Download Knee Extension and Flexion Weakness in People With Knee

Griffith Research Online
https://research-repository.griffith.edu.au
Knee Extension and Flexion Weakness
in People With Knee Osteoarthritis: Is
Antagonist Cocontraction a Factor?
Author
L. Heiden, Tamika, G. Lloyd, David, R. Ackland, Timothy
Published
2009
Journal Title
Journal of Orthopaedic and Sports Physical Therapy
Copyright Statement
Copyright 2009 Journal of Orthopaedic and Sports Physical Therapy. Reproduced with permission of
the Orthopaedic Section and the Sports Physical Therapy Section of the American Physical Therapy
Association (APTA). Please refer to the journal's website for access to the definitive, published
version.
Downloaded from
http://hdl.handle.net/10072/40315
Link to published version
http://www.jospt.org/issues/articleID.2366,type.2/article_detail.asp
[
RESEARCH REPORT
]
TAMIKA L. HEIDEN, PhD¹š:7L?:=$BBEO:"PhD²šJ?CEJ>OH$79AB7D:"PhD²
Knee Extension and Flexion Weakness
in People With Knee Osteoarthritis:
Is Antagonist Cocontraction a Factor?
uadriceps weakness is one of the most common and disabling
impairments seen in individuals with knee osteoarthritis
(OA).19 Sufficient quadriceps and hamstrings strength, both
isometric and dynamic, is essential for undertaking basic
activities of daily living such as standing and walking.36 Muscle
strength testing has revealed that those with knee OA have a 25% to
45% loss of knee extension strength6,15,39 and a 19% to 25% loss of knee
Q
flexion strength,6,50 compared with similarly aged controls. There are 3 factors
thought to contribute to knee extension
and flexion weakness in those with knee
TIJK:O:;I?=D0 Controlled laboratory study,
cross-sectional data.
TE8@;9J?L;I0 To investigate isometric knee
flexion and extension strength, failure of voluntary
muscle activation, and antagonist cocontraction of
subjects with knee osteoarthritis (OA) compared
with age-matched asymptomatic control subjects.
T879A=HEKD:0 Quadriceps weakness is a
common impairment in individuals with knee OA.
Disuse atrophy, failure of voluntary muscle activation, and antagonist muscle cocontraction are
thought to be possible mechanisms underlying this
weakness; but antagonist cocontraction has not
been examined during testing requiring maximum
voluntary isometric contraction.
TC;J>E:I0 Fifty-four subjects with knee
OA (mean SD age, 65.6 7.6 years) and 27
similarly aged control subjects (age, 64.2 5.1
years) were recruited for this study. Isometric knee
flexion and extension strength were measured,
and electromyographic data were recorded, from
OA: muscle atrophy, failure of voluntary
muscle activity, and apparent weakness from increased antagonist muscle
cocontraction.5
7 muscles crossing the knee and used to calculate
cocontraction ratios during maximal effort knee
flexion and extension trials. The burst superimposition technique was used to measure failure of
voluntary activation.
TH;IKBJI0 Knee extension strength of subjects
with knee OA (mean SD, 115.9 6.7 Nm) was
significantly lower than for those in the control
group (152.3 9.6 Nm). No significant betweengroup difference was found for failure of voluntary
muscle activation, or the cocontraction ratios during maximum effort knee flexion or extension.
T9ED9BKI?ED0 These results demonstrate that
the reduction in isometric extension strength, measured with a 90° knee flexion angle, in subjects
with knee OA is not associated with increased antagonist cocontraction. J Orthop Sports Phys Ther
2009;39(11):807-815. doi:10.2519/jospt.2009.3079
TA;OMEH:I0 burst superimposition, OA, quadriceps strength, voluntary muscle activation
Decreases in muscle cross-sectional
area (CSA) have been established in subjects with early OA degenerative changes22 and in those with severe knee OA.10
Ikeda et al22 found that women with early
degenerative changes in the knee joint
had reductions in quadriceps CSA of
up to 12%, compared with age-matched
women without radiographic changes.
