A Physiologic Rationale

for Orthotic Prescription

in Paraplegia
by Robert L. Waters, M.D.
Leslie Miller, R.P.T.

INTRODUCTION
A difficult clinical decision to be made when
treating a paraplegic patient is deciding if
walking is a realistic goal, if orthoses should be
prescribed, and what the functional outcome
will be. It has been demonstrated that the en
ergy expenditure for paraplegics, utilizing a
crutch assisted swing-through gait pattern, is
markedly elevated. Many patients have learned
to walk with crutches and orthoses, but discon
tinued their use after discharge from a rehabili
tation center. , , Studies of other forms of
bracing also reveal elevated energy expendi
ture.
In this review, we will describe the indica
tions for prescribing ankle-foot orthoses and
knee-ankle-foot orthoses. We will then outline
the criteria used at Rancho Los Amigos Med
ical Center to determine whether or not a para
plegic is a candidate for ambulation. These cri
teria are based on the results of energy expen
diture measurements of 150 patients with
traumatic paraplegia. Further investigation of
the data collected revealed that the propriocep
tion level or pattern seemed a reliable indicator
of which patients would achieve ambulation,
while muscle function seemed to determine the
quality of their ambulation. These results have
helped us to develop guidelines for projecting
the functional outcome of ambulation of para
plegics.
2

12

10

ORTHOTIC PRESCRIPTION
The goal of orthotic management in para
plegia is to provide the external support neces

sary to compensate for the motor and sensory

deficits. Joint range of motion, muscle
strength, proprioception, sensation, and spas
ticity are evaluated when determining the or
thotic prescription.

Knee-Ankle-Foot
Orthosis
(KAFO)
Quadriceps strength less than " F a i r + " on
manual muscle testing is the most common in
dication for a KAFO. The KAFO is locked at
the knee while walking. Although some pa
tients with less than "Fair+ " strength are able
to ambulate a short distance without a locked
knee (knee stabilization), knee instability
usually occurs after a few steps. The exception
is the patient with severe quadriceps spasticity
which maintains the knee in extension, elimi
nating the need for external support.
Another indication for a KAFO is impaired
or absent knee proprioception. The lack of pro
prioception can result in knee instability even
when the quadriceps strength is " F a i r + " or
greater, as the patient is unable to monitor joint
position. If light touch sensation is present on
the front of the thigh, a KAFO which allows
knee flexion is usually sufficient to control the
knee. The anterior stop of the knee mechanism
limits extension at 180 degrees and the patient
feels pressure from the anterior thigh cuff. In
this regard, the brace serves as a transducer that
converts proprioception (which is not per
ceived) into pressure (which is perceived).
The final indication for extending bracing
above the knee is a severe hyperextension
thrust during stance. Paraplegics whose gait is

characterized by a hyperextension thrust may

develop ligamentous instability, due to atten
uation of the posterior cruciate ligament and
posterior knee capsule resulting in hyperexten
sion deformity.
Range of motion at the hip from 0 degrees of
extension to 110 degrees of flexion should be
present. In the absence of hip extensor
muscles, full hip extension range is necessary
to allow the patient to lean backwards and
move the center of gravity of the trunk poste
rior to the hip joint (Figure 1). Hip flexion to
110 degrees, with the knee extended, enables
the patient to come to standing with locked
KAFO's and rise from the ground after a fall.
Full knee extension is required for optimal fit.
Ten degrees of dorsiflexion at the ankle is
the minimum necessary for unassisted upright
balance (Figure 1). Normal proprioception in at
least one hip also facilitates unassisted
standing.
Inability to meet the joint range requirements
described above commonly occurs and is most
often due to spasticity or contracture. If satis
factory orthotic fit and posture cannot be
achieved, a physical therapy regime that in
cludes stretching exercises or serial casting is
often successful when spasticity is mild and the
deformity is not longstanding. When severe
spasticity or deformity is present, or the de
formity has been present for an extended time,
the patient should be referred to an orthopedic
surgeon.
Good trunk strength is necessary to maintain
an erect posture in the standing position
without excessive weight bearing in the arms.
High level paraplegics without adequate trunk
strength must exert a strong upwards force by
the arms throughout the entire gait cycle to pre
vent forward collapse and accomplish forward
progression. This contributes to the high energy
demand. (All swing-through gait candidates are
required to perform 50 consecutive dips on par
allel bars to insure they have sufficient upper
extremity strength and endurance.)
Ankle-Foot

Orthosis

(AFO)

Quadriceps strength greater than "Fair"

should be present to stabilize the knee if an
AFO is prescribed. The patient must also have
adequate hip flexion strength to swing the leg

Figure 1. Standing posture.

forward to achieve a reciprocal gait pattern.

