ANTERIOR VERSUS POSTERIOR WALKERS:

A GAIT ANALYSIS STUDY

Lynne Logan
Kathleen flyers-Hinkley
Charles D. Ciccone

Therapeutic habilitation of young child-
ren has often included the use of walkers
as mobility aids (Bleck 1987). When
properly prescribed and implemented, a
walker can help provide mobility and
enhance the ambulation skills of children
with cerebral palsy, Down syndrome and
similar developmental disabilities (Cusick
1980, Click and Davies 1985, Wilson
00
1988). Traditional walkers have four
d vertical legs, a horizontal bar in front of
0
the child, and horizontal bars on either
3- side which also serve as hand-grips (Licht
m- 1969). This type of walker can be
m
considered an ‘anterior’ walker, since the
0-
OI child pushes it and walks behind it
Q‘
(Fig. 1). There are numerous variations to
-e5
0 this traditional anterior design, including
fitting wheels on two or more of the legs,
2 varying the position of the hand-grips,
2 providing forearm or trunk support, and
is adjusting the height, weight and shape of
2 the walker’s base (Finnie 1975, Bergen
and Colangelo 1985).
QJ
.- One recurrent problem with anterior
2
-u walkers is that the child often tends to
5
\
lean forward while pushing the walker
k (Carmick 1985, Wilson 1988), and the
23
.
entire body is held in a forward, flexed
position. This forward flexion can be
tendency for the child to lose his balance
and fall.
Recently walkers have been introduced
which the child pulls along behind himself
during ambulation i.e. ‘posterior’ walker
(Fig. 2). These walkers have a horizontal
bar at the rear and two horizontal bars
which extend forward and serve as hand-
grips. Preliminary clinical observations
have suggested that children have a more
erect, safer gait using a posterior walker.
Some investigators have also postulated
that the horizontal bar at the back pro-
vides a facilitatory effect by striking the
hip extensors while ambulating (Carmick
1985, Wilson 1988). However, there is
little documented evidence that either type
of walker offers any advantage in pro-
viding improved posture, normalized gait
pattern, or a safer means of ambulation.
The purpose of this investigation was to
compare specific gait characteristics of
children with developmental disabilities,
while they ambulated with anterior and
posterior walkers. In particular, we
were interested in determining whether
posterior walkers were associated with
improved posture and a more typical gait
pattern in these children.

z quite pronounced, and excessive flexion Method

4 of the trunk, hips and other joints Eight children (two females, six males)
compromises the integrity and alignment participated in the study, ranging in age
1044 of these joints. There is also an increased from 2.5 to 8.0 years. Seven had

Fig. 1. Ambulating with anterior walker.
moderate to severe cerebral palsy and
one had Down syndrome. All usually
ambulated with a walker, and none had
undergone surgical intervention.
High-speed cinematography was used
to analyze the children's gait when using
both anterior and posterior walkers. They
were filmed in a lateral (sagittal plane)
view as they ambulated along a 10-
foot walkway. A tripod-mounted 16mm
camera with a film speed of 60 frames/
second was used (Sutherland and Hage
1972, Skrotzky 1983). The choice of
which walker was used first was
randomized. The anterior walker was the
Pediatric Guardian' (Fig. 1) and the
posterior walker was the Kaye Posture
Control (two-wheeled) walker2 (Fig. 2).
The children wore only undershorts. They
also wore the lower-leg orthotics they
usually used during ambulation, or
walked barefoot if they did not require
bracing. They walked at their usual pace,
and occasionally verbal encouragement
was provided. Several practice trials were
'Guardian Products Inc., Fairfield, New Jersey.
2Kaye Products Inc., Durham, North Carolina.
ri
m
Fig. 2. Ambulating with posterior walker.
given with each type of walker before
filming began. During filming, the
children passed four to six times in front
of the camera with each type of walker,
except for one child who only completed
one pass with each walker.
Films were analyzed using a Vanguard
Motion Analyzer. For each child, one
representative pass using each type of
walker was selected for analysis. For each
pass, distancekime measurements of
stride length, gait velocity and percent of
gait cycle spent in double support were
determined. One gait cycle was also
chosen for detailed analysis. At selected
points of initial contact, mid-stance, pre-
swing and mid-swing in the gait cycle,
trunk, hip and knee flexion were deter-
mined. These respective points in the gait
cycle correspond t o the Rancho Los
Amigos criteria, as described by Hoffer
and Perry (1983) and Norkin and
LeVangie (1 983).
Statistical analysis
Distanceltime measurements during
ambulation with both types of walker
were compared using a paired t test. A 1045
 TABLE 1
Distance and time measurements during ambulation
with anterior and posterior walkers
cj
Anterior Posterior
walker walker
Mean (SEM)* Mean (SEMI

