JRRD

Volume 41, Number 5, Pages 707–712
September/October 2004

Journal of Rehabilitation Research & Development

A new hip-knee-ankle-foot sling: Kinematic comparison with a traditional
ankle-foot orthosis
Arturo Zancan, MD; Maria V. Beretta; Micaela Schmid, PhD; Marco Schieppati, MD
Division of Rehabilitation, Salvatore Maugeri Foundation, Work and Rehabilitation, Institute of Care and Scientific
Research, Pavia, Italy; Human Movement Laboratory (CSAM), Salvatore Maugeri Foundation, Work and Rehabilitation, Institute of Care and Scientific Research, Pavia, Italy; Human Physiology Section, Department of Experimental
Medicine, University of Pavia, Pavia, Italy

AFOs are mainly used to counteract the foot plantar flexion, they are also reported to enhance the paretic quadriceps muscle activity [6] and lower-limb swing in
hemiplegic subjects [9].
This study developed a device for the lower limb that
could correct the abnormal foot posture, while simultaneously facilitating the swing phase of the lower limb. In
our opinion, this could be done by “linking” the patient’s
affected lower limb and the opposite body side. Following this idea, we developed a strap-based linkage system,
which is a “sling-like” product. We then performed a
kinematic analysis of the new, simple, low-cost sling in
one representative hemiplegic patient and compared the
analysis of the sling to a commonly prescribed AFO.

Abstract—In this study, we performed a kinematic analysis of
a new, low-cost sling for the lower limb, compared to a common ankle-foot orthosis (AFO). Gait with no orthosis, with the
AFO, and with the new sling was analyzed in one hemiplegic
subject. Both the AFO and the sling reduced the mean angle
and ROM (range of movement) of the ankle and the vertical
displacement of the center of mass. The sling, but not the AFO,
restored the normal sequence heel-strike, forefoot contact of
the affected side. The sling, but not the AFO, reduced the
affected limb stance and stride duration, increased stride
length, and improved walking speed. In conclusion, the proposed sling for the lower limb equally improved the affected
ankle kinematics in contrast to the traditional AFO, and it also
improved some gait variables in this hemiplegic subject.

Key words: AFO, gait, hemiplegia, kinematic, lower limb,
orthosis, sling.
Abbreviations: AFO = ankle-foot orthosis, ANOVA = analysis of variance, ASIS = anterior superior iliac spine, COM =
center of mass, KAFO = knee-ankle-foot orthosis, LSD = least
significant difference, RGO = reciprocating gait orthosis,
ROM = range of movement, SD = standard deviation.
The work published in this manuscript was unfunded at
the time of manuscript preparation.
Address all correspondence to Dr. Arturo Zancan, Division of
Rehabilitation, Salvatore Maugeri Foundation, Work and
Rehabilitation, Institute of Care and Scientific Research, via
Ferrata 8, 27100 Pavia, Italy; 39-038-259-2626; fax: 39-038259-2085; email: azancan@fsm.it.
DOI: 10.1682/JRRD.2003.06.0101

INTRODUCTION
Hemiplegic patients always show alterations of walking pattern kinematics [1,2] and slow walking speed,
even if the gait is tripod- or cane-assisted [3]. Ankle-foot
orthoses (AFO) are often prescribed to those patients to
obtain a better heel-strike, a more dynamic and balanced
gait, and a lower energy expenditure during walking [4–
7]. Knee-ankle-foot orthoses (KAFO) are rarely used in
hemiplegic patients, because of their weight and their
supposed disturbance of walking patterns [8]. While
707