The indications for AFO are numerous and in
clude any or all of the following: plantarflexion
strength less than " G o o d , " dorsiflexion
strength less than "Fair," impaired ankle pro
prioception, and moderate to severe plantarflexion spasticity.
During normal walking, the plantarflexors
are active during the stance phase of gait to
prevent excessive forward advancement of the
tibia. As a result of forward momentum, the
knee passively extends as the body advances
forward over the stabilized tibia, and the de
mand placed on the quadriceps is minimal.
Customary manual muscle testing methods fail
to place a sufficient load on the plantarflexors
to evaluate the force required during gait. The
strength required to provide ankle and knee sta
bility is present in patients who can perform 15
to 20 toe raises on one leg. Failure to provide
adequate orthotic stabilization of the ankle in
patients with inadequate plantarflexion strength
may result in ankle instability and knee insta
bility, if the quadriceps and/or hip extension
strength is also inadequate.

Knee wobble can be a sign of impaired ankle

proprioception and/or weakness. This can be
eliminated by an AFO with a rigid ankle or an
terior ankle stop, which provides distal stability
and kinesthetic information via the calf cuff.
An AFO may be utilized to hold the ankle in
the neutral position when dorsiflexion strength
is impaired or there is excessive plantarflexion
spasticity. When spasticity is severe, it may not
be possible to maintain the foot in neutral, and
the patient should be referred to an orthopedic
surgeon if non-operative measures prove inade
quate.
When the ankle is held in a rigid orthosis,
ankle stability is gained during midstance.
However, a forward thrust is imposed, forcing
the knee into flexion at the moment of heel
contact. (This knee flexion torque is generated
because the rigidly immobilized ankle rotates
forward about the point of heel contact.)
During normal gait, this torque is avoided by
ankle plantarflexion, minimizing the effect of
the heel lever.
There are two courses of action available to
provide ankle stability during stance, while
still maintaining knee stability at heel strike. If
the patient has "Fair+ " or better ankle dorsi
flexion strength and intact proprioception, we
fit a metal AFO with a double adjustable ankle
joint. A set screw in the anterior channel pro
vides an adjustable stop that prevents excessive
dorsiflexion. The posterior stop is left open to
allow free ankle plantarflexion. Springs can be
added posteriorly if dorsiflexion strength is less
than " F a i r + . " The advantage gained is that
restriction of motion during terminal stance is
maintained while the normal plantarflexion
motion during heel contact is preserved,
avoiding the undesired knee flexion torque. If
the patient has less than "Fair" dorsiflexors or
absent proprioception at the ankle, then the
ankle is locked and either metal or plastic is
used. To avoid the excessive knee flexion
torque when the AFO is locked, the heel of the
shoe is undercut. This decreases the heel lever
and, thus, the knee flexion torque.

ORTHOSIS WEIGHT
Weight is an important factor to some pa
tients, as is the availability of joint motion of
the orthotic system. Plastic, because of its po

tential to be lighter than metal, is sometimes

preferable. For the patient with weak hip
flexors, efforts to minimize weight are war
ranted since any extra weight at the end of the
limb will make it more difficult to lift the foot
and advance the leg. Lehneis, et al. found that
improving orthotic stability at the ankle reduces
energy costs. It follows, then, that in any or
thotic design, stability (control about a joint)
should not be sacrificed merely to achieve
lighter weight.
8

