Decreased gastrocnemius temporal muscle activation during gait in elderly

women with history of recurrent falls

Renata Noce Kirkwood
a,
*, Renato Guilherme Trede
b
, Bruno de Souza Moreira
c
,
Scott Alexander Kirkwood
d
, Leani Souza Ma ximo Pereira
e
a
Department of Physical Therapy, Universidade Federal de Minas Gerais, Minas Gerais, Brazil
b
Department of Physical Therapy, Universidade Federal do Vale do Jequitinhonha e Mucur, Minas Gerais, Brazil
c
Rehabilitation Science, Universidade Federal de Minas Gerais, Minas Gerais, Brazil
d
Software Engineering, ForUsers Tecnologia Ltda., Minas Gerais, Brazil
e
Universidade Federal de Minas Gerais, Minas Gerais, Brazil
1. Introduction
Falling is an important public health problem, particularly
among the elderly population. A greater susceptibility of the
elderly to falls is due to the high prevalence of comorbidities
associated with functional decline resulting from the aging process
[1]. In the United States, falls are the main cause of injuries and
hospitalization among people over the age of 65 [2]. In Brazil, the
incidence of falls is similar to these international gures, with
approximately 29% of the elderly population falling at least once a
year, 14% of whom become recurrent fallers [3].
Gait dysfunction is among the main risk factors for falls [4]. An
unstable gait with inconsistent cycles is common during the aging
process, even in the absence of any specic diseases [5]. A
reduction in stride length and velocity and an increase in the
support phase are conservative strategies that elderly individuals
employ to gain stability [4,6,7]. In addition, a slight increase in
within-participant standard deviation variability of the temporal
and spatial parameters increases instability and the chance of
falling nearly twofold [4].
If gait variability is a strong predictor of falls, what might be the
behavior of muscle activity underlying this variability? We can
examine studies of the normal elderly to get some insight. For
example, elderly individuals often increase coactivation across the
ankle and knee joints as a strategy to stiffen the limb during single
support to compensate for reduced muscle strength and joint
laxity [8,9]. In threatening situations, higher levels of tibialis
anterior and gastrocnemius muscles activity were observed in the
elderly group in comparison to a young group with the objective to
increase gait stability [2]. In addition, elderly women adopted
different strategies, increasing tibialis anterior and rectus femoris
muscle activity, when their walking speed was increased [10,11].
Despite the growing number of studies involving electromyog-
raphy during gait of elderly individuals, there are no studies
comparing the temporal activation of the ankle muscles during gait
in elderly individuals with a history of recurrent falls. Temporal
activation curves reveal the time contraction, onset and termina-
Gait & Posture 34 (2011) 6064
A R T I C L E I N F O
Article history:
Received 9 July 2010
Received in revised form 23 February 2011
Accepted 9 March 2011
Keywords:
Gait
Falls
Human locomotion
Electromyography
Elderly
A B S T R A C T
Gait dysfunction is a strong issue in elderly women with a history of falls. The purpose of this study was
to compare the temporal activity of the ankle muscles during gait in elderly women with and without a
history of recurrent falls. Eighty-nine (89) elderly women one group with a history of falls (45) and
another group without (44) participated in the study. The mean range of temporal activation of the
gastrocnemius, tibialis anterior and soleus muscles during gait was obtained using electromyography.
The muscles were considered active when the signal magnitude surpassed two standard deviations of
the minimal magnitude of the average signal per individual. The results showed that the mean range of
gastrocnemius muscle activation of the group of recurrent fallers was signicantly shorter, 2.9%
(16.9 5.7%) compared to the group without recurrent falls (19.8 6.6%) (p = 0.004). The shorter duration in
the gastrocnemius muscle activation during stance could possibly affect stability in the support phase, since
the gastrocnemius is the main decelerator of the trunk. Clinically, this nding shows the importance of
rehabilitation programs for elderly women that focus on strengthening the plantar exor musculature
aiming to reestablish the function and stability of gait and possibly avoiding falls.
2011 Elsevier B.V. All rights reserved.
* Corresponding author. Tel.: +55 31 3409 4783; fax: +55 31 3409 4781;
mobile: +55 31 9985 0707.
E-mail addresses: renata.kirkwood@gmail.com (R.N. Kirkwood),
renato.trede@gmail.com (R.G. Trede), onurbsm@yahoo.com.br
(B. de Souza Moreira), scott@forusers.com (S.A. Kirkwood),
leanismp.bh@terra.com.br (L.S.M. Pereira).