708
JRRD, Volume 41, Number 5, 2004

METHODS
A single-case study design was chosen, avoiding
intersubject variability and confounding factors. One
subject was evaluated: a 74-year-old male who suffered
from ischemic stroke and subsequent right-side hemiplegia without aphasia, apraxia, or spatial neglect. The
ischemic lesion was located in the left-middle cerebral
artery territory, and it was documented by a computed
tomography (CT) scan. Walking analysis was performed
30 days after stroke and 3 weeks after the beginning of
the rehabilitation program, which was performed according to Bobath principles.
At the moment of the walking tests, the subject presented an upper-limb recovery classified as Phase 1,
according to Brunnstrom [10]:
1. A spasticity of affected triceps surae muscle was
present (Ashworth score = 2) with no equinus
deformity.
2. No limitation in the range of motion (ROM) of the
lower-limb joints was present.
3. The lower limb presented flexion and extension synergies without capability of isolated movements.
4. The walking pattern was classified as Phase 2-A,
according to Winters et al. [1,11].
The custom-made sling consisted of inextensible textile stripes, as shown in Figure 1(a) and 1(b):
1. A belt with a stripe across the shoulder.
2. A connecting stripe lying on the anterior face of the
lower limb, consisting of (Figure 1(a)) an upper segment, connected to the belt in correspondence to the
anterior superior iliac spine (ASIS), and a lower segment, which is connected to the upper segment by
means of a buckle and fixed by little straps around
the knee and around the metatarsal heads.
All the parts of the sling are buckle-adjustable to fit
patient’s body size. The AFO, also custom-made, was a
commonly prescribed type (posterior leaf spring,
(Figure 1(c)), which was already in use at the time of the
rehabilitative treatment. Like the AFO, the sling counteracts the plantar flexion, while allowing the “normal” foot
dorsal flexion during gait. The sling was used by the
patient for the first time during this test. The patient was
not able to put on the AFO or the sling without help. The
walking test was performed either without the orthosis,
with the AFO, and with the sling. The sling was set on

Figure 1.
Hip-knee-ankle-foot sling, (a) front and (b) rear views, respectively,
and (c) ankle-foot orthosis (AFO).

the patient in an upright position, with the ankle at an
approximately 90° angle.
Experimental Procedure
In a single session, three conditions were tested (use
of no orthosis, use of AFO, and use of sling), with a total
of six walking tests and 2 min intervals between tests.
Every test consisted of the patient walking alone for a
distance of 6 m, with the paretic upper limb supported
and a tripod on the unaffected side. Two tests at the same
walking condition were necessary to collect a sufficient
number of steps for the analysis. The first two consecutive tests were performed with no orthosis, two consecutive tests with the spring AFO, and the last two
consecutive tests with the sling. The subject was placed
at the beginning of the 6 m line, previously traced on the
laboratory floor. At the operator’s signal, the patient
started walking at his own spontaneous speed, and data
recording started.
Data Acquisition and Analysis
The walking tests were recorded by the ELITE
movement analysis system (BTS, Italy). This system

709
ZANCAN et al. New hip-knee-ankle-foot sling

consisted of six infrared cameras (100 Hz frequency)
positioned at the vertexes of an hexagonal space surrounding the line used for the walking tests. Infraredreflecting markers were positioned on the following
points of body sides: acromion, ASIS, trochanter major,
head of fibula, lateral malleolus, fifth metatarsal head.
The last marker was attached to the shoe, corresponding
to the fifth metatarsal head. The patient wore his shoes
under all conditions tested. The position of the markers
was acquired throughout the 6 m walking test. Data were
recorded on electronic media and later analyzed with the
use of the standard ELITE system software. The following variables were recorded and measured:
1. Hip, knee, and ankle mean angle and ROM.
2. Frontal and horizontal displacement of the mean
point between the ASIS markers approximately corresponding to the subject’s center of mass (COM).
3. Temporal sequence of the heel-strike forefoot contact
and heel-off-forefoot-off, calculated on the basis of the
time course of the position of the markers on the malleolus and fifth metatarsal bone.
4. Duration of stance phase.
5. Stride length.
6. Stride duration, calculated for both limbs from the
temporal succession of heel strikes.
7. Walking speed.
Statistical Analysis
Mean values and standard deviations (SDs) of the
variables were recorded during the patient’s walking with
no orthosis, with the spring AFO, and with the new sling.
The statistics were compared with the use of a one-way
analysis of variance (ANOVA) and a posthoc NeumannKeuls Multiple Comparison test, with p < 0.05 considered for significance. To evaluate the effect of the devices
on the heel-strike-forefoot-contact and heel-off-forefootoff intervals, on both the affected and the unaffected side,
we used a two-way ANOVA (no orthosis, AFO, sling on
the affected and unaffected side), followed by Fisher’s
Least Significant Difference (LSD) tests. Fisher’s LSD
tests were chosen because of the small size of samples; p
< 0.05 was considered significant.