EXERCISE PHYSIOLOGY
It is necessary to understand several funda
mental principles of exercise physiology to in
terpret the results of energy expenditure mea
surements in paraplegic patients. The use of
oxygen consumption is based on the fact that
during sustained exercise, most of the ATP for
muscle contraction is generated by aerobic met
abolic pathways. After several minutes of exer
cising at a constant submaximal workload, the
rate of oxygen consumption rises until it
reaches a level sufficient to meet the metabolic
demands of the exercising muscle. Measure
ment of the rate of oxygen consumption at this
time reflects the energy cost of the activity and
indicates the exercise intensity. The oxygen
cost per meter walked determines the efficiency
of ambulation.
The principle fuels for aerobic metabolism
are carbohydrates and fats. The oxidation of
glucose can be summarized by the following
equation:
1

GLUCOSE + 36 ADP + 6 O ->

6 CO + 44 H O + 36 ATP
2

During exercise, the extent to which anaerobic

pathways contribute to the production of energy
depends upon the intensity of the effort. In mild
to moderate exercise (approximately 50 percent
of the maximal aerobic capacity for untrained
individuals), the oxygen supplied to the tissue
for the aerobic energy producing reactions is
usually sufficient to meet energy requirements.
During more strenuous exercise, anaerobic oxi
dation processes also occurs.
The amount of energy that can be produced
by anaerobic means is limited. Nineteen times
more energy is produced by the aerobic oxida
tion than by anaerobic oxidation. Also, accu
mulation of lactate in muscle and blood leads to

acidosis, limiting the extent to which intense

exercise can be performed. From a practical
standpoint, anaerobic oxidation provides an
extra supply of energy for sudden bursts of
strenuous effort, but these pathways cannot be
routinely utilized for a prolonged time. In con
trast, when exercise is performed below anaer
obic threshold, an individual can sustain exer
cise for many hours without exhaustion.

MAXIMAL
AEROBIC CAPACITY
2

In able bodied subjects, the VO max also

depends on the type of exercise. During lower
limb exercise, the VO max is greater than the
VO max for the upper limbs. Since paraplegic
patients rely on the upper extremities to walk
with the aid of crutches, their energy produc
tion capability is inherently limited. The
problem in paraplegics is further compounded
by the effects of the spinal injury. The upper
2

6,11

ENERGY EXPENDITURE

The maximal aerobic capacity (VO max) is

the single best indicator of physical work ca
pacity and fitness. It measures the individual's
maximum energy production capability. Gener
ally, an individual is able to reach the VO
maximum within two to three minutes of insti
tuting strenuous exercise. Any disorder of the
respiratory-cardiovascular muscle or metabolic
systems that restricts the supply of oxygen to
the muscle decreases the VO max. A physical
conditioning program can increase aerobic ca
pacity by several processes which include im
proving cardiac output, increasing the capacity
of the muscle to extract oxygen from the blood,
increasing the level of hemoglobin, and in
creasing the muscle mass. On the other hand,
the maximal aerobic capacity can be reduced
due to blood loss, paralysis, surgery, negative
nitrogen balance, or bed rest. The important
clinical implication is that the paraplegic pa
tient is usually severely deconditioned as a con
sequence of the injury. The prescription of
orthoses and a walking program should not be
initiated until the patient has sufficient strength
and maximal aerobic capacity to meet the re
quired energy demand. The deconditioned
paraplegic patient will respond to a physical
conditioning program just as an able bodied
subject with respect to increased strength, en
durance, and maximal aerobic capacity.

extremity VO max for paraplegics is lower

than for able bodied subjects, presumably due
to the effects of paralysis and interruption of
the autonomic neurological pathways which
regulate blood flow and cause venous pooling
in the lower extremities. For the typical
adult male paraplegic, we establish a VO max
of 20 ml/kg-min during upper arm cranking as
the minimal criteria acceptable for entering gait
training if a swing-through crutch assisted gait
pattern will be required.

