Journal of Applied Biomechanics, 2013, 29, 443-452
© 2013 Human Kinetics, Inc.
An Official Journal of ISB
www.JAB-Journal.com
ORIGINAL RESEARCH
Kinematic and Kinetic Evaluation of the Stance Phase
of Stair Ambulation in Persons with Stroke
and Healthy Adults: A Pilot Study
Alison C. Novak and Brenda Brouwer
Queen’s University, Kingston
This study describes and contrasts the kinematics and kinetics of stair ambulation in people with chronic stroke
and healthy control subjects. Three-dimensional motion data were collected from 10 persons with stroke (7
males) and 10 sex and age-matched older adults as they ascended and descended an instrumented staircase
at self-selected speed with and without a handrail. Ankle, knee and hip joint angle and moment profiles were
generated during stance and range of motion and peak moments were contrasted between groups, sides (stroke
only) and condition. Cadence was lower in stroke than controls, although the kinematic profiles appeared similar
during ascent and decent. Notable differences in joint kinetics were evident as the peak extensor moments were
typically lower on the affected side in stroke compared with controls and the less affected side. These differences accounted for the lower magnitude net extensor support moment. The lower affected side hip abductor
moments likely limited lateral stability. Handrail use tended to reduce the peak moments on the affected side
only leading to more side-to-side differences than occurred without the handrail. The findings reveal differences in task performance between stroke and healthy groups that help inform rehabilitation practice.
Keywords: stair negotiation, mobility, gait, biomechanics
Stair negotiation is a challenging and demanding
locomotor task that is essential for independent ambulation. Biomechanical analyses have shown that compared
with level walking, greater lower limb joint ranges of
motion and muscle moments are generally required to
ascend and descend stairs, with the knee extensors having
a dominant role.1–3 Following stroke, reduced neuromuscular capabilities and limitations in joint mobility can
increase the challenge of stair negotiation which can
present as a barrier to active community living.4
There is ample evidence that reduced joint mobility poses risks for stair negotiation.5 Researchers have
reported that healthy, older adults show reduced ankle
and knee motion compared with young adults when
descending stairs, accompanied by increased hip and
pelvic motion in the frontal plane.6 During stair ascent,
age-related reductions in ankle dorsiflexion could result
in a trip if the toes fail to clear the step.7 Recent evidence
has shown that older adults adopt alternative strategies to
meet the high demands of the task.8–11 A redistribution of
joint moments,8–10 an exaggerated net support moment,
and sustained abductor moments throughout stance are
evident in older adults likely to compensate for declining
Alison C. Novak and Brenda Brouwer (Corresponding Author)
are with the School of Rehabilitation Therapy, Queen’s University, Kingston, ON, Canada.
muscle strength and/or to enhance stability.8 In clinical
populations such as stroke, where impairment is typically
superimposed on aging, the ability to redistribute loads
in a way that accomplishes the task of stair ambulation
may be compromised.
Level ground gait of persons with stroke is characterized by a decrease in self-selected speed and altered
sagittal and frontal plane kinematic and kinetic profiles
in both magnitude and pattern.12–16 These deviations
from the normal gait pattern result from residual deficits
in strength, sensation, poor coordination, loss of balance
control and deconditioning; the nature and extent of the
impairments related to the severity and location of the
stroke (see Novak and Brouwer17 for review). It is the case
that many individuals regain the ability to walk following stroke,18,19 however it is stair negotiation that is the
single best predictor of community living activity.4 The
higher physical demands associated with stair ambulation compared with walking likely explains why stair use
has been rated as the most difficult task following stroke
rehabilitation.20 Like walking, stair negotiation requires
substantial sagittal plane motion as the body advances in
forward and vertical directions, but also demands greater
medio-lateral control for stability;2 the biomechanics of
which have not been well described. Knowledge of lower
limb kinematics and kinetics in sagittal and frontal planes
is essential to characterize the requirements of stair ascent
and descent5 and is important to understand stroke related
443
444 Novak and Brouwer
compensations compared with nondisabled controls. This
information can guide intervention strategies aimed at
improving stair mobility and community access.
This study characterizes the sagittal and frontal plane
lower limb kinematics and kinetics of stair ascent and
descent in people with stroke in comparison with similarly aged healthy controls. We hypothesize that persons
with stroke will demonstrate reduced ankle mobility on
the paretic side associated with altered kinematic patterns
contralaterally that differ from healthy adults. Lower peak
moments on the paretic side will result in asymmetries
and necessitate compensation from other muscle groups
to produce adequate support. We further postulate that
handrail use will be associated with reduced lower limb
joint moments, particularly in the stroke group.