Stride length
(cm) 23.6 (3.4) 33.3 (5.7)**

Gait cycle spent

in double
support (070) 6 5 . 3 (9.6) 4 0 . 0 (4.8)*
Gait velocity
(cm/s) 31.4 (6.4) 4 3 . 9 (18.9)

*standard error of the mean; * * p ~ 0 . 0 5 .

TABLE I 1
Trunk, hip and knee flexion while ambulating with anterior walker (AW) and posterior walker (PW)

Initial contact Mid-stance Pre-swing Mid-swing

AW PW AW PW AW PW AW PW
Mean (SEMI Mean (SEM) Mean (SEMI Mean (SEMI Mean (SEMI Mean (SEM) Mean (SEMI Mean (SEMI

Trunk flexion
(degrees) 28.6 (2.9) 18.4 (2.3)* 31.1 ( 4 . 7 ) 18.5 ( 3 . 3 ) * 31.7 ( 2 . 9 ) 19.2 (2.9)' 28.7 ( 4 . 6 ) 18.5 (2.3)'
Hip flexion
(degrees) 50.2 (6.4) 42.5 (6.0) 37.4 (7.7) 2 3 . 6 (4.3)' 40.6 (5.8) 2 8 . 8 (4.8). 59.7 (6.4) 4 3 . 4 (5.4)*
Knee flexion
(degrees) 44.5 (6.4) 40.8 (7.1) 26.0 (7.1) 22.7 (5.0) 47.7 (6.2) 44.0 (7.6) 67.0 (4.8) 58.6 (8.3)