Wheeling

Versus Normal

Walking

On a hard, level surface paraplegic wheel

chair use is as efficient as normal walking. A
comparison of the data in Figure 2 indicates
that when propelling a chair around a 60.5
meter circular track, the speed was almost as
fast as normal walking (72 versus 80 m/min).
The oxygen rate was approximately the same
(11.5 versus 11.9 ml/kg/min) (Figure 3), as
was the oxygen cost (.16 versus .15 ml/kg/
min). The heart rate was higher in paraplegics
using the wheelchair than in normal walking
(123 versus 100 BPM) (Figure 4). As pre
viously mentioned, this relates to the lower
upper maximal aerobic capacity in paraplegics
during arm exercise. From a clinical stand
point, it may be concluded that the wheelchair
is a highly efficient means of transportation
whose speed and energy requirements are com
parable to that of normal walking.
10

Swing Through

Gait

Crutch walking with a swing-through gait re

quires the arms and shoulder girdle to lift the
entire weight of the body and swing it forward
with each step. The average speed in para
plegics trained to use a swing-through crutch
assisted gait was 64 percent lower than normal
walking (20 versus 80 m/min) (Figure 2); the
rate of oxygen consumption was 38 percent
greater (16.5 versus 11.9 ml/kg/min) (Figure
3); the oxygen cost was 560 percent greater
(.84 versus .15 ml/kg/min); and the heart rate
was increased 46 percent (145 versus 99 BPM)
(Figure 4 ) . This rate of energy expenditure
requires most of the aerobic capacity of the
typical adult male paraplegic with a complete
T12 lesion and is well above the anaerobic
10

F i g u r e 2. A v e r a g e v e l o c i t y i n n o r m a l s u b j e c t s a n d in p a t i e n t s u s i n g w h e e l c h a i r s o r o r t h o s e s .

F i g u r e 3 . R a t e o f o x y g e n c o n s u m p t i o n in n o r m a l s u b j e c t s a n d i n p a t i e n t s u s i n g w h e e l
chairs or orthoses.

F i g u r e 4. H e a r t r a t e in n o r m a l s u b j e c t s a n d in p a t i e n t s u s i n g w h e e l c h a i r s o r o r t h o s e s .

threshold. The extreme exertion required for a

swing-through gait demands a greater intensity
of physical effort than a normal individual cus
tomarily expends on sports activity such as rec
reational jogging. Consequently, it is not sur
prising that while the athletic paraplegic may
be willing to expend this level of exertion for
recreational purposes, he is unwilling to sustain
these efforts for normal activities of daily
living. Even those patients, who are physiolog
ically capable of sustaining the intense physical
effort of a swing-through gait for a sustained
time period to travel longer distances, find
tachypnea (rapid breathing), tachycardia (rapid
heart rate), and hidrosis (sweating), unaccept
able for routine activities of daily living.
We believe that the highly motivated para

plegic who is willing to exercise strenuously

should not be discouraged from walking, but a
more realistic approach should be taken for the
average patient. The average patient should be
given walking training and bilateral kneeankle-foot orthoses only if walking is necessary
for psychological reasons, for purposes of ex
ercise, or because of architectural barriers in
the living environment. It should be clearly ex
plained that the wheelchair should be consid
ered as the primary means of mobility.
We have tested three patients with " F a i r + "
hip flexors who used bilateral KAFO's and pre
ferred a reciprocal gait pattern. Interestingly,
the effort expended by these patients was just
as great as in swing-through gait (Figures 2, 3,
and 4).
10

F i g u r e 5 . A v e r a g e d i s t a n c e s n e c e s s a r y t o p e r f o r m c u s t o m a r y a c t i v i t i e s of d a i l y
living.

Energy Expenditure:

Reciprocal

Gait

In a review of spinal cord injured patients,

Hussey and Stauffer found that those patients
who were able to walk in the community had
pelvic control with at least "Fair" hip flexor
strength and at least "Fair" extensor strength
in one knee so that a maximum of one KAFO
was required, enabling the patient to achieve a
reciprocal gait pattern. Having " F a i r + " or
greater quadriceps strength sufficient to stabi
lize one knee eliminates the need for one
KAFO and enables the patient to walk with a
crutch assisted reciprocal gait pattern at a rate
of energy expenditure and heart rate that are
significantly below that required for a swingthrough gait pattern (Figures 3 and 4). Surpris
ingly, we found no difference in the speed and
rate of energy expenditure in patients with one
free knee or two free knees and requiring
bracing only below the knee (Figures 2 and 3).
Nevertheless, paraplegics who have intact
hip flexion and knee extension bilaterally re
quire orthoses only below the knees, and those
who use a reciprocal crutch assisted gait pattern
6

are still severely impaired (Figures 2, 3, and 4).