Methods
Subjects
Ten community-dwelling chronic stroke survivors (7
male, 3 female) at least 6 months poststroke (mean =
28.1 ± 16.3 months) were recruited through newspaper
advertisements and from monthly clinics held at the rehabilitation hospital for this pilot study. All reported residual
weakness, with six presenting with right hemiparesis and
four had left hemiparesis; the mean age of the group was
60.1 ± 10.3 years, age- and sex-matched controls (mean =
59.4 ± 8.7 years) were drawn from those individuals who
responded to newspaper advertisements and postings at
the local senior’s center. All subjects self-reported good
health and were able to ascend and descend at least 4
steps with or without the use of a handrail. In addition,
subjects with stroke were independently ambulatory with
or without an assistive device though all presented with
mobility deficits. This was evidenced from their performance on the community balance and mobility scale
(mean score = 53.9 ± 20.3 points; max score = 96), which
relates to motor recovery and lower limb strength.21,22
All individuals were screened to exclude those with
conditions (other than stroke) that affected ambulation
(eg, mobility limiting arthritis) or who were unable to
follow instructions. All procedures were approved by the
university’s research ethics board and subjects provided
their informed consent before participation.
Gait Assessment
Three-dimensional motion data were obtained as subjects
ascended and descended an instrumented staircase consisting of four steps (rise: 15 cm; run: 26 cm; width: 56
cm). Bilateral kinematic data were acquired at a sampling
rate of 50 Hz using two Optotrak 3020 sensors (RMS
accuracy ≤ 0.15 mm at a distance of 2.5m, Northern
Digital Inc, Waterloo, ON) which were positioned within
2 m of either side of the staircase to track the position of
infrared emitting diodes. Clusters of three to four infrared
emitting diodes were mounted noncollinearly in molded
plastic and secured on each segment of the lower body
(midfoot, midshank and midthigh) bilaterally and over
the sacrum at the level of S2 using a fin which projected
outwards. A force platform (AMTI, Newton, MA model
OR6-6) mounted on concrete blocks formed the middle
of the second step and recorded ground reaction forces
at a sampling rate of 100 Hz.
Limb segment lengths, joint centers and rotational
axes were defined using an instrumented probe embedded
with four infrared emitting diodes fixed relative to the
tip.8,23 In addition, bilateral virtual landmarks were identified from a static standing reference trial representing the
approximate locations of the first and fifth metatarsals,
lateral and medial malleoli, lateral and medial epicondyles, greater trochanters, and points aligned vertically
with each greater trochanter at the level of the anterior
superior iliac spine.
Procedure
Once instrumented, subjects performed several practice trials to ensure lead wires were nonrestricting and
adequately secured. Subjects were instructed to ambulate
at a self-selected pace and to place one foot on each step
(step-over-step). Subjects were asked to perform trials and
without the use of a handrail as they were able. Because
of constraints on the experimental setup, the handrail was
always placed opposite the limb contacting the forceplate.
Three successful trials were acquired for each condition
(with handrail, without handrail) for ascent and descent.
Data Processing and Analysis
All kinematic and force platform data were filtered (second-order, low pass, Butterworth, cutoff frequency 6 Hz),
and synchronized using postprocessing software (Visual
3D, C-Motion, Inc., Rockville, MD). Kinematic (joint
angles) and kinetic data (internal joint moments) were
computed using a seven segment, link-segment model
and an inverse dynamics approach. Spatial coordinates
were transformed into Cardan angles by determining the
orientation of the distal segment with respect to the reference proximal segment using the x, y, z ordered sequence
of rotations (representing flexion/extension, abduction/
adduction and axial rotation, respectively). Kinetic data
were normalized to body mass (kg) and all data were
normalized to 100% of the stance phase. Most affected
and less affected sides were analyzed in stroke subjects
and the dominant side only in healthy controls since no
side-to-side differences have been reported in this group.8
Cadence (steps/min) and stance time (s) were computed from the kinematic data and joint ranges of motion
determined. Peak values for joint moments were obtained
from individual curve profiles and where appropriate,
distinct peaks associated with specific phases of stance
were also identified. Outcome measures for individual
trials were averaged for each condition to avoid attenuation due to minor temporal shifts. Kinematic and kinetic
data are reported for the ankle, knee and hip joints in the
sagittal and frontal planes of movement. To illustrate the
temporal aspects of the data, curves were normalized to
Stair Ambulation: Stroke and Healthy Adults 445
100% of the stance phase and averaged for the control
group and the affected and less affected sides in stroke.
Descriptive statistics (means and standard deviations) were calculated for all outcome measures (SPSS
version 19.0, San Rafael, CA). T-tests were used to
analyze between group effects for each condition (with
handrail, without handrail) rather than a factorial analysis because of unequal cell counts (two subjects with
stroke could not manage stairs without handrail use).
Paired t tests were conducted to examine the effect of
side (most affected vs. less affected) in the stroke group.