Contents lists available at ScienceDirect
Gait & Posture
j o u r n al h omep age: www. el s evi er . co m/ l oc at e/ g ai t p ost
0966-6362/$ see front matter 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.gaitpost.2011.03.012
tion, which is important to demonstrate the electromyographic
(EMG) patterns between groups of individuals [12]. Knowledge of
the temporal activation of the tibialis anterior, gastrocnemius and
soleus muscles will give us insight into the strategies used by
elderly women with a history of falling.
Therefore, the present study compared the temporal behavior
of ankle muscles during gait in elderly women with and without a
history of recurrent falls. Our hypothesis is that recurrent fallers
will have a longer activation of synergists and/or antagonists in an
attempt to maintain stability during the support phase of gait.
2. Method
2.1. Design
A cross sectional observational study was conducted in elderly females with and
without history of recurrent falls. The individuals were recruited from the
community and assisted-living centers in the city of Belo Horizonte-Brazil.
Temporal muscle activity during walking was collected in one day in a laboratory.
The present study received approval from the Ethics Committee of the Universidade
Federal de Minas Gerais under process number ETIC 088/04. All participants signed
the informed consent.
2.2. Participants
Eighty-nine (89) elderly women, 45 with a history of recurrent falls and 44 with no
history of recurrent falls participated in the study. A fall was dened as an unexpected
event in which an individual comes to rest on a lower level [13]. The inclusion criteria
were as follow: female gender, age equal to or above 60 years and ability to walk
without assistance. The criteria of exclusion for both groups were orthopedic or
neurological diseases that could affect gait performance. A history of recurrent falls
was dened as at least two episodes of falls in the previous twelve months, whereas no
history of recurrent falls was dened as either the absence of falls or only one fall in the
previous twelve months [14]. Characteristics of the participants, age (years), height
(m) and body mass index (BMI) (kg/m
2
) were collected in order to describe the
differences between the groups. Fear of falls was also assessed through a simple
question (Are you afraid of falling?), with a yes/no answer.
2.3. Outcome measures
An electromyograph (MP150WSW Biopac Systems
1
, CA, USA) with an input
impedance of 2 M Ohm and a common-mode rejection capacity of 110 dB with four
bipolar pre-amplied active surface electrodes (TSD-150A/B Biopac System), was used
for capturing the signals. Data was captured by the AcqKnowledge program (Biopac
Systems
1
, CA, USA) with a frequency of 1000 Hz and gain of 350 times, for 10 s.
The assessed limb was the dominant one, dened by asking the volunteer which
limb she would kick a ball with. The electrodes had a diameter of 11.4 mm and were
attached to the greatest muscle volume: tibialis anterior (TA), placed parallel to the
tibia; medial muscle belly of the gastrocnemius (GAS) on the posteriomedial side of
the leg; soleus (SOL) below the GAS muscle belly and the reference electrode was
placed on the tibial anterior tuberosity [15]. The participants walked barefoot on a
rubber mat six meters in length by one meter in width.
Gait events were obtained using two footswitches (Heel/Toe Strike Transducer
Biopac Systems
1
, CA, USA) attached to the centre of the heel and ball of the foot. The
footswitches were connected to the EMG system and synchronized with the EMG
data through the AcqKnowledge program.
2.4. Data reduction
The total temporal muscle activity during the stance phase of the gait cycle (in %)
and latency (s) (interval between heel contact and the beginning of muscle activity)
of the GAS, TA and SOL muscles were obtained. Data collected was transferred to the
PlotEMG program (ForUsers Tecnologia Ltda, Brazil) and processed as follows: the
signal was rectied and cut with a low-pass lter at 6 Hz frequency generating the
linear envelope [16]. The signal was then normalized at 101 points from initial
contact of the foot until the next contact of the same foot, using information from
the footswitches. The magnitude of the electromyographic signal was normalized
by the highest peak in these 101 points. The muscle was considered active when the
signal magnitude surpassed two standard deviations of the minimal magnitude of
the average signal per individual and was considered inactive when the signal was
below the established standard deviations [12].
The footswitch signals and the EMGdata were detected by the PlotEMG program,
which was exible enough to allow the user to select which strides to analyze. To
avoid acceleration and deceleration effects, the rst and last strides were not used
in the analyses. On average, 3.0 (1.2) cycles per individual in the group with
recurrent falls, and 3.7 (1.4) cycles in the group without recurrent falls were
analyzed.