Fink et al10 found atrophy of type 2 fibers
and type 1 fibers of the vastus medialis
in patients undergoing total joint arthroplasty. Importantly, a smaller lean-muscle CSA in the diseased limb, compared
to the contralateral limb, in subjects with
OA contributes to muscle weakness but is
not the only factor involved.40
Several investigators have examined
failure of voluntary muscle activity as
a contributing factor in reduced knee
extension strength.12,29,33,40,48 Failure of
voluntary activity is the measure of an
individual’s ability to fully activate their
muscles during maximal voluntary contractions.20 Reductions in muscle activation are due to either an inability to
recruit all motor units or a reduction in
the motor unit discharge rate.25 In healthy
adults, simulated failure of voluntary activity by saline injection into the knee
joint to cause effusion resulted in decreased isometric extension strength and
knee extension torque during strength
testing23,31 and increased muscle cocon-
1
Research Associate, School of Sport Science, Exercise and Health, The University of Western Australia, Crawley, Western Australia, Australia. 2 Associate Professor, School of
Sport Science, Exercise and Health, The University of Western Australia, Crawley, Western Australia, Australia. 3 Professor, School of Sport Science, Exercise and Health, The
University of Western Australia, Crawley, Western Australia, Australia. The research was given ethical approval from the University of Western Australia Human Research Ethics
Committee. Address correspondence to Dr David Lloyd, School of Sports Science, Exercise and Health, The University of Western Australia, 35 Stirling Highway, Crawley, Western
Australia 6009, Australia. E-mail: [email protected] or [email protected]
journal of orthopaedic & sports physical therapy | volume 39 | number 11 | november 2009 |
807
[
traction during gait.52 As such, failure of
voluntary activity is thought to contribute to disability and has been proposed
as an important element in the pathogenesis and further progression of knee
OA.21,41,44
The results of investigations examining failure of voluntary muscle activity in subjects with knee OA have been
somewhat mixed, depending on disease
severity and the measurement methods
used. Two methods have been employed
to assess failure of voluntary activity in
subjects with knee OA: the burst superimposition technique and the twitch interpolation technique. In these measures,
electrical stimuli are superimposed onto
a volitionally contracted muscle group,
thereby recruiting any inactive motor
units and subsequently generating additional force. Each method uses a different
type of stimulus. The twitch interpolation
technique uses a single pulse on various
levels of muscle contractions, which is
calculated as 1 – (superimposed twitch
force at maximum voluntary isometric
contraction ÷ twitch force at rest) and
reported as a percentage of voluntary
activation.43 The burst superimposition
technique uses a train of stimuli and is
calculated as the ratio between the maximal volitional effort prior to stimulus and
the total force produced with the stimulus, termed the central activation ratio
(CAR).25 A CAR of 1.0 implies that the
muscle is fully activated through voluntary contraction, with no additional force
obtained from the stimulus. Although
both methods have been used in subjects
with knee OA, the burst superimposition
technique is the recommended technique
during maximal contractions,38,45 due to
trains of stimuli having greater sensitivity
to activation deficiencies than superimposed single or double stimuli.25
A loss of knee extension and/or flexion strength, often interpreted as muscle
weakness, may be a consequence of increased antagonist muscle activation. 9
Knee muscle cocontraction is thought to
occur to stabilize joints24 and decrease
anterior/posterior translation.26 For
RESEARCH REPORT
example, during isometric knee extension strength tests, the increases in joint
torque production coincide with the increases in antagonist (hamstring) muscle
activity, thereby generating a flexion moment that counteracts the extension moment generated by the quadriceps. This
hamstring antagonist cocontraction appears as a loss of knee extension strength.
In healthy adults (20-30 years of age),
hamstring antagonist cocontraction
during maximum isometric extension is
between 5% to 10% of maximum activation.5,9 Others have found that elderly
female subjects (mean age, 69.5 years)
have approximately 15% more hamstring antagonist cocontraction during
maximal isometric knee extension trials
than younger female subjects (mean age,
22.8 years).30 So the aging process alone
results in greater levels of hamstring antagonist cocontraction. Subsequently,
subjects with knee OA who tend to be
older must be compared with similarly
aged, unaffected people.
Increased agonist-antagonist cocontraction levels have been reported in
subjects with knee OA while walking17
and performing other activities of daily
living.16 During activities of daily living, subjects with knee OA have greater
hamstring antagonist cocontraction than
their healthy counterparts, most likely in
compensation for weak quadriceps or because of pain.16 Therefore, it is possible
that they use this mechanism to protect
the knee joint and minimize pain during isometric strength testing, which in
turn would decrease the measured knee
extension and flexion strength, giving the
appearance of weakness. Lewek et al29
observed low knee extension strength
in subjects with knee OA and, by using
the burst superimposition technique,
showed that failure of voluntary activity was not exhibited in their cohort.
They then suggested that the apparent
decrease in knee extension strength was
due to muscle atrophy. However, they did
not examine the possibility for hamstring
antagonist cocontraction. To date, the
level of antagonist cocontraction during
]
maximum voluntary isometric contractions (MVICs) has not been assessed
in subjects with knee OA, especially in
regard to low isometric knee extension
strength. Antagonist cocontraction could
be a factor contributing to apparent knee
extension and flexion weakness reported
in this population.
The purpose of this study was to examine knee extension and flexion strength,
failure of voluntary muscle activation for
knee extension, and antagonist cocontraction during maximal effort isometric
knee flexion and extension in subjects
with knee OA. Consistent with what has
been reported previously, we hypothesized that subjects with knee OA would
have reduced isometric knee extensor
and knee flexor strength compared with
age-matched asymptomatic controls. We
also hypothesized that subjects with knee
OA would not display greater failure of
voluntary activity during maximal voluntary isometric knee extension compared
to same-aged controls. However, we hypothesized that subjects with knee OA
would exhibit higher levels of antagonist
cocontraction during maximum effort
isometric knee flexion and extension trials than the age-matched controls.
C;J>E:I
F
ifty-four subjects diagnosed
with knee OA (30 female) were recruited through public advertisement and local orthopaedic outpatient
clinics. These subjects had radiographic
signs of knee OA, body mass index less
than 35 kg/m2, morning knee stiffness
less than 30 minutes, could walk unassisted, and had not received steroid injections in the previous 6 months or regular
physiotherapy in the previous 12 months.
Subjects with knee OA were required to
fill in the Knee Osteoarthritis Outcomes
Survey (KOOS)42 as part of the inclusion
criteria and were included based on an
average score of less than 70 for each of
4 subscales: pain, symptom, activities of
daily living, and quality of life. Difficulties
associated with responses to the KOOS
808 | november 2009 | volume 39 | number 11 | journal of orthopaedic & sports physical therapy
have previously been used as inclusion
criteria when examining subjects with
knee OA.35 Twenty-seven (18 female)
asymptomatic controls, in the same age
range as those with knee OA, were enlisted through local newspaper advertisements and university mailing lists. People
diagnosed with any form of arthritis were
excluded from the control group. Both
those with knee OA and age-matched
controls were screened and excluded
if they suffered from any neuromuscular disease, cardiovascular disorder, or
recent surgery or injury (within last 2
years) to the back or lower extremities.