A significance level of P < .05 was adopted for all
analyses. Trends (P = .05–.10) were explored post hoc to
determine whether the inclusion of more subjects might
have produced a conclusive finding. This was reported
as appropriate.
Results
Cadence was in all cases (ascent, descent and with
handrail, without handrail) lower in stroke subjects than
controls (P < .001). This was accompanied by longer
stance times (most affected and less affected sides) during
ascent (P < .04), but not descent (P > .10) (Table 1). A
larger sample size (14/group) may have provided the
power to detect a difference in the no handrail condition
during stair descent.
During stair ascent the mean joint angle profiles
in the sagittal plane were qualitatively similar between
groups (Figure 1, left). It was the case however that the
range of motion (ROM) at the ankle and knee associated
with the less affected side were greater than in controls
(P < .029) in both handrail conditions by 5.2° to 9.7°.
At the ankle, the ROM on the less affected side (36.7°
± 6.8°) also exceeded the ROM exhibited on the most
affected side (25.6° ± 6.9°; P = .026). For the without
handrail condition, the ROM at the hip (most affected
side) was about 5° greater than observed in control subjects (P = .014).
In the frontal plane, no differences in ROM were
detected between groups, conditions, or sides (stroke)
(P > .072) when climbing stairs even though the illustrated mean profiles suggest disparity (Figure 1, right).
The joint ROMs were small (< 15°) with substantial
variability making it difficult to detect any differences
across conditions.
The kinetics of stair ascent indicated that the magnitude of the peak extensor moments were generally smaller
on the affected side compared with the less affected side
and controls (Table 2, Figure 2). This was particularly
evident at the ankle during late stance (P < .001) and
at the knee in early stance or weight acceptance (P <
.002). The latter paired with a lower peak hip extensor
moment compared with the less affected side (P = .018)
contributed to a markedly lower peak support moment in
early stance on the most affected side compared with the
less affected side (P < .001) and controls (P < .024). The
peak support moment in late stance (pull-up) was lower in
stroke than controls (P < .003). Handrail use effectively
diminished the peak support moment generated in both
groups during weight acceptance (P < .001) corresponding to lower peak plantar flexor moments (P < .005).
In the frontal plane, differences between stroke and
control groups were limited to the ankle, which exhibited
much lower peak invertor moments in stroke (both sides,
P < .033) during ascent. Side-to-side differences were
noted at the knee (P < .047) and hip (P < .026) during
handrail use only as the magnitudes of the peak abductor
moments on the affected side appeared to be somewhat
less than observed without handrail use. Handrail use,
however, was not associated with any significant change
in peak moment output at any joint (P > .37) (Table 2).
As with stair ascent, the joint angular displacement
profiles associated with stair descent were qualitatively
similar in shape for both groups (Figure 3). There was
however, in the sagittal plane markedly smaller ROM
at the ankle on the most affected side (40.8° ± 9.4°)
compared with the less affected side (52.9° ± 16.4°) and
controls (51.7° ± 4.6°), P < .021. Greater hip ROM on the
most affected side (18.1° ± 3.9°) compared with controls
(13.9° ± 3.2°) was also detected (P < .025). On the less
affected side, the mean ROM at all joints tended to be
Table 1 Mean (SD) values for cadence (steps/min) and stance time (seconds) during stair ascent
and descent
Cadence (steps/min)
Ascent
Descent
Stance time (seconds)
Stroke
Control
Aff
LAff
Control
NH
71.11(10.30)*
94.72 (10.32)*
1.33 (.30)†
1.32 (.25)‡
1.05 (.22)†‡
H
65.60 (10.82)*
95.48 (10.84)*
1.38 (.29)†
1.45 (.26)‡
1.11 (.23)†‡
NH
75.56 (13.43)*
106.82 (16.98)*
1.11 (.14)
1.12 (.33)
.97 (.19)
H
68.13 (10.58)*
102.77 (21.35)*
1.12 (.16)
1.24 (.33)
.96 (.39)
Note. Aff, most affected side (stroke); LAff, less affected side (stroke); NH, no handrail; H, handrail.
* Significant difference between groups (P < .05).
† Significant difference between affected and control (P < .05).
‡ Significant difference between less affected and control (P < .05).
446 Novak and Brouwer
Figure 1 — Mean sagittal plane (left) and frontal plane (right) angles of the ankle, knee and hip during stair ascent without handrail
use (left) and with handrail use (right). Positive values represent eversion and flexion angles. All curves are normalized to 100% of the
stance phase. Thin line = control group; solid thick line = most affected side, stroke group; dashed line = less affected side, stroke group.
greater than in controls, significantly so at the ankle (P
= .029) and knee (P = .001); the latter for the without
handrail condition only. The use of a handrail had no
effect on ROM for either group.