2.5. Data analysis
Baseline characteristics of participants are presented as values, means and
standard deviations (SD). The coefcient of variation showing on each graph was
obtained according to the literature [17]. The mean difference between the groups
with a 95% CI is also presented. Age, BMI and cycle times were compared between
groups using an independent t test. Due to a skewed distribution of the other
outcomes (height, muscle activity, latency, swing and stance percentage) the
MannWhitney U test and the KruskalWallis test were conducted. A Bonferronis
adjustment was applied to correct for multiple comparisons, which resulted in a
signicant limit of <0.00625 to attain a nominal estimate of <0.05 to avoid Type I
error.
3. Results
The recurrent fallers group with an average age of 74.0 years
(5.6) was signicantly older than the non recurrent fallers with a
mean age of 70.7 years (5.4) (p = 0.006). There were no signicant
differences between groups with regard to BMI and height (Table
1). In the recurrent fallers group, a history of 3 falls was more
prevalent and only 8 individuals in the non recurrent group had a
history of a fall in the previous year. Recurrent fallers answered
afrmatively 64.4% (29/45) compared to 47.7% (21/44) of non
recurrent fallers group, of being afraid of falling (Table 1).
Total temporal muscle activity demonstrated a signicant
difference between groups in the mean range activation of the GAS
during stance (p = 0.004). The group with recurrent falls presented
shorter GAS activity (2.9% less) during the stance phase (Table 2).
Table 1
Mean (SD), mean (95% CI), p value and number of falls and fear of falling of the baseline characteristics between the groups in the beginning of the study.
Characteristics of the participants Groups Difference between groups
Recurrent fallers N = 45 Non recurrent fallers N = 44 Recurrent fallers minus non recurrent fallers
Anthropometrics
Age (years) 74.0 (5.6) 70.7 (5.4) 3.3 (1.05.3)
a
p = 0.006
Height (m) 1.6 (0.7) 1.6 (0.7) 0 (0.3 to 0.3)
b
p = 0.546
BMI (kg/m
2
) 25.6 (4.0) 26.5 (4.2) 0.9 (0.3 to 0.8)
a
p = 0.308
Number of falls
1 in the past year 0 8
2 14 0
3 27 0
>4 3 0
Fear of falling
Yes/no 29/16 21/23
Bolded values indicate a statistically signicant difference.
a
Independet t test signicant at p < 0.025.
b
MannWhitney U signicant at p < 0.05.
R.N. Kirkwood et al. / Gait & Posture 34 (2011) 6064 61
All individuals exhibited GAS activation in the gait support phase,
with the earliest activation taking place at 14% of the cycle and the
latest at 57%. The mean range was from 29% to 47% of the gait cycle
for the recurrent fallers group and 25% to 46% for the non recurrent
fallers group. Four women in the group with a history of recurrent
falls and ve in the group without recurrent falls also exhibited
activation of the GAS muscle in the swing phase that ranged from
65% to 98% of the cycle. Latency of the GAS, TA and SOL and muscle
activity of the TA and SOL were not different between groups (Table
2). The cycle time differed signicantly between groups (p = 0.02),
with the recurrent fallers group showing a slight decrease, 0.1 s in
duration, with no reected changes in the other gait measurements
(Table 2). Fig. 1 displays the linear envelope and the coefcient of
variation of the muscles assessed during the gait cycle [17].
4. Discussion
This is the rst study to analyze the behavior of muscle
activation during gait between groups of elderly women with and
without a history of recurrent falls. Older womens issues are
important since women live longer than men and the risk for
across their life course is escalating [18]. Appropriate care and
support for this vulnerable group is very important.
The hypothesis that elderly women with a history of recurrent
falls would exhibit mechanisms of compensatory muscle activa-
tion during gait was not supported. It was expected that the
recurrent fallers group, 64.4% of them with a fear of falling, and
more history of falls, would increase temporal activation of the GAS
and TA muscle in an attempt to maintain stability. However, from
the EMG data the recurrent fallers activated the GAS muscle for a
signicantly shorter period of time during stance (2.9% reduction)
and no difference was found in the activation time of the TA muscle
between groups.