paired t test was also used to compare
analogous joint angles at each selected
point in the gait cycle while the children
ambulated with each type of walker.
Differences were considered significant at
p < 0.05.
Results
Using the posterior walker, stride length
increased by 41 per cent (p < 0.05) and the
period in the gait cycle spent in double-
limb support decreased by 39 per cent
(p<O.O5) (Table I). Hence children am-
bulating with the posterior walker took
longer strides and spent a relatively
smaller proportion of the gait cycle with
both feet on the ground. Gait velocity
increased by an average of 40 per cent
with the posterior walker, but there
was considerable variability among the
children, so the mean increment in gait
1046 velocity was not significant.
Table I1 lists the angles of trunk, hip
and knee flexion recorded at specific
points of the gait cycle. Trunk flexion was
significantly less (Le. the trunk was held
more erect) throughout the gait cycle
when the children used the posterior
walker, compared with the anterior
walker. Hip flexion was also decreased at
each point in the gait cycle in the posterior
walker, with the difference between the
posterior and anterior walker being
significant at mid-stance, pre-swing and
mid-swing (see Table 11). Knee flexion
was slightly less throughout the gait cycle
using the posterior walker, but the
differences in knee flexion between the
two types of walkers were not significant.
Figure 3 illustrates the effects of using
an anterior versus posterior walker on
postural alignment throughout the gait
cycle. The diagrams are a composite of
the trunk, hip and knee angles listed in
 IC MSt
Fig. 3. Composite stick-figure diagrams of mean trunk, hip and knee flexion of eight children using
anterior walker (A W) and posterior walker (PW). Gait cycle subdivisions are initial contact (IC), mid-
stance (MSt), pre-swing (PSw) and mid-swing (MSw).
Table 11. Ambulation with a posterior
walker was associated with a more
upright, vertical posture throughout the
gait cycle, primarily because of decreased
trunk and hip flexion.
Discussion
In order t o progress to mature and
independent gait, an upright posture
against gravity must be achieved
(Sutherland et al. 1980). The results of
this study indicate that the posterior
walkers influenced both postural align-
ment and gait characteristics of these
neurologically impaired young children.
The decreased time in double support
suggests increased stability, thereby al-
lowing longer swing phase, increased
stride length, and a more relaxed,
controlled gait. The key change toward a
more upright posture allows the center of
mass to move directly over the base of
support, rather than in front of the child,
thus mirroring more normal stance.
Significantly decreased trunk and hip
flexion resulted in more upright posture.
Any significant change in knee flexion
may have been masked by the fixed ankle
orthoses that some of the children wore
(Simon et al. 1978).
PSW MSw
Subjective changes in ambulation were
also noted. There were fewer incidences
of loss of balance or of ‘chasing’ the
walker when ambulating with the pos-
terior walker. The child stayed within the
four legs of the walker (base of support)
during ambulation. Some children seemed
to use the support of their upper rather
than their lower extremities, and the dif-
ferences in their gait were less dramatic.
Using an anterior walker usually requires
raising the arms forward and pushing.
When using a posterior walker, the arms
are down at the side and the walker is
pulled along. It was not determined
whether the posterior horizontal bar
striking the hip extensor muscles caused
an activation of these muscle groups.
Clinical decisions are often based on
subjective analysis. The design of this
study allowed us to investigate objective
kinematic measures of the children using
posterior and anterior walkers. Our ob-
servations suggest that posterior walkers
produce beneficial effects in posture and
gait, and may be the device of choice for
children with cerebral palsy. Further
investigation using EMG and force-plate
data would be helpful in distinguishing
differences in walking abilities. 1047
 Accepted f o r publication 3rd May 1990.
Authors’ Appointments
*Charles Ciccone, P.T., Ph.D., Associate
cj Professor, Department of Physical Therapy, lthaca
College, School of Physical Therapy, Ithaca, New
York 14850.
Lynne Logan, M.A., P.T., Physical Therapist, The
Special Children’s Center, Inc.; Lecturer, Depart-
ment of Physical Therapy, Ithaca College, lthaca,
New York.
Kathleen Byers-Hinkley, M.S., P.T., Physical
Therapist, The Special Children’s Center, Inc.,
Ithaca, New York.
*Correspondence to first author.

SUMMARY
Eight children, who needed walkers to ambulate and who were able to use either anterior or
posterior walkers, were selected for a comparative study of both types of walker. Data for gait
analysis, using high-speed cinematography, were collected to determine stride length, time in double
support and velocity. Trunk, hip and knee angles were also measured during four phases of the gait
cycle. The results demonstrated significant improvements in both postural alignment and gait
characteristics with the posterior walker.

RESUME
Deambulateurs anterieurs ou posterieurs pour les enfants: etude analyfique d e la demarche
Huit enfants ayant besoin d’un deambulateur pour se deplacer et pouvant utiliser des deambulateurs
anterieurs ou posterieurs, ont ete choisis pour une etude comparative des deux types de
deambulateur. Des donnees d’analyse de la demarche, utilisant le cinema a grande vitesse, ont ete
recueillies pour determiner la longueur du pas, le temps de double appui et la vitesse. Les angles du
tronc, des hanches et des genoux ont ete tgalement mesures durant quatre phases du cycle de la
demarche. Les resultats demontrerent des ameliorations significatives a la fois dans l’alignement
postural et les caractkristiques de la demarche avec le deambulateur posterieur.