All test procedures were approved by the
University of Western Australia Human
Research Ethics Committee, the rights
of all subjects were protected, and subjects gave their informed, written consent
prior to testing.
Electromyographic (EMG) data were
recorded using a 16-channel EMG system
(Delsys, Boston, MA), with custom-made,
in-lead, double-differential preamplifiers
(input impedance, 1 M8; bandwidth,
20-450 Hz; common-mode rejection ratio, 140). After cleaning and abrasion of
the skin, ClearTrace Ag/AgCi disposable
circular (25-mm diameter) surface EMG
electrodes (ConMed Corporation, Utica,
NY) were applied with an interelectrode
distance of 30 mm, in line with the muscle
fibers over the midmuscle bellies. The following muscles of the test limb were tested: the quadriceps (rectus femoris, vastus
lateralis, vastus medialis), the hamstrings
(biceps femoris and semimembranosus),
and the gastrocnemius (medial gastrocnemius and lateral gastrocnemius). The
test limb was the most painful or functionally limited limb in subjects with knee
OA, and for the controls, their test limb
was selected to ensure matching with
the body mass and height of the subjects
with knee OA. A reference electrode was
placed over the head of the fibula. The
raw EMG data were checked for artifacts
after placement of electrodes and prior to
data collection.
Quadriceps, hamstrings, and gastrocnemius strength, and failure of voluntary
muscle activation of the test limb were
measured isometrically using an isokinetic dynamometer (Biodex Medical
Systems, Inc, Shirley, NY). Subjects were
seated with a hip angle of 90° and a knee
angle of 90° for the knee strength testing.
During the gastrocnemius testing, the
knee angle was set at 65°. The subjects
were secured with Velcro straps to reduce
movement of the body and test limb. The
knee joint was visually aligned with the
dynamometer axis of movement and the
device locked at 90° of knee flexion for
quadriceps and hamstring testing. Following familiarization and warm-up trials, the subject underwent 3 MVICs for
each of the quadriceps, hamstrings, and
gastrocnemius muscles. Subjects were
instructed to maximally extend the knee
for quadriceps MVIC testing, maximally
flex the knee for hamstrings MVIC testing, and flex the knee while maximally
plantar flexing the ankle for gastrocnemius MVIC trials. A computer provided
real-time feedback and displayed a visual
target line for each trial. The visual target
line was set just above each subject’s maximum force attained during the warm-up
trials and was adjusted as needed to ensure a maximum effort was undertaken.
To motivate the subject, verbal encouragement was provided during each trial.
EMG and torque data were sampled simultaneously at 2000 Hz. Subjects were
given 2 minutes rest between each trial,
to prevent fatigue. EMG data were normalized to the maximum value obtained
for each muscle during the agonist MVIC
trial. The trial with the strongest recorded torque for each muscle group was used
in determining stimulus amplitude and
target line values for the failure of voluntary activation testing described below.
The muscle cocontraction values were
calculated during the maximum torque
trial to avoid stimulus artifact that would
occur during the burst superimposition
testing, affecting the EMG signal. The
cocontraction was calculated across a
400-millisecond period, when the maximum torque was achieved and when the
force was stable.
We chose to use hip and knee angles
of 90° during knee strength testing for 4
reasons. Firstly, we did not want to expose the subjects with knee OA to too
many strength tests, as well as electrical stimulation tests, over long periods.
Secondly, even though the greatest knee
extension strength occurs between 60°
and 70°, with hip at 90°, the greatest
maximum voluntary quadriceps activation is possible at knee and hip angles of
90°.8 Thirdly, the hamstrings tend to have
the greatest activation with knee and hips
angles at 90°.34 Finally, and most importantly, we wanted to utilize the same
angle used by Lewek et al29 to examine
the role of cocontraction in the possible
absence of failure of voluntary activity.
To test failure of voluntary muscle activity, 2 aluminium-plate electrodes (70
50 mm), covered with saline-soaked
pads, were attached proximally over the
vastus lateralis and distally over the vastus medialis. Stimulation was performed
by a Grass S88 square pulse stimulator,
with a Grass Model SIU5A stimulus
isolation unit (Grass Instruments, West
Warwick, RI). The amplitude of the stimulation was adjusted to produce a torque
equivalent to 25% of MVIC when applied
to the resting muscle of each subject. A
response equivalent to 20% to 25%
maximum torque has been suggested
adequate to measure failure of voluntary
muscle activity.15,21 Initial stimuli began at
low amplitudes and were increased over
3 or 4 trials until a sufficient response
(20% to 25% of maximum torque) was
elicited. Following stimulus familiarization and amplitude determination, the
subjects were instructed to undertake
an MVIC and aim for a visual target line
on the computer screen in front of them.
The visual target was adjusted between
trials as needed, to ensure that subjects
were continuing to work at a maximum
effort in each trial. During the MVICs,
the stimulus (amplitude, 100-150 V;
pulse duration, 600 microseconds; pulse
interval, 10 milliseconds; train duration,
100 milliseconds) was delivered automatically by computer 350 milliseconds after
journal of orthopaedic & sports physical therapy | volume 39 | number 11 | november 2009 |
809
[
the visual target line had been reached.