The literature shows plantar exors activity from mid stance to
terminal stance (12% to 50%) [19], which is a phase where the GAS
and SOL muscles eccentrically control the anterior rotation of the tibia
over the foot, followed by concentric contraction for the push-off
phase, accelerating the limb for swing [20]. The adequate control of
the anterior rotation of the tibia is important mainly for stabilizing
gait during the unipodal phase and ensuring knee exion, which
starts at the terminal stance and the beginning of the pre-swing phase
[20]. The shorter duration in the GAS muscle activity in the group with
a history of recurrent falls could probably reduce the anterior rotation
of the tibia over the foot, limiting the duration of the support phase
and leading to an early push-off. Consequently, a decrease in ankle
dorsiexion range of motion and likely increase in the demand of the
hip and knee exors would occur in order to help release the foot off
the ground. The result would be the reduction in the forward
momentum of the body, reduction in step length of the opposite limb,
and excess hip and knee exion in the initial swing phase [20]. All
these changes would generate a more unstable gait.
The importance of the GAS and SOL muscles in the maintenance of
stability and the forward progression of the body during gait has
been demonstrated [21]. In the mid stance phase, both muscles work
as decelerators of the trunk; however the contribution of the GAS is
greater in decelerating the trunk horizontally. At the end of the mid
support phase, the plantar exor muscles have an antagonistic
action, the SOL muscle transfers energy from the leg to the trunk and
the GAS muscle transfers energy fromthe trunk to the lower limbs.
The result of these reaction forces promotes the support of the trunk
and lower limbs, thereby ensuring a stable gait with progression. At
the end of the stance phase, the SOL accelerates the trunk and the GAS
transfers energy to the limbs, contributing to the beginning of the
oscillation phase. Although the study was a biomechanical simula-
tion, direct dynamics by means of power and acceleration analysis is
capable of informing the real cause between muscle activity and
executed activity [21]. From the results of the present study, the
shorter duration in the GAS muscle activity in the group of recurrent
fallers affected the stability in the support phase of the gait, as the
GAS is the main decelerator muscle of the trunk [21]. It could also
have an effect on the initiation of the swing phase, challenging other
muscles and joints to accomplish the task, since no difference was
found in stance % between groups. Therefore, lack of activity of the
GAS muscle activity could clearly impact on walking performance.
In the absence of GAS, an increase in SOL activity would take
place and vice versa. In the absence of both, an increase in the
activity of the femoral biceps and adductor longus and shorter
activity of the tibialis anterior muscle would occur [22]. The aim of
which would be to initiate the rotation of the foot, to restrict the
extension of the hip and ankle dorsiexion and to allow knee
Table 2
Mean (SD) of the groups, mean (95% CI) and p value of the difference between groups for muscle activity, latency and cycle parameters of the 88 elderly participants.
Outcomes Groups Difference between groups
Recurrent fallers N = 45 Non recurrent fallers N = 44 Recurrent fallers minus non recurrent fallers
Muscle activity during stance (%)
GAS 16.9 (5.7) 19.8 (6.6) 2.9 (5.5 to 0.3)
b
p = 0.004
TA 9.2 (5.0) 9.3 (5.3) 0.1 (2.1 to 2.3)
b
p = 0.933
SOL 22.1 (7.0) 24.9 (8.5) 2.8 (6.1 to 0.5)
b
p = 0.308
Latency of muscle activity (s)
GAS 0.3 (0.1) 0.3 (0.1) 0.0 (0.04 to 0.04)
b
p = 0.614
TA 0.04 (0.2) 0.03 (0.1) 0.01 (0.1 to 0.1)
b
p = 0.897
SOL 0.3 (0.1) 0.3 (0.1) 0.0 (0.04 to 0.04)
b
p = 0.973
Cycle
Time (s) 1.2 (0.1) 1.3 (0.1) 0.1 (0.14 to 0.06)
a
p = 0.02
Swing (%) 40.2 (3.6) 39.7 (4.8) 0.5 (1.3 to 2.3)
b
p = 0.751
Stance (%) 59.8 (3.6) 60.3 (4.8) 0.5 (2.3 to 1.3)
b
p = 0.751
Bolded values indicate a statistically signicant deference.
a
Independet t test signicant at p < 0.025.
b
KruskalWallis signicant at p < 0.00625.
R.N. Kirkwood et al. / Gait & Posture 34 (2011) 6064 62
extension in the support phase [22]. Based on these, it could be
expected that the recurrent fallers group would compensate for the
shorter GAS activation by increasing the SOL muscle activation for
a longer period of time, but such a difference between groups did
not occur. It is possible that the recurrent fallers group used
another strategy, such as activating the biceps femoris muscle or
even the adductors, as postulated in the biomechanical model [22].