RESULTS
We evaluated a total of 12 steps with no orthosis, 10
steps with the AFO and 10 steps with the sling. The hip,

knee, and ankle mean angles showed only little differences
across no orthosis, AFO, and sling conditions. The ankle
ROM with the sling showed a significant decrease compared to the use of no orthosis (–30.5%, p < 0.001). It also
showed a considerable decrease with the AFO compared
to the use of no orthosis (–31.6%, p < 0.001), while there
was no statistical difference between ankle ROM with the
AFO and with the sling (Figure 2(a)). The COM vertical
displacement showed a sizeable decrease (–34.2%, p <
0.001) with the sling compared to the use of no orthosis.
The AFO also had a substantial decrease compared to no
orthosis (–33.2%, p < 0.001), and no significant difference
existed between COM vertical displacement with the AFO
and the sling (Figure 2(b)).
No noteworthy difference was found in the subject’s
COM mediolateral displacement with either the AFO,
sling, or no orthosis. During free walking on the affected
side, the normal sequence heel-strike forefoot contact was
inverted (forefoot landing before heel-strike), and a negative value resulted from the difference between the instants
of heel-strike and forefoot contact (Figure 2(c)). This
abnormal value turned to normal (heel-strike before forefoot contact) with the sling, compared to no orthosis (p <
0.001), and the difference between the affected and unaffected side turned to a nonsignificant value (p = 0.607,
posthoc test).
With the AFO, the timing difference between heelstrike and forefoot contact sizeably diminished (p <
0.001) and was almost canceled. However, this difference
was still abnormal compared to no orthosis, most likely
because of the large compliance of the AFO material,
which did not fully support the foot weight or oppose the
plantar flexors muscles spasticity [9]. Although much
improved, the sequence was still appreciably different
between the affected and unaffected side (p < 0.05, posthoc test). The improving effects produced by the sling
on the foot contact sequence proved to be almost significantly different (p = 0.05008, posthoc test) from those
obtained with the AFO (Figure 2(c)). The temporal
sequence of the heel-off-forefoot-off did not show any
measurable difference with the AFO, sling, or no orthosis
(Figure 2(d)).
The stance duration on the affected side was noticably
smaller with the sling (–20.8%, p < 0.001) compared to no
orthosis, and the AFO also considerably decreased stance
duration (–22.7%, p < 0.05) compared to no orthosis, but
no difference was found between the AFO and sling (Figure 2(e)). The stride length of the affected side increased
sizeably with the sling (+18.5%, p = 0.024), but not the

710
JRRD, Volume 41, Number 5, 2004

Figure 2.
(a) Mean ± standard deviations (SDs) of hip, knee, and ankle range of motion (ROM). ROM “0” represents no variation of articular angle during
locomotion. Both AFO and sling significantly limited mobility of affected side ankle. (b) Mean ± SDs of vertical displacement of subject’s center
of mass (COM). Both AFO and sling significantly limited elevation of affected pelvis side (* = p < 0.05). (c) Mean ± SDs of heel-strike forefoot
contact temporal sequence. Negative value indicates forefoot landing before heel. (d) Heel-off-forefoot-off temporal sequence: sling induces a
trend toward normalization of sequence. (e) Mean ± SDs of stance duration, calculated for unaffected and affected lower limb: both AFO and
sling orthoses significantly reduced stance duration on affected lower limb (* = p < 0.05), with minor effects on unaffected side. (f) Mean ± SDs
of stride length, calculated for unaffected and affected lower limb: with sling, but not AFO, significantly increased stride length of both affected
and unaffected lower limb (* = p < 0.05). (g) Mean ± SDs of stride duration, calculated for unaffected and affected lower limb: with sling, but not
AFO, significantly diminished stride duration of both affected and unaffected lower limb (* = p < 0.05). (h) Mean walking speed with no orthosis,
AFO, and sling. Walking speed with sling increased 64.7% and 29.4% with AFO.

711
ZANCAN et al. New hip-knee-ankle-foot sling

AFO, compared to no orthosis (Figure 2(f)). The stride
duration of the affected leg was significantly smaller with
the sling (–26%, p = 0.002), but not the AFO, compared to
no orthosis (Figure 2(g)). The mean walking speed
increased with both the sling (+64.7%) and the AFO
(+29.4%), compared to no orthosis (Figure 2(h)).