The 350-millisecond interval was selected to ensure the likelihood of applying the stimuli when the subjects’ MVIC
torques were stable. This is particularly
important for the subjects with knee OA,
as they may not have been able to hold an
MVIC for any longer durations.2 The trial
was not used if a subject did not reach the
visual target line. Each subject completed
3 extension MVIC trials with the stimulus
superimposed over the quadriceps. CAR
values were used to express the level of
inhibition and were calculated by dividing the subject’s voluntary torque immediately prior to stimulation by the torque
value with the stimulus applied.25
Raw EMG and torque data were processed using custom-written software
(Matlab, Version R2007a; The MathWorks, Inc, Natick, MA). Raw EMG data
were full-wave rectified and filtered with
a second-order, low-pass Butterworth,
with a 6-Hz cutoff to create linear envelopes. EMG data were normalized for
amplitude to maximum EMG values
calculated from the Biodex trials. Raw
torque data were filtered using a secondorder, low-pass Butterworth filter with a
2-Hz cutoff. Knee extension and flexion
strength were taken from the trial with
the greatest measured torque.
Cocontraction has 2 components: the
ratio of agonist and antagonist activation,
and the total amount of activation of the
agonists and antagonists. Therefore, we
constructed 2 variables—cocontraction
ratios (CCR) and net activation—that
were examined during the maximum
knee flexion and extension trials. CCRs
used in this experiment were the ratio of
hamstring (semimembranosus, biceps
femoris) to quadriceps (vastus lateralis,
vastus medialis, rectus femoris), and
the ratio of flexion (semimembranosus,
biceps femoris, medial gastrocnemius,
lateral gastrocnemius) to extension
(vastus lateralis, vastus medialis, rectus
femoris). These ratios differed in that the
hamstring-quadriceps CCR used only 2
flexors (semimembranosus and biceps
femoris), while the flexion-extension CCR
RESEARCH REPORT
J78B;'
Age (y)
Height (m)
]
Demographics (Mean p SD) and
t Test Comparison of Subjects With
Knee OA and Controls
E7 9edjhebi†
tIYeh[
65.6 (7.6)
64.2 (5.1)
0.82
.42
–0.12
.90
1.70 (0.09)
1.70 (0.09)
PLWbk[
Body Mass (kg)
81.4 (14.2)
71.3 (13.8)
3.06
.01
BMI (kg/m2)
28.1 (4.2)
24.4 (3.6)
3.88
.01
KOOS Pain (0-100)‡
57.6 (18.2)
96.2 (6.2)
14.04
.01
KOOS symptom (0-100)‡
53.6 (17.6)
93.7 (8.1)
14.04
.01
KOOS ADL (0-100)‡
60.0 (20.0)
96.6 (5.7)
12.46
.01
KOOS QOL (0-100)‡
33.7 (15.8)
88.4 (15.3)
14.92
.01
Abbreviations: ADL, activities of daily living; BMI, body mass index; KOOS, Knee Osteoarthritis
Outcomes Survey; OA, osteoarthritis; QOL, quality of life.
* n = 54 ( female, n =30).
†
n = 27 ( female, n = 18).
‡
Lower KOOS scores indicate greater functional declines and increase levels of pain and symptoms.
included the medial gastrocnemius and
lateral gastrocnemius. These ratios were
calculated as follows: if agonist mean
EMG antagonist mean EMG, CCR = 1
– (antagonist mean EMG ÷ agonist mean
EMG); else, CCR = (agonist mean EMG ÷
antagonist mean EMG) – 1.
In these equations, if agonists were
more active than the antagonists, the
CCR would be above zero, and vice versa.
Maximum cocontraction would be represented with a CCR equal to zero, while
a minimum cocontraction is indicated
with a CCR of 1 or –1. In addition to the
CCR, net muscle activation values (ie, the
sum of all agonist and antagonist activity) were calculated from the normalized
EMG data for each flexion and extension
trial.
Between-group differences for age,
height, body mass, and KOOS subscales
were examined using t tests. Isometric knee flexion and extension torque
were examined between groups, using
an analysis of covariance (ANCOVA) to
account for the effect of body mass that
was entered as the covariate. Regression
analysis between body mass and torque
showed equivalent slopes between the
groups for extension (P = .691) and flexion (P = .437) tests; thus the assumption
of homogeneity of slopes was not violated.