Compared to normal walking, the rate of oxy
gen expenditure is 20 percent greater (16.3
versus 11.9 ml/kg/min) (Figure 3), the heart
rate 31 percent greater (131 versus 100 BPM)
(Figure 4), and the speed 67 percent slower (80
versus 20 m/min) (Figure 2). The typical para
plegic who uses crutches and a reciprocal gait
still exerts a force of 25 to 50 percent of total
body weight on the crutches with each step, ac
counting for the increased rate of energy ex
penditure. The only spinal cord injured patients
we have tested whose energy expenditure
during walking does not exceed normal values
are those patients with minimal involvement
who have intact sacral function (in addition to
lumbar function) and a sufficient hip abductor
and extensor strength to maintain an erect pos
ture without crutches.
The average distances necessary to perform
different daily living activities are listed in
Figure 5 and were obtained from numerous
measurements made in different types of urban
areas in Los Angeles. Since the average speed
of walking in low lumbar paraplegics who
10

used bilateral ankle-foot orthoses and a recip

rocal crutch assisted gait pattern was only 26
m/min, it would take more than five minutes to
travel 150 meters. Because five minutes of
walking will require a strenuous effort, it is ap
parent why even the typical low lumbar para
plegic is a limited walker outside the home and
is not able to routinely ambulate comfortably
for activities which require walking a longer
distance. In this regard, clinicians are justified
in prescribing a wheelchair to any spinal injury
patient who requires crutch assistance. The pa
tients should be encouraged to use the wheel
chair as necessary and be reassured that reli
ance on the wheelchair, when necessary,
should not be considered a failure.
AUTHORS
Robert L. Waters, M.D. is Chairman of the Department
of Surgery at Rancho Los Amigos Medical Center,
HB-121, 7601 E. Imperial Highway, Downey, California
90242.
Leslie Miller, R.P.T. is a spinal cord injury clinical spe
cialist at Rancho Los Amigos Medical Center.

Cerny, K., R. Waters, H. Hislop and J. Perry,

"Walking and Wheelchair Energetics in Persons with Para
plegia," Physical Therapy, 60:1133-1139, 1980.
Clinkingbeard, J.R., J.W. Gersten, and D. Hoehn,
"Energy Cost of Ambulation in the Traumatic Para
plegic," American Journal of Physical Medicine, 43:157165, 1964.
Gordon, E.E., "Physiological Approach to Ambula
tion in Paraplegia," JAMA, 161:686-688, 1956.
Huang, C.T., A.B. McEachran, K.V. Kuhlemeier,
M.J. DeVivo, and P.R. Fine, "Prescriptive Arm Ergometry to Optimize Muscular Endurance in Acutely Injured
Paraplegic Patients," Arch. Phys. Med. Rehab., 64:578582, 1983.
Hussey, R.W. and E.S. Stauffer, "Spinal Cord Injury:
Requirements for Ambulation," Arch. Phys. Med. Rehab.,
3

5 4 : 5 4 4 - 5 4 7 , 1973.
7

Lehneis, H.R., E. Bergofsky, and W. Fresina, "En

ergy Expenditure with Advanced Lower Limb Orthoses and
with Conventional Braces," Arch. Phys. Med. Rehab.,
57:20, 1976.
Lerner-Frankiel, M.B., S. Vargas, M. Brown, L. Krusell, and W. Schoneberger, "Functional Community Am
bulation: What Are Your Criteria?" Clin. Man. in Phys.
Ther., 6:12-15, 1986.
Waters, R.L., H.J. Hislop, J. Perry, and D. Antonelli,
"Energetics: Application to the Study and Management of
Locomotor Disabilities," Orthop. Clin. North America,
9:351-377, 1978.
Waters, R.L. and B.R. Lunsford, "The Energy Cost
of Paraplegic Locomotion," Journal of Bone Joint Sur
gery, 67A: 1245-1250, 1985.
Wolf, E. and A. Magora, "Orthostatic and Ergometric Evaluation of Cord-injured Patients," Scandinavian
Journal of Rehabilitation Medicine, 8:93-96, 1976.
8

10

REFERENCES
1

Aastrand, P.O. and K. Rodahl, Textbook Work Physi

ology, Ed. 2, McGraw-Hill, Inc., New York, 1977.

11