In the frontal plane, the joint ROMs were relatively
small (<14°) during stair descent. There were no differences detected between groups, conditions, or sides
(stroke) (P > .138).
The kinetics of stair descent revealed that the mean
peak moment magnitudes were typically smaller on the
most affected side as compared with control subjects
(Table 3, Figure 4). However, statistical significance was
limited to the peak knee extensor moment during early
stance (P < .030) and the peak support moment during
controlled lowering (P < .039). Side-to-side differences
in stroke were evident in the peak extensor moments
generated at the knee (P < .012) and the peak support
moment (P < .013), both in late stance reflecting lower
values on the most affected side. With handrail use,
interlimb differences were also evident at the ankle with
a lower peak plantar flexor moment on the most affected
side during controlled lowering (P = .037). The impact
of handrail use was to lower the peak magnitudes of the
moments generated, though the effect was significant only
with respect to the peak support moment during weight
acceptance (P < .012).
In the frontal plane, the peak ankle invertor and hip
abductor moments on the most affected side were smaller
than those of control subjects (P < .014), but only for the
without handrail condition during stair descent. With
handrail use, a side-to-side difference in the peak hip
abductor moment was evident (P < .013) and attributed
to the reduction in muscle output on the most affected
side relative to the without handrail condition (Table 3).
Discussion
This paper provides a detailed kinematic and kinetic
description of stair ambulation comparing stroke and
healthy adults. The major findings are that while people
with stroke present with kinematic and kinetic profiles
that are similar in shape, there are marked differences
Stair Ambulation: Stroke and Healthy Adults 447
Figure 2 — Mean sagittal plane (left) and frontal plane (right) moments of the ankle, knee, hip and support during stair ascent without
handrail use (left) and with handrail use (right). Positive values represent internal eversion and flexion moments. All curves are normalized to 100%
of the stance phase. Thin line = control group; solid thick line = most affected side, stroke group; dashed line = less affected side, stroke group.
in ROM and the magnitudes of the peak moments produced between groups and between sides in stroke that
are important in understanding deficiencies associated
with stroke. Further, the impact of handrail use in stroke
provides insight into its use in compensating physical
limitations, likely serving to enhance stability.
In overground walking, the positive relationships of
both joint ROM and muscle output with walking speed
are well established.12,16,24–26 In stroke, walking speed is
markedly reduced compared with similarly aged healthy
adults secondary to stroke-related sensorimotor impairments including paresis.16,25 The limitations in lower
limb strength and associated reductions in power output
result in slower walking speeds and a restricted capacity
to walk faster.16 The lower cadence observed in our stroke
subjects during stair ascent and descent may reflect a
similar phenomenon. Controlling for cadence (by treating it as a covariate) eliminated the significance of the
between group differences; however, given that the lower
cadence in stroke did not uniformly affect kinematic and
kinetic parameters (ROM and peak moments) across
joints suggests that following stroke, stair negotiation is
achieved by select compensation strategies that cannot be
explained by speed alone. Indeed understanding where
significant differences lie using self-selected speed as the
standard instruction may guide intervention strategies that
could ultimately lead to more normal speeds of ascent
and descent following stroke.
Except for reduced ankle ROM on the affected
side, lower limb joint excursions were similar to the less
affected side. Despite the slower cadence, the less affected
ankle and knee showed a greater degree of extension than
was observed in controls during midstance (see Figure
1). The net effect would be to raise the vertical position
of the center of mass and facilitate clearance of the
intermediate step by the affected leg as it progresses
through swing12 which could minimize concerns of
tripping. The corresponding extensor moments were of
similar magnitude to those generated by control subjects
suggesting that any anticipated reduction due to slower
speed may have been offset by the exaggerated vertical
displacement.