However, these muscles were not investigated in the present
study. It is possible that other factors may have contributed to the
lack of compensatory actions. For example, reduced ankle plantar
exor strength in the elderly population has been described in the
literature, along with slower muscle contractile properties and
decreased dorsiexion range of motion [23].
Therefore, regaining the 2.9% of gastrocnemius muscle activa-
tion, in a group of individuals that would probably present shorter
muscle activity due to the aging process, is important for elderly
females as benets in improving stability in the support phase of
the gait and possibly avoiding falls. Resistance training programs
have shown improvement in the force of the gastrocnemius muscle
in a group of elderly males over the age of 70 [24]. Other studies
have shown the positive results of strength training in reducing or
even reversing the sarcopenia effects in older adults [25].
Clinically, our nding together with the evidences from the
literature reinforces the importance of strategies focused on the
rehabilitation of the plantar exor musculature in elderly females
with history of recurrent falls.
The average number of strides used in the analysis varied from
about 3 to 3.7 strides. Signicant errors would be introduced in the
analysis when too few strides are analyzed, but a great number of
strides dramatically increased the variability of the data [26]. On
average, the literature considers that three to four strides are
sufcient for a consistent result; therefore, by eliminating the
initial and nal strides, we achieved the recommended number of
strides [26,27].
One limitation of this study is that elderlysubjects walked with
their preferred walking velocity but velocity was not measured,
only cycle time. It is known that velocity has signicant effect over
most of gait measurements and could be reected in the results of
the present study [10,28,29]. However, recently Kang and
Dingwell [11] reported in a group of health elderly subjects, that
the electromyographic patterns of the gastrocnemius muscles did
not vary with age or speed during gait measurements, when
compared to a younger group. Although cycle time was different
between groups (0.1 s), the percentage of swing and stance was
similar. In addition, most falls occur during preferred walking
speed; therefore, we consider that our experimental condition
reects the situation that most of the recurrent elderly fallers face
every day.
Future studies involving the electromyographic analysis of
other muscles during gait are important for the understanding of
the mechanisms of muscle activation of elderly women with a
history of recurrent falls. Extending the ndings to other groups of
individuals it is also important, as the gender factor changes the
association of gait parameters and falls [30]. Moreover, it would be
interesting to assess gait in a more complex manner, kinetic
parameters associated to electromyography, in order to gain an
understanding of the biomechanical changes resulting from the
shorter temporal gastrocnemius muscle activation during stance
phase.
Ethics approval
The Ethics Committee of the Universidade Federal de Minas
Gerais (Brazil) under process number ETIC 088/04 approved this
study. All participants gave written informed consent before data
collection began.
Source(s) of support
None. The resources necessary to conduct the research were
available at the Physical Therapy Department at Universidade
Federal de Minas Gerais.
Acknowledgement
None.
Conict of interest statement
The authors declare that they have no conict of interest,
nancial or otherwise, related to the submitted manuscript.
References
[1] Ganz DA, Bao Y, Shekelle PG, Rubenstein LZ. Will my patient fall? JAMA
2007;297:7786.
[2] Brown LA, Gage WH, Polych MA, Sleik RJ, Winder TR. Central set inuences on
gait. Age-dependent effects of postural threat. Exp Brain Res 2002;145:286
96.
[3] Ramos LR. Determinant factors for healthy aging among senior citizens in a
large city: the epidoso project in Sao Paulo. Cad Saude Publica 2003;19:7938.
Fig. 1. Normalized EMG linear envelopes of the tibialis anterior, gastrocnemius and
soleus muscles during gait. Average (straight line) and STD (dotted line) of the non
recurrent fallers group (N = 44) and average temporal patterns (dashed line) of the
recurrent faller group (N = 45). The coefcients of variation are shown on the side of
each graph.
R.N. Kirkwood et al. / Gait & Posture 34 (2011) 6064 63
[4] Maki BE. Gait changes in older adults: predictors of falls or indicators of fear? J
Am Geriatr Soc 1997;45:31320.
[5] Callisaya ML, Blizzard L, Schmidt MD, McGinley JL, Srikanth VK. Ageing and
gait variability a population-based study of older people. Age Ageing
2010;39:1917.
[6] Chamberlin ME, Fulwider BD, Sanders SL, Medeiros JM. Does fear of falling
inuence spatial and temporal gait parameters in elderly persons
beyond changes associated with normal aging? J Gerontol 2005;60:
11637.