The covariate and dependant variables
were significantly correlated (P.001)
and scatter plots with regression lines
fitted showed the existence of nonzero y
intercepts, indicating the need for the use
of the ANCOVA. Because body mass and
muscle activation data were not significantly correlated, an analysis of variance
(ANOVA) was sufficient to compare between group differences on the remaining
dependant variables (flexion-extension
CCR, hamstring-quadriceps CCR, net
muscle activation for flexion and extension strength, and failure of voluntary
activity). All statistical analyses were
done using SPSS for Windows, Version
12.0 (SPSS, Inc, Chicago, IL). The analysis of 6 dependant variables required
multiple statistical tests. A Bonferonni
adjustment (0.05/6) resulted in a value
of 0.008; therefore, a slightly higher
significance level of P.01 was used, as
Bonferonni adjustment is known to be
conservative.49
H;IKBJI
A
ge and height did not differ
significantly between the groups,
but the control subjects were 12.5%
lighter than the subjects with knee OA
(J78B; '). KOOS scores revealed significant differences between the groups on
all 4 subscales (J78B;'). These values are
810 | november 2009 | volume 39 | number 11 | journal of orthopaedic & sports physical therapy
180
*
160
J78B;(
140
Torque (Nm)
120
100
80
Control and Subjects With Knee OA
Cocontraction and Net Muscle Activity
ANOVA Results During Isometric Knee
Extension and Flexion Trials
9edjhebi Ad[[E7IkX`[Yji <LWbk[
PLWbk[
.223
Knee extension trials (mean SD)
60
Hamstring-quadriceps CCR
0.74 0.17
0.78 0.12
1.510
40
Flexion-extension CCR
0.77 0.15
0.78 0.12
0.354
.554
20
Net activation
2.01 0.37
1.97 0.31
0.272
.603
Hamstring-quadriceps CCR
0.89 0.13
0.91 0.05
0.926
.339
Flexion-extension CCR
0.77 0.15
0.90 0.58
0.885
.350
Net activation
2.30 0.54
2.12 0.34
3.336
.072
Knee flexion trials (mean SD)
0
Extension
Control
Flexion
OA
<?=KH;'$Adjusted (body mass) mean and standard
error values for maximum knee extension and flexion
strengths of the controls and subjects with knee OA.
*Significant at P.01.
similar to those previously reported for individuals with moderate OA (grade 2-3).18
Gender ratios were different between the
control and OA groups. An analysis examining the effect of gender and group on
each of the dependant variables showed
nonsignificant interactions for gender
(P.05); thus the observed differences
were not due to the affect of gender.
Using body mass as a covariate, the
knee extension strength of the subjects
with knee OA was significantly less
(P.01) than that of the control group
(<?=KH;'). Knee flexion strength did not
differ between the knee OA and control
groups (P = .026). The covariate of body
mass was significantly associated with
knee extension and knee flexion strength
(P.01).
Mean SD quadriceps CAR values
were 0.942 0.041 and 0.952 0.033
for the knee OA and control groups, respectively, and no significant difference
existed between the groups (P = .267).
There were no significant differences
between the control subjects and subjects with knee OA in the hamstringquadriceps CCR, flexion-extension CCR,
or net muscle activation measured in either the isometric knee flexion or extension trials (J78B; (). Individual muscle
activation during the flexion and extension MVICs (<?=KH;() revealed no differ-
Abbreviations: ANOVA, analysis of variance; CCR, cocontraction ratio; net activation, activity summation of all muscles; OA, osteoarthritis.
* Values are mean SD.
ences between the controls and subjects
with OA.
:?I9KII?ED
A
s hypothesized, and as shown
previously,11,44 the maximum knee
extension strength of subjects with
knee OA was less than that of control
subjects. The knee flexion strength, although reduced, was not significantly
lower in this group of subjects with knee
OA, and there was no evidence to suggest
a failure of voluntary quadriceps activity. In addition, our hypothesis that the
magnitude of antagonist cocontraction
during maximum effort isometric knee
flexion and extension trials would differ
from controls was not supported.
Previous studies have shown a large
variation in the percentage deficit in knee
extension strength of subjects with knee
OA compared with healthy adults (26%45%).6,28,50 In individuals with moderate
to severe knee OA, these strength deficits
are related to functional declines and the
amount of voluntary quadriceps activation failure.12 Our 23% reduction in knee
extension strength is smaller than that
previously reported, but still substantially
different to controls. The difference between our knee extension strength findings and those previously reported could
be due to differences in the testing meth-
ods used, the reporting of torque values,
or alterations in the muscle moment arms
of the subjects with knee OA.
Methodologically, some previous studies have had insufficient rest between
trials,6,15,21,50 no visual feedback or verbal encouragement,6,15,37,50 no warm-up
trials,13,15,50 and have used different knee
angles than those used in this investigation.33,48 Additionally, differences could be
caused by different units of measure4 and
normalization procedures.53 The units of
measurement for strength should reflect
muscle generated torque (Nm), which is
a product of force and the lever arm distance (m).4 Strength measurements, within the knee OA literature, have, in some
cases, reported only the applied force (N),
thus neglecting limb length differences
between subjects.13,32,40 Another concern
with some studies is the use of “ratio normalization”7 of torque by simply dividing
the torque measure by body mass3,6,13 or
body mass index (BMI).29,32,48 This ratio
normalization7 fails to eliminate the confounding effects of a subject’s body size
and structure on the peak torque values
for 2 reasons: it assumes (1) a linear relationship (in this case, one between peak
torque and BMI or body mass), and (2)
a y intercept of zero, which is rarely the
case in biological experiments.51 ANCOVA
has been effective for treating body mass
as a confounding factor of knee strength
journal of orthopaedic & sports physical therapy | volume 39 | number 11 | november 2009 |
811
[
RESEARCH REPORT
A
0.90
0.80
Muscle Activation (V)
0.70
0.60
0.50
0.40
0.30
0.20
0.10
0.00
B
0.90
0.80
Muscle Activation (V)
0.70
0.60
0.50
0.40
0.30
0.20
0.10
0.00
SM
BF
RF
VM
VL
MG
LG
Muscle
Control
OA subjects
<?=KH;($Mean of individual muscle EMG data during (A) maximum extension trial and (B) maximum flexion trial.