448 Novak and Brouwer
Table 2 Sagittal and frontal plane peak moments (N·m/kg) during the stance phase of stair ascent, mean (SD)
No Handrail
Handrail
Aff
LAff
Control
Aff
Laff
Control
PF 1
–.62 (.10)‡
–.60 (.13)‡
–.69 (.24)‡
–.48 (.14)‡†
–.48 (.18)‡
–.61 (.28)‡†
PF 2
–.85 (.17)†
–.96 (.14)†
–1.17 (.09)†
–.83 (.14)*†
–.99 (.12)*†
–1.17 (.11)†
.15 (.10)
.25 (.10)
.22 (.07)
.12 (.08)*†
.21 (.09)*
.22 (.12)†
Extensor
–.80 (.15)*†
–1.09 (.20)*
–1.14 (.21)†
–.78 (.15)*
–1.17 (.19)*
–1.12 (.15)
Flexor
.23 (.10)*†
.09 (.09)*
.12 (.07)†
.21 (.07)*†
.11 (.08)*
.11 (.10)†
Extensor
–.45 (.27)*
–.59 (.32)*
–.41 (.11)
–.41 (.21)
–.46 (.22)
–.40 (.14)
Ext 1
–1.70 (.35)*‡†
–2.13 (.41)*‡
–2.08 (.29)‡†
–1.56 (.27)*‡†
–1.92 (.34)*‡
–2.00 (.25)‡†
Ext 2
–.83 (.22)†
–.82 (.14)†
–1.16 (.19)†
–.85 (.20)†
–.82 (.15)†
–1.16 (.16)†
Sagittal Plane
Ankle
Knee
Hip
Support
Flexor
Frontal Plane
Ankle
Inverter
–.15 (.06)†
–.16 (.07)†
–.29 (.13)†
–.12 (.06)
–.14 (.07)
–.36 (.39)
Knee
Peak abd
.29 (.13)
.35 (.11)
.29 (.09)
.27 (.10)*
.36 (.11)*
.28 (.14)
Hip
Peak abd
.64 (.17)
.69 (.13)
.74 (.14)
.54 (.20)*
.66 (.21)*
.71 (.27)
Note. Aff, most affected side (stroke); LAff, less affected side (stroke); PF, plantar flexion; Ext, extension; Abd, abduction. Positive values indicate dorsiflexion,
flexion, eversion, or abduction; Negative values indicate plantar flexion, extension, inversion, or adduction.
* Significant difference between sides (stroke only).
† Significant difference between groups (stroke vs control).
‡ Significant difference between conditions (handrail vs no handrail).
More so than the kinematic profiles, examination of
how the task of stair ascent was accomplished revealed
qualitative disparities in pattern between sides (in stroke)
and groups. The peak moments produced on the affected
side were mostly lower than on the less affected side and
at all joints were lower than those of control subjects. In
healthy adults, stair climbing places considerable demand
on the knee extensors2 and together with the ankle plantar
flexors, contribute significantly to generating positive
work2,8,27 as we have shown in the current study. In stroke,
the slower speed reduced the ankle and knee moment
requirements; however, while the plantar flexor moments
were lower bilaterally, the net knee extensor output
was reduced on the affected side only. Furthermore,
the mean peak extensor moments at the hip appeared
higher in stroke (both sides) though the trend was not
borne out statistically and post hoc analysis indicated
this was not a function of the small sample. Nonetheless, the combined hip and knee extensor output served
to generate a net extensor support moment on the less
affected side during weight acceptance comparable
to that of controls; a strategy that would enhance stability during transition as the affected limb is off-loaded.8
Consistent with this interpretation, when stability was
enhanced by the use of a handrail, the magnitude of the
net support moment was lower. Though this was primarily associated with a decrease in the plantar flexor
moment, a general pattern of somewhat lower extensor
moments was observed in all the joint profiles and the
side-to-side difference in hip extensor moment magnitudes disappeared. In control subjects, only the plantar
flexor moment showed a reduction in magnitude during
weight acceptance (and a corresponding decrease in the
net support moment) but no s were detected elsewhere
when a handrail was used.
In the frontal plane, only the ankle inverter moments
were lower in stroke (both sides) than controls, secondary
to the group’s slower cadence during stair ascent in the
case of the affected side. The relatively large moments
at the hip in both groups is consistent with the abductors helping to stabilize the upper body over the base of
support.2,8 That a side-to-side difference was detected in
subjects with stroke during handrail use only suggests
that while the hip abductors on the affected side may be
able to provide stability, the challenge may be lessened
if an alternate solution is presented. In the case of controls, the use of a handrail yielded a negligible impact.
Considering that those with stroke showed deficits in
balance and mobility, it may not be surprising that the
availability of a handrail influenced the task. Certainly in
the case of two individuals with stroke, it allowed them
to perform the task.
In stair descent, the ROM required at all lower limb
joints is larger during stance than during swing which is
the reverse of the pattern during ascent.3,9 Consequently
limited or impaired joint mobility is more likely to
Stair Ambulation: Stroke and Healthy Adults 449
Figure 3 — Mean sagittal plane (left) and frontal plane (right) angles of the ankle, knee and hip during stair descent without handrail
use (left) and with handrail use (right). Positive values represent eversion and flexion angles. All curves are normalized to 100% of the
stance phase. Thin line = control group; solid thick line = most affected side, stroke group; dashed line = less affected side, stroke group.
alter the pattern of movement in stance during descent.
Our stroke group exhibited reduced dorsiflexion on the
affected side during late stance or controlled lowering.