ZUSAMMENFASSUNG
Anteriore und posteriore Gehhilfen f u r Kinder: eine Ganganalyse
Acht Kinder, die Gehhilfen brauchten und die sowohl anteriore oder posteriore Gehhilfen benutzen
konnten, wurden fur eine Vergleichsstudie der beiden Gehhilfearten ausgesucht. Mit der
Hochgeschwindigkeits-Cinematographiewurden Daten fur die Ganganalyse gesammelt, um die
Schrittlange und die Zeit fur Abstutzung und Geschwindigkeit zu bestimmen. Aunerdem wurden
Rumpf-Hiift- und Kniewinkel in vier Schrittphasen gemessen. Die Ergebnisse zeigten bei der
Benutzung der posterioren Gehhilfen eine signifikante Besserung der Korperhaltung, sowie der
Gangcharakteristika.

RESUMEN
Andadores anteriores frente a posteriores en niiios: estudio analitico de la marcha
Ocho niilos que necesitaban andadores para ambular y eran capaces de usar igualmente andadores
anteriores o posteriores fueron seleccionados para un estudio comparativo entre ambos grupos de
andadores. Se recogieron datos para el analisis de la marcha con cinema de aka velocidad para
determinar la longitud de la zancada, el tiempo en soporte doble y la velocidad. Tambien se
midieron el tronco, cadera, rodilla y tobillos en las cuatro fases del ciclo de la marcha. Los
resultados mostraron mejorias significativas tanto en el alineamiento, como en las caracteristicas de
la marcha usando un andador posterior.

References Ankle, 4, 128-134.

Bergen, A., Colangelo, C. (1985) Positioning of the
Client With Central Nervous System Deficits, 2nd
Edn. Valhalla, NY: Valhalla Rehabilitation.
Bleck, E. E. (1987) Orthopaedic Management in
Cerebral Palsy. Clinics in Developmental Medicine,
Nos. 99-100. London: Mac Keith Press with
Blackwell Scientific; Philadelphia: Lippincott.
Carmick, J. (1985) ‘Alternative standing and
ambulating support.’ Totline, 2, 20.
Click, M., Davies, J. (1985) Moving Right Along.
Developmental Goals f o r Physically Disabled
Children. Mesa, AZ: Edcorp Publications.
Cusick, B. D. (1980) ‘Developmental programs for
children in below-knee casts.’ In Wilson, J. M.
(Ed.) Orthopedic Aspects of Developmental
Disabilities. Chapel Hill, NC: University of North
Carolina at Chapel Hill.
Finnie, N. R. (1915) Handling the Young Cerebral
Palsied Child at Home. New York: Dutton.
Hoffer, M . M., Perry, J. (1983) ‘Pathodynamics of
gait alterations in cerebral palsy and the
significance of kinetic electromyography in
I048 evaluating foot and ankle problems.’ Foot and
Licht, E. (1969) Therapeutic Exercise. Baltimore:
W averly .
Norkin, C., LeVangie, P. (1983) ‘Gait.’ In Joint
Structure and Function: a Comprehensive
Analysis. Philadelphia: Davis.
Simon, S. R., Deutsch, S. D., Nuzzo, R. M.,
Mansour, M. J., Jackson, J. L., Koskinen, M.,
Rosenthal, R. K. (1978) ‘Genu recurvatum in
spastic cerebral palsy.’ Journal of Bone and Joint
Surgery, 60A, 882-894.
Skrotzky, K. (1983) ‘Gait analysis in cerebral palsied
and nonhandicapped children.’ Archives of Physi-
cal Medicine and Rehabilitation, 64, 291-295.
Sutherland, D. H., Hage, J. L. (1972) ‘Measure-
ments of gait movements from motion picture
film.’ Journal of Bone and Joint Surgery, 54A,
181-797.
- Olshen, R., Cooper, L., Woo, S. L.-Y. (1980)
‘The development of mature gait.’ Journal of
Bone and Joint Surgery, 62A, 336-353.
Wilson, J. (1988) ‘Selecting and using qosture walkers
for children with cerebral palsy. Totline, 14,
15-16.