[7] Thies SB, Richardson JK, Ashton-Miller JA. Effects of surface irregularity and
lighting on step variability during gait: a study in healthy young and older
women. Gait Posture 2005;22:2631.
[8] Schmitz A, Silder A, Heiderscheit B, Mahoney J, Thelen DG. Differences in
lower-extremity muscular activation during walking between healthy older
and young adults. J Electromyogr Kinesiol 2008;19:108591.
[9] Hortobagyi T, DeVita P. Mechanisms responsible for the age-associated
increase in coactivation of antagonist muscles. Exerc Sport Sci Rev
2006;34:2935.
[10] Chung MJ, Wang MJ. The change of gait parameters during walking at different
percentage of preferred walking speed for healthy adults aged 2060 years.
Gait Posture 2010;31:1315.
[11] Kang HG, Dingwell JB. Dynamics and stability of muscle activations during
walking in healthy young and older adults. J Biomech 2009;42:22317.
[12] Chang WN, Lipton JS, Tsirikos AI, Miller F. Kinesiological surface electromyog-
raphy in normal children: range of normal activity and pattern analysis. J
Electromyogr Kinesiol 2006;17:43745.
[13] Lamb SE, Jorstad-Stein EC, Hauer K, Becker C. Development of a common
outcome data set for fall injury prevention trials: the Prevention of Falls
Network Europe consensus. J Am Geriatr Soc 2005;53:161822.
[14] Swanenburg J, de Bruin ED, Uebelhart D, Mulder T. Falls prediction in elderly
people: a 1-year prospective study. Gait Posture 2010;31:31721.
[15] Cram JR, Kasman GS, Holts J. Introduction to surface electromyography.
Gaithersburg: An Aspen Publication; 1998.
[16] Winter DA. Biomechanics and motor control of human movement. Waterloo:
John Wiley & Sons, Inc.; 2005.
[17] Winter DA, Yack HJ. EMG proles during normal human walking: stride-to-
stride and inter-subject variability. Electroencephalogr Clin Neurophysiol
1987;67:40211.
[18] Restrepo HE, Rozental M. The social impact of aging populations: some major
issues. Soc Sci Med 1994;39:132338.
[19] Neumann DA. Kinesiology of the musculoskeletal system. Foundations for
physical rehabilitation. St. Louis: Mosby, Inc.; 2002.
[20] Perry J, Burneld LM. Gait analysis. Normal and pathological function. Thoro-
fare: SLACK Incorporated; 2010.
[21] Neptune RR, Kautz SA, Zajac FE. Contributions of the individual ankle plantar
exors to support, forward progression and swing initiation during walking. J
Biomech 2001;34:138798.
[22] Jonkers I, Stewart C, Spaepen A. The complementary role of the plantar exors,
hamstrings and gluteus maximus in the control of stance limb stability during
gait. Gait Posture 2003;17:26472.
[23] McGibbon CA. Toward a better understanding of gait changes with age and
disablement: neuromuscular adaptation. Exerc Sport Sci Rev 2003;31:1028.
[24] Morse CI, Thom JM, Mian OS, Birch KM, Narici MV. Gastrocnemius specic
force is increased in elderly males following a 12-month physical training
programme. Eur J Appl Physiol 2007;100:56370.
[25] Holviala JH, Sallinen JM, Kraemer WJ, Alen MJ, Hakkinen KK. Effects of strength
training on muscle strength characteristics, functional capabilities, and bal-
ance in middle-aged and older women. J Strength Cond Res 2006;20:33644.
[26] Shiavi R, Frigo C, Pedotti A. Electromyographic signals during gait: criteria for
envelope ltering and number of strides. Med Biol Eng Comput 1998;36:1718.
[27] Arsenault AB, Winter DA, Marteniuk RG, Hayes KC. How many strides are
required for the analysis of electromyographic data in gait? Scand J Rehabil
Med 1986;18:1335.
[28] Jordan K, Challis JH, Newell KM. Walking speed inuences on gait cycle
variability. Gait Posture 2006.
[29] Kang HG, Dingwell JB. Effects of walking speed, strength and range of motion
on gait stability in healthy older adults. J Biomech 2008;41:2899905.
[30] Callisaya ML, Blizzard L, Schmidt MD, McGinley JL, Srikanth VK. Sex modies
the relationship between age and gait: a population-based study of older
adults. J Gerontol A Biol Sci Med Sci 2008;63:16570.
R.N. Kirkwood et al. / Gait & Posture 34 (2011) 6064 64