Reeves and colleagues9 identified this period of stance
as one of high risk for falls among the elderly because
they are operating near their maximum available range
of motion to lower the body to allow the opposite limb
to contact the next step. By compensating with exaggerated hip flexion (most affected side), subjects with
stroke were able to adequately shorten the limb to lower
the swing limb onto the next step.28 The correspondingly
low extensor support moment however, may compromise stability. Interestingly, even with handrail use, the
net extensor moment was sustained at a comparable
magnitude which may be a strategy to offset the greater
instability associated with descent5 compared with ascent
and at this point in stance in particular.9 Note that during
weight acceptance, the net extensor moment was lower
with handrail use, but only on the less affected side and
for control subjects, which further supports the notion
that on the most affected side, people with stroke are not
inclined to reduce the net lower limb extensor support.
Stair descent is accomplished primarily through
negative work as muscles contract eccentrically to control
the lowering of the body with gravity and the knee extensors are a major contributor.8,27 In stroke, markedly lower
knee extensor moments were produced on the affected
side compared with the less affected side and control
subjects. The lower cadence explains the difference
from controls although the fact that the less affected side
compared well to controls regardless of the slower speed
suggests other factors also played a role. The knee exhibits the greatest range of motion of all lower limb joints
during descent, which necessarily requires considerable
dynamic control.29,30 To promote stability it is possible
that people with stroke increase their knee stiffness by
co-contracting the knee flexors, which would reduce
the net extensor moment, but increase joint stability.31
Electromyographic recordings would have been useful
in determining if this was the case.
450 Novak and Brouwer
Figure 4 — Mean sagittal plane (left) and frontal plane (right) moments of the ankle, knee, hip and support during stair descent without handrail use (left) and with handrail use (right). Positive values represent internal eversion and flexion moments. All curves are
normalized to 100% of the stance phase. Thin line = control group; solid thick line = most affected side, stroke group; dashed line =
less affected side, stroke group.
Similar to stair ascent, the frontal plane moments
revealed the highest output at the hip as the abductors
stabilize the trunk over the support limb. With handrail
use a significant side-to-side difference in the mean peak
hip abductor moment was evident which qualitatively
appeared to be associated with a lower moment on the
most affected side. Conceivably the handrail may serve
to provide lateral stability enabling less reliance on the
paretic abductors.
It is not possible to determine whether alterations
in moment magnitudes associated with handrail use
were attributable to the redistribution of loads through
the hand and upper extremity. Such conclusions would
require knowledge of the forces transmitted to the handrail and an analytical approach including the upper limb
segments. Future studies should consider this given that
the magnitudes of some peak moments were different in
relation to handrail use.
It warrants stating that this comprehensive biomechanical analysis of stair ambulation in stroke is
limited to individuals with mild to moderate mobility
impairment (as indicated by scores on the Community
Balance and Mobility Scale) and are not generalizable
to the entire stroke population with a range of sensorimotor deficits. Even so, it is clear that compared with
age-matched healthy adults, significant alterations in
movement patterns and movement control were evident
thus highlighting the challenge of stair ambulation in this
group.
There has been a paucity of information describing
how people with stroke manage stairs and the strategies
they employ. The findings presented describe the mechanics and highlight stroke-related differences in lower limb
mobility and motor output that are not attributable to
aging. Slower cadence accounts for the lower muscle
work required in stroke though lower extremity extensor
weakness limits the overall net support compared with
healthy adults. It would be of interest to explore whether
strategies aimed at augmenting the output of those muscle
groups identified as being abnormally low would translate
Stair Ambulation: Stroke and Healthy Adults 451
Table 3 Sagittal and frontal plane peak moments (N·m/kg) during the stance phase of stair descent, mean
(SD)
No Handrail
Handrail
Aff
LAff
Control
Aff
Laff
Control
PF 1
–.61 (.18)
–.65 (.25)
–.66 (.26)
–.67 (.17)
–.63 (.10)
–.63 (.23)
PF 2
–.91 (.15)
–1.02 (.11)
–.99 (.16)
–.90 (.10)*
–1.02 (.11)*
–.98 (.12)
Ext 1
–.58 (.19)†
–.69 (.22)
–.92 (.37)†
–.45 (.17)†
–.50 (.20)
–.82 (.33)†
Ext 2
–.99 (.09)*
–1.19 (.17)*
–1.13 (.22)
–.90 (.17)*†
–1.18 (.24)*
–1.07 (.14)†
Flexor
.33 (.08)
.41 (.20)
.37 (.34)
.29 (.13)
.36 (.27)
.31 (.17)
Extensor
–.15 (.25)
–.22 (.28)
–.21 (.23)
–.14 (.16)
–.23 (.22)
–.19 (.21)
Ext 1
–1.23 (.53)
–1.22 (.49)‡
–1.47 (.46)‡
–1.18 (.36)
–1.09 (.34)‡
–1.33 (.46)‡
Ext 2
–1.59 (.15)*†
–1.87 (.29)*
–1.83 (.26)†
–1.55 (.16)*†
–1.84 (.16)*
–1.77 (.24)†
Sagittal Plane
Ankle
Knee
Hip
Support
Frontal Plane
Ankle
Inverter
–.19 (.08)†
–.18 (.09)†
–.33 (.09)†
–.16 (.07)
–.16 (.09)
–.40 (.41)
Knee
Peak abd
.30 (.12)
.38 (.16)
.39 (.15)
.31 (.11)
.40 (.08)
.39 (.20)
Hip
Peak abd
.69 (.11)†
.75 (.07)
.86 (.16)†
.59 (.18)*
.74 (.09)*
.79 (.31)
Note. Aff, most affected side (stroke); LAff, less affected side (stroke); PF, plantar flexion; Ext, extension; Abd, abduction. Positive values indicate dorsiflexion,
flexion, eversion or abduction; negative values indicate plantar flexion, extension, inversion, or adduction.