Abbreviations: BF, biceps femoris; LG, lateral gastrocnemius; MG, medial gastrocnemius; RF, rectus femoris; SM,
semimembranosus; VL, vastus lateralis; VM, vastus medialis.
in unaffected subjects and subjects with
knee joint pathologies.49 This is because
it removes the linear effect of the covariate on torque and does not assume a zero
y-intercept,53 making it the ideal solution
in statistical tests of torque data.
Smaller muscle moment arms in those
with knee OA could affect the maximum
extension moment. However, there is no
evidence to suggest that subjects with
knee OA have changes in, or smaller,
knee flexion and extension moment arms.
In addition, the subjects in this study did
not have varus or valgus deformity, suggesting that the knee joint may not have
been dramatically different from that
of the controls. Nevertheless, the lower
strength in subjects with knee OA may
still be due to lower knee flexion and extension moment arms and needs to be
the subject of future research.
Failure of voluntary muscle activity
was not responsible for the measured
reductions in knee extension strength
]
between subjects with knee OA and controls in this study. We chose to use the
burst superimposition technique due to
its recommendation for use when undertaking maximal strength trials. 38,45
Studies of muscle activation in individuals with knee OA using this method have
provided mixed results, possibly due to
methodological, disease status, and participant age differences. Three studies of
muscle inhibition in subjects with endstage knee OA, using the burst superimposition technique, have reported CAR
values substantially less than the 0.94 to
0.95 values considered normal for healthy
older adults.47 Stevens et al48 found a
CAR value of 0.85 for the involved limb
of subjects with knee OA, but this was
not significantly different to the subject’s
uninvolved limb. Mizner et al32 found a
mean preoperative CAR value of 0.87 in
subjects with knee OA but did not compare their data to a control group. Petterson et al40 reported CAR values of 0.76 in
their knee OA group; however, this value
was determined using the methods of
Stackhouse et al,46 where the curvilinear
relationship between CAR and maximal
voluntary contraction are modeled using a submaximal contraction, making a
comparison between their results and the
present study impossible. The disparity
between these previous studies and the
results of the current investigation may
be due to the differences in disease severity of the subject groups. However, alternative reasons could be due to differences
in the stimulus amplitude and the knee
flexion angle used during testing.
It is difficult to compare the precise
values of voluntary activation failure between studies due to differences in the
stimulus parameters used.12 Studies that
examined voluntary activation failure in
subjects with knee OA, using the burst
superimposition technique, have failed
to clarify how their voltage level was determined for testing and have used the
same voltage for each subject, despite
possible differences in body composition.12,29,33 Interestingly, Stevens et al48 did
not report the voltage used during their
812 | november 2009 | volume 39 | number 11 | journal of orthopaedic & sports physical therapy
testing. Based on differences in subject
response to the stimulus due to muscle
mass and adipose tissue, we believe that
determining the appropriate voltage for
each individual was necessary to ensure
that all inhibited motoneurons were activated when the stimulus was applied.
Additionally, given that our subjects had
CAR of 0.95, a stimulus producing 25%
of maximum torque from “at rest” would
have been more than sufficient to maximally stimulate the muscles.
Differences in knee flexion angles used
in studies of voluntary activation failure
could be another reason for the differences found among studies. Several studies
of voluntary quadriceps activation failure
have used a knee flexion angle of 75°, due
to reductions in the knee flexion range of
their participants,32,40,48 compared to the
90° knee flexion angle used in the current
investigation. Failure of voluntary muscle
activity measured at knee flexion angles
of less than 80° has been shown in subjects who were normal, with quadriceps
activation levels of 85% compared to 95%
when tested at knee flexion angles greater
than 80°.27 This may account for the differences between the current investigation and these aforementioned studies of
subjects with knee OA.32,40,48 Indeed, our
study is similar to the study by Lewek
et al,29 enrolling subjects with moderate
knee OA and using a 90° knee flexion angle with the burst superimposition technique. This may explain the agreement
of our results (CAR, 0.94) with those
of Lewek et al,29 who also failed to find
muscle inhibition in a group of subjects
with knee OA (CAR, 0.93) compared
with controls. Importantly to the current
study, Lewek et al29 then surmised that
the lower knee extension strength was
due to muscle atrophy. However, they did
not examine the possibility of hamstring
cocontraction reducing the apparent isometric knee extension strength.
To our knowledge, this is the first examination of antagonist cocontraction
during maximum isometric flexion and
extension trials in a knee OA population. In this study, antagonist hamstring
cocontraction during maximal isometric
extension did not differ between controls and subjects with knee OA (22%26%), nor were there any differences in
the net muscle activation level between
groups. These hamstring cocontraction
values were higher than those recorded
for healthy adults in studies using identical muscle groups (9%-13%).5,14 This is
not surprising, as antagonist cocontraction of the hamstrings during maximum
isometric knee extension has been shown
to be greater in older women (40%) as
compared to young women (20%).30 Our
cocontraction values are much lower than
those previously found for healthy older
women.30 This difference in cocontraction level between the current investigation and that of Macaluso et al30 may be
due to our use of biceps femoris, medial
hamstrings, and 3 quadriceps muscles,
where they only used biceps femoris and
vastus lateralis to examine antagonist cocontraction. In isokinetic knee extension,
the antagonist cocontraction of the biceps
femoris, a more lateral hamstring muscle,
has been found to have greater values
than that measured from the medial hamstring.1 If this variation in medial and lateral hamstring muscle activity also applies
to an isometric contraction, then the use
or inclusion of other flexor and extensor
muscles in the calculation of cocontraction, could account for these differences.
