. Devel. Med. and Child Neurol. . in press.

7. HICKS, R.; TASHMAN, S.; CARY. JM; ALTMAN. RF; and GAGE, JR: Swing Phase Control with Knee Friction in Youth Amputees. J.

Orthop. Res., 3: 198-201, 1985.

8. KNUTSSON. EVF.RT, and RICHARDS, CAROL: Different. Types of Disturbed Engine Control in Gait of Hemiparetic Patients. Brain. 102: 405-430.
1979.

9. PERRY, JACQUELIN: Kinesiology of Lower Extremity Bracing. Clin. Orthop., 102: 18-31, 1974.
10. PERRY, JACQUELIN: GI0vAN, PETER; HARRIS, L. J. ; MONTGOMERY, JACQUELINE; and AZARIA, MORRIS: The Determinants of Muscle action in
the Hemiparetic Lower Extremity (and Their RL; Effect on the Examination Procedure). Clin. Orthop. , 131: 71-89, 1978.
11. WATERS. PERRY, J.; and GARLAND, DOUGLAS: Surgical Correction of Gait Abnormalities following Stroke. Clin. Orthop.. 131: 54-63.
1978.
12. WATERS. RL; GARLAND, OF; PERRY. JACQUELIN: 61- HABIG, TERRY; and SLABAUGH. PETER: Stiff-Legged Gait in Hemiplegia: Surgical Correction.

J. Bone and Joint Surg. , A: Sept. 927-933. 1979.

VOL. 69-A, NO. 3. MARCH 1987