* Significant difference between sides (stroke only).
† Significant difference between groups (stroke vs control).
‡ Significant difference between conditions (handrail vs no handrail).
into faster cadence stair ascent and descent while preserving stability.
Acknowledgments
We acknowledge Samantha Reid for her help with data collection. Funded in part from the Heart and Stroke Foundation
of Ontario (grant NA 7369) and a Natural Sciences and Engineering Research Council (NSERC) Postgraduate Doctoral
Award (ACN).
References
1.Andriacchi TP, Galante JO, Fermier RW. The influence of total knee-replacement design on walking and
stair-climbing. J Bone Jt Surg Am. 1982;64:1328–1335.
PubMed
2. Nadeau S, McFadyen BJ, Malouin F. Frontal and sagittal plane analyses of the stair climbing task in healthy
adults aged over 40 years: what are the challenges compared to level walking? Clin Biomech (Bristol, Avon).
2003;18(10):950–959. PubMed doi:10.1016/S02680033(03)00179-7
3. Protopapadaki A, Drechsler WI, Cramp MC, Coutts FJ,
Scott OM. Hip, knee, ankle kinematics and kinetics during
stair ascent and descent in healthy young individuals. Clin
Biomech (Bristol, Avon). 2007;22(2):203–210. PubMed
doi:10.1016/j.clinbiomech.2006.09.010
4.Alzahrani MA, Dean CM, Ada L. Ability to negotiate
stairs predicts free-living physical activity in communitydwelling people with stroke: an observational study. Aust
J Physiother. 2009;55(4):277–281. PubMed doi:10.1016/
S0004-9514(09)70008-X
5. Startzell JK, Owens DA, Mulfinger LM, Cavanagh PR.
Stair negotiation in older people: a review. Am Geriatr
Soc. 2000;48(5):567–580. PubMed
6.Mian OS, Thom JM, Narici MV, et al. Kinematics of
stair descent in young and older adultsand the impact of
exercise training. Gait Posture. 2007;25(1):9–17. PubMed
doi:10.1016/j.gaitpost.2005.12.014
7. Benedetti MG, Berti L, Maselli S, Mariani G, Giannini
S. How do the elderly negotiate a step? A biomechanical
assessment. Clin Biomech (Bristol, Avon). 2007;22(5):567–
573. PubMed doi:10.1016/j.clinbiomech.2007.
01.010
8.Novak AC, Brouwer B. Sagittal and frontal lower limb
joint moments during stair ascent and descent in young
and older adults. Gait Posture. 2011;33(1):54–60. PubMed
doi:10.1016/j.gaitpost.2010.09.024
9. Reeves ND, Spanjaard M, Mohagheghi AA, Baltzopoulos
V, Magnaris CN. The demands of stair descent relative
to maximum capacities in elderly and young adults. J
Electromyogr Kinesiol. 2008;18(2):218–227. PubMed
doi:10.1016/j.jelekin.2007.06.003
10. Reeves ND, Spanjaard M, Mohagheghi AA, Baltzopoulos
V, Magnaris CN. Older adults employ alternative strategies to operate within their maximum capabilities when
452 Novak and Brouwer
ascending stairs. J Electromyogr Kinesiol. 2009;19(2):e57–
e68. PubMed doi:10.1016/j.jelekin.2007.09.009
11.Reid SM, Graham RB, Costigan PA. Differentiation of
young and older adult stair climbing gait using principal
component analysis. Gait Posture. 2010;31(2):197–203.
PubMed doi:10.1016/j.gaitpost.2009.10.005
12. Kim CM, Eng JJ. Magnitude and pattern of 3D kinematic
and kinetic gait profiles in persons with stroke: relationship to walking speed. Gait Posture. 2004;20(2):140–146.
PubMed doi:10.1016/j.gaitpost.2003.07.002
13. Novak AC, Olney SJ, Bagg S, Brouwer B. Gait changes following botulinum toxin A treatment in stroke. Top Stroke
Rehabil. 2009;16(5):367–376. PubMed doi:10.1310/
tsr1605-367
14.Olney SJ, Griffin MP, McBride I. Temporal, kinematic
and kinetic variables related to gait speed in subjects
with hemiplegia: a regression approach. Phys Ther.