Antagonist quadriceps activity, in the knee
flexion trials in this study, did not differ
between controls and subjects with knee
OA and does not differ with age.30
The subjects with knee OA in the
present study demonstrated reduced
knee extensor strength in the absence
of voluntary muscle activation failure
and hamstring antagonist cocontraction.
These results suggest that muscle atrophy
was the most likely cause of quadriceps
weakness. However, the knee extension
strength of the subjects with knee OA in
this study may have been better than in
some studies.28,36 A lack of testing and
reporting standards makes comparisons
between studies difficult. Given the moderate severity level of our knee OA cohort,
large declines in strength and high levels
of inhibition were expected. It is possible
that a group of subjects with knee OA,
with even greater decline in quadriceps
strength compared to controls, could
exhibit both failure of voluntary muscle
activation and increased levels of antagonist cocontraction; but this has not been
measured. Alternatively, subjects with
greater knee varus or valgus deformity
may exhibit different levels of failure of
voluntary activity and/or cocontraction.
The lack of difference in failure of voluntary muscle activation and antagonist
cocontraction between the controls and
subjects with knee OA suggests that the
knee joint deterioration does not alter
the level of muscle activation needed in
an isometric task. It is important to note
that these results may only be applicable
to this task, and that in more functional
tasks, antagonist cocontraction, failure
of voluntary activation, and muscle atrophy may all contribute to knee extension
weakness.
There are several limitations of this
study. Firstly, although patients were
physician diagnosed or waiting for knee
replacement surgery, we were unable to
obtain radiographic data for OA grading. However, the self-perceived data
obtained from the KOOS was similar
to that of subjects with moderate OA.18
Secondly, the collection of EMG data and
the calculation of cocontraction ratios
during the MVIC trials and not during
the stimulation trials could have affected
the results; but the protocols within each
trial were the same (visual target line and
verbal encouragement, and sufficient rest
between trials to provide consistency in
the testing). The rationale behind using
the MVIC trials for the collection of cocontraction data was to ensure that EMG
data were adequately sampled while the
subject was producing a steady MVIC.
Because antagonist cocontraction is altered in the gait patterns of subjects with
knee OA,17 it is thought that a more dynamic isokinetic test may reveal different
results. Due to the differences shown in
voluntary muscle activation failure based
journal of orthopaedic & sports physical therapy | volume 39 | number 11 | november 2009 |
813
[
on the knee flexion angle, it is possible
that antagonist cocontraction levels may
also differ between subjects with knee OA
and controls at knee flexion angles other
than 90°. Clearly, further investigation is
required to examine the levels of antagonist cocontraction in subjects with knee
OA at different knee angles and during
more dynamic tasks such as isokinetic
strength testing.
9ED9BKI?ED
T
his study revealed no differences between the subjects with
OA and control subjects in muscle
cocontraction and failure of voluntary
quadriceps muscle activation during
maximal isometric knee extension trials.
These findings suggest that altered muscle cocontraction and failure of voluntary
quadriceps muscle activation are not associated with the reductions in isometric
knee extension torque exhibited in this
group of subjects with knee OA.T
A;OFE?DJI
<?D:?D=I0 Reductions in isometric knee
extension strength in subjects with knee
OA are not due to quadriceps muscle
inhibition. Further, antagonist cocontraction is not increased in subjects
with knee OA and is not thought to contribute to the isometric knee extension
weakness.
?CFB?97J?EDI0 The findings of this investigation highlight the need for pure
muscle-strengthening exercises in individuals with knee OA.
97KJ?ED0 These results may be specific
to this group of subjects with knee OA.
Additionally, increases in antagonist
cocontraction could potentially occur
during more dynamic strength testing
or activities.
ACKNOWLEDGEMENTS: This work was made
possible by funding from the University of
Western Australia. Special thanks to Drs
Riaz Khan and Sean Williams for allowing
us to recruit from their orthopaedic outpatient clinics.
RESEARCH REPORT
H;<;H;D9;I
1. Aagaard P, Simonsen EB, Andersen JL, Magnusson SP, Bojsen-Moller F, Dyhre-Poulsen P.
Antagonist muscle coactivation during isokinetic knee extension. Scand J Med Sci Sports.
2000;10:58-67.
2. Allen GM, Gandevia SC, McKenzie DK. Reliability of measurements of muscle strength and
voluntary activation using twitch interpolation.
Muscle Nerve. 1995;18:593-600. http://dx.doi.
org/10.1002/mus.880180605
3. Bennell KL, Hinman RS, Metcalf BR, et al. Efficacy of physiotherapy management of knee
joint osteoarthritis: a randomised, double
blind, placebo controlled trial. Ann Rheum Dis.
2005;64:906-912. http://dx.doi.org/10.1136/
ard.2004.026526
4. Bennell KL, Hunt MA, Wrigley TV, Lim BW, Hinman RS. Role of muscle in the genesis and
management of knee osteoarthritis. Rheum Dis
Clin North Am. 2008;34:731-754. http://dx.doi.
org/10.1016/j.rdc.2008.05.005
5. Busse ME, Wiles CM, van Deursen RW. Co-activation: its association with weakness and specific neurological pathology. J Neuroeng Rehabil.