DISCUSSION
The findings indicate that both types of orthoses produced similar effects on hip, knee, and ankle ROM. Both
the sling and the AFO equally limited the subject’s plantar flexion during walking. The analysis of COM vertical
displacement shows a diminished need for elevation of
the affected pelvis side with both the AFO and the sling.
Despite the similarity in kinematic data for both devices,
the walking speed was substantially higher with the sling.
Therefore, the reduction in the affected ankle ROM alone
does not seem to be sufficient to explain the subject’s
increase in walking speed [12]. An element of explanation could be derived from the change in the interval of
heel-strike forefoot contact. During the subject’s walking
without orthosis, as well as with the AFO, the normal
heel-strike forefoot-contact sequence was inverted (the
affected limb forefoot hit the ground before the heel did).
The AFO significantly improved the timing of the foot
stance phases; in particular, it changed the foot-strike
from a first-forefoot contact to an almost flat-foot contact. Using the sling, the subject resumed a correct and a
close-to-normal pattern sequence interval (the heel-strike
was performed before forefoot contact), with no significant difference compared to subject’s unaffected lower
limb. Even if the mechanism of action is not simple to
elucidate, the “linkage” between normal body side and
the affected lower-limb may be another explanation of
the increase in walking speed. The sling may also
enhance the effects of the inward hip rotation of the unaffected limb. As the affected side ASIS rotates forward
toward the midsagittal plane, the sling tightens and pulls
forward the whole limb through its foot attachment. This
recalls the action of reciprocating gait orthosis (RGO),
with a difference. In this case, the RGO action turns the
extension of the stance leg into the flexion of the swinging leg [13], while pelvic thrust and inward rotation on
the supporting unaffected limb are presumably the main
mechanism of action of the sling.

CONCLUSIONS
The results indicated to us that the proposed sling
improved the affected side’s ankle kinematics equally to
the traditional AFO. In addition, the sling also improved
some variables of walking pattern in this hemiplegic subject, leading to a greater increase of walking speed than
the AFO. These data could contribute to a background for
future studies that would evaluate the possibility of clinical application of the proposed sling.

ACKNOWLEDGMENTS
We thank Dr. Luigi Tesio for his useful comments on
the paper. We would also like to acknowledge Mauro
Serra and Ortopedia Noli Voghera for their assistance in
device manufacturing.

REFERENCES
1. Rodda J, Graham HK. Classification of gait patterns in spastic hemiplegia and spastic diplegia: a basis for a managment
algorithm. Eur J Neurol. 2001;8 Suppl 5:98–108.
2. Roth EJ, Merbitz C, Mroczek K, Dugan SA, Suh WW.
Hemiplegic gait. Relationships between walking speed and
other temporal parameters. Am J Phys Rehabil. 1997;
76(2):128–33.
3. Chen CL, Chen HC, Wong MK, Tang FT, Chen RS. Temporal stride and force analysis of cane-assisted gait in people with hemiplegic stroke. Arch Phys Med Rehabil. 2001;
82(1):43–48.
4. Mauritz KH. Gait training in hemiplegia. Eur J Neurol.
2002;9 Suppl 1:23–29.
5. Hesse S, Luecke D, Jahnke MT, Mauritz KH. Gait function
in spastic hemiparetic patients walking barefoot, with firm
shoes and with ankle-foot orthosis. Int J Rehabil Res. 1996;
19(2):133–41.
6. Hesse S, Werner C, Matthias K, Stephen K, Berteanu M.
Non-velocity related effects of a rigid double-stopped
ankle-foot orthosis on gait and lower limb muscle activity
of hemiparetic subjects with an equinovarus deformity.
Stroke. 1999;30(9):1855–61.
7. Bean J, Walsh A, Frontera W. Brace modification improves
aerobic performance in Charcot-Marie-Tooth disease: a single subject design. Am J Phys Med Rehabil. 2001;80(8):
578–82.
8. Kakurai S, Akai M. Clinical experiences with a convertible
thermoplastic knee-ankle-foot orthosis for post stroke hemiplegic patients. Prosthet Orthot Int. 1996;20(3):191–94.

712
JRRD, Volume 41, Number 5, 2004

9. Gok H, Kucukdeveci A, Altinkaynak H, Yavuzer G, Ergin S.
Effects of ankle-foot orthoses on hemiparetic gait. Clin
Rehabil. 2003;17(2):137–39.
10. Brunnstrom S. Movement therapy in hemiplegia. New
York (NY): Harper & Row; 1970.
11. Winters TF, Gage JR, Hicks R. Gait patterns in spastic
hemiplegia in children and young adults. J Bone Joint Surg.
1987;69:437–41.
12. Hachisuka K, Ogata H, Tajima F, Ohmine S. Clinical evaluations of dorsiflexion assist controlled by spring ankle-foot
orthosis for hemiplegic patients. J UOEH. 1998;20(1):1–9.

13. Dall PM, Muller B, Stallard I, Edwards J, Granat MH. The
functional use of the reciprocal hip mechanism during gait
for paraplegic patients walking in the Louisiana State University reciprocating gait orthosis. Prosthet Orthot Int.
1999;23(2):152–62.

Submitted for publication June 24, 2003. Accepted in
revised form February 2, 2004.