1994;74(9):872–885. PubMed
15. Olney SJ, Richards C. Hemiparetic gait following stroke.
Part I: Characteristics. Gait Posture. 1996;4(2):136–148.
doi:10.1016/0966-6362(96)01063-6
16. Parvataneni K, Olney SJ, Brouwer B. Changes in muscle
group work associated with changes in gait speed of
persons with stroke. Clin Biomech (Bristol, Avon).
2007;22(7):813–820. PubMed doi:10.1016/j.clinbiomech.2007.03.006
17.Novak AC, Brouwer B. Mobility Function Following
Stroke: The Impact of Multiple Impairments. In: Pelligrino
LT, ed. Handbook of motor skills: development, impairment and therapy. New York: Nova Science Publishers;
2010:117–137.
18. Lord SE, McPherson K, McNaughton HK, Rochester L,
Weatherall M. Community Ambulation after stroke: how
important and obtainable is it and what measures appear
predictive. Arch Phys Med Rehabil. 2004;85(2):234–239.
PubMed doi:10.1016/j.apmr.2003.05.002
19.Thorngren M, Westling B, Norrving B. Outcome after
stroke in patients discharged to independent living.
Stroke. 1990;21(2):236–240. PubMed doi:10.1161/01.
STR.21.2.236
20. Tsuji T, Sonada S, Domen K, Saitoh E, Liu M, Chino N.
ADL structure for stroke patients in Japan on the functional independence measure. Am J Phys Med Rehabil.
1995;74(6):432–438. PubMed doi:10.1097/00002060199511000-00007
21. Knorr S, Brouwer B, Garland SJ. Validity of the community
balance and mobility scale in community-dwelling persons
after stroke. Arch Phys Med Rehabil. 2010;91(6):890–896.
PubMed doi:10.1016/j.apmr.2010.02.010
22. Howe J, Inness E, Venturini A, William JI, Verrier MC.
The Community Balance and Mobility Scale: a balance
measure for individuals with traumatic brain injury. Clin
Rehabil. 2006;20(10):885–895. PubMed
23.Brouwer B, Parvataneni K, Olney SJ. A comparison of
gait biomechanics and metabolic requirements of overground and treadmill walking in people with stroke. Clin
Biomech (Bristol, Avon). 2009;24(9):729–734. PubMed
doi:10.1016/j.clinbiomech.2009.07.004
24.Kerrigan DC, Todd MK, Della Croce U, Lipsitz LA,
Collings JJ. Biomechanical gait alterations independent of
speed in the healthy elderly: evidence for specific limiting
impairments. Arch Phys Med Rehabil. 1998;79(3):317–
322. PubMed doi:10.1016/S0003-9993(98)90013-2
25. Nadeau S, Gravel D, Arsenault B, Bourbonnais D. Plantarflexor weakness as a limiting factor of gait speed in
stroke subjects and the compensating role of the hip flexors. Clin Biomech (Bristol, Avon). 1999;14(2):125–135.
PubMed doi:10.1016/S0268-0033(98)00062-X
26.Nadeau S, Arsenault B, Bertrand A, Gravel D, Bourbonnais D. Analysis of the clinical factors determining
natural and maximal gait speeds in adults with a stroke.
Am J Phys Med Rehabil. 1999;78(2):123–130. PubMed
doi:10.1097/00002060-199903000-00007
27. McFadyen BJ, Winter DA. An integrated biomechanical
analysis of normal stair ascent and descent. J Biomech.
1988;21(9):733–744. PubMed doi:10.1016/00219290(88)90282-5
28.Radtka SA, Oliveira GB, Lindstrom KE, Borders MD.
The kinematic and kinetic effects of solid, hinged, and no
ankle-foot orthoses on stair locomotion in healthy adults.
Gait Posture. 2006;24(2):211–218. PubMed doi:10.1016/j.
gaitpost.2005.09.005
29. Riener R, Rabuffetti M, Frigo C. Stair ascent and descent
at different inclinations. Gait Posture. 2002;15(1):32–44.
PubMed doi:10.1016/S0966-6362(01)00162-X
30. Zachazewski JE, Riley PO, Krebs DE. Biomechanical analysis of body mass transfer during stair ascent and descent
of healthy subjects. J Rehabil Res. 1993;30(4):412–422.
PubMed
31.Hortobagyi T, DeVita P. Altered movement strategy
increases lower extremity stiffness during stepping down in
the aged. J Gerontol A Biol Sci Med Sci. 1999;54(2):B63–
B70. PubMed doi:10.1093/gerona/54.2.B63