2006;3:26. http://dx.doi.org/10.1186/17430003-3-26
6. Cheing GL, Hui-Chan CW. The motor dysfunction of patients with knee osteoarthritis
in a Chinese population. Arthritis Rheum.
2001;45:62-68. http://dx.doi.org/10.1002/15290131(200102)45:1<62::AID-ANR85>3.0.CO;2-W
7. Davies MJ, Dalsky GP. Normalizing strength for
body size differences in older adults. Med Sci
Sports Exerc. 1997;29:713-717.
8. de Ruiter CJ, Kooistra RD, Paalman MI, de Haan
A. Initial phase of maximal voluntary and electrically stimulated knee extension torque development at different knee angles. J Appl Physiol.
2004;97:1693-1701. http://dx.doi.org/10.1152/
japplphysiol.00230.2004
9. Ebenbichler GR, Kollmitzer J, Glockler L, Bochdansky T, Kopf A, Fialka V. The role of the
biarticular agonist and cocontracting antagonist
pair in isometric muscle fatigue. Muscle Nerve.
1998;21:1706-1713.
10. Fink B, Egl M, Singer J, Fuerst M, Bubenheim
M, Neuen-Jacob E. Morphologic changes in
the vastus medialis muscle in patients with
osteoarthritis of the knee. Arthritis Rheum.
2007;56:3626-3633. http://dx.doi.org/10.1002/
art.22960
11. Fisher NM, Gresham G, Pendergast DR. Effects
of a quantitative progressive rehabilitation program applied unilaterally to the osteoarthritic
knee. Arch Phys Med Rehabil. 1993;74:13191326.
12. Fitzgerald GK, Piva SR, Irrgang JJ, Bouzubar
F, Starz TW. Quadriceps activation failure as a
moderator of the relationship between quadriceps strength and physical function in individuals with knee osteoarthritis. Arthritis Rheum.
]
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
2004;51:40-48. http://dx.doi.org/10.1002/
art.20084
Gapeyeva H, Buht N, Peterson K, Ereline J,
Haviko T, Paasuke M. Quadriceps femoris
muscle voluntary isometric force production
and relaxation characteristics before and 6
months after unilateral total knee arthroplasty in
women. Knee Surg Sports Traumatol Arthrosc.
2007;15:202-211. http://dx.doi.org/10.1007/
s00167-006-0166-y
Grabiner MD, Owings TM. EMG differences
between concentric and eccentric maximum
voluntary contractions are evident prior to
movement onset. Exp Brain Res. 2002;145:505511. http://dx.doi.org/10.1007/s00221-002-11292
Hassan BS, Mockett S, Doherty M. Static postural sway, proprioception, and maximal voluntary
quadriceps contraction in patients with knee
osteoarthritis and normal control subjects. Ann
Rheum Dis. 2001;60:612-618.
Hortobagyi T, Westerkamp L, Beam S, et al. Altered hamstring-quadriceps muscle balance in
patients with knee osteoarthritis. Clin Biomech
(Bristol, Avon). 2005;20:97-104. http://dx.doi.
org/10.1016/j.clinbiomech.2004.08.004
Hubley-Kozey C, Deluzio K, Dunbar M. Muscle
co-activation patterns during walking in those
with severe knee osteoarthritis. Clin Biomech
(Bristol, Avon). 2008;23:71-80. http://dx.doi.
org/10.1016/j.clinbiomech.2007.08.019
Hubley-Kozey CL, Deluzio KJ, Landry SC,
McNutt JS, Stanish WD. Neuromuscular alterations during walking in persons with moderate
knee osteoarthritis. J Electromyogr Kinesiol.
2006;16:365-378. http://dx.doi.org/10.1016/j.
jelekin.2005.07.014
Hurley MV. The role of muscle weakness in the
pathogenesis of osteoarthritis. Rheum Dis Clin
North Am. 1999;25:283-298, vi.
Hurley MV, Newham DJ. The influence of arthrogenous muscle inhibition on quadriceps
rehabilitation of patients with early, unilateral osteoarthritic knees. Br J Rheumatol.
1993;32:127-131.
Hurley MV, Scott DL, Rees J, Newham DJ. Sensorimotor changes and functional performance
in patients with knee osteoarthritis. Ann Rheum
Dis. 1997;56:641-648.
Ikeda S, Tsumura H, Torisu T. Age-related
quadriceps-dominant muscle atrophy and incident radiographic knee osteoarthritis. J Orthop
Sci. 2005;10:121-126. http://dx.doi.org/10.1007/
s00776-004-0876-2
Jensen K, Graf BK. The effects of knee effusion
on quadriceps strength and knee intraarticular
pressure. Arthroscopy. 1993;9:52-56.
Kellis E. Quantification of quadriceps and
hamstring antagonist activity. Sports Med.
1998;25:37-62.
Kent-Braun JA, Le Blanc R. Quantitation of
central activation failure during maximal voluntary contractions in humans. Muscle Nerve.
1996;19:861-869. http://dx.doi.org/10.1002/
(SICI)1097-4598(199607)19:7&lt;861::AID-
814 | november 2009 | volume 39 | number 11 | journal of orthopaedic & sports physical therapy
MUS8&gt;3.0.CO;2-7
26. Kingma I, Aalbersberg S, van Dieen JH. Are
hamstrings activated to counteract shear
forces during isometric knee extension efforts
in healthy subjects? J Electromyogr Kinesiol.
2004;14:307-315. http://dx.doi.org/10.1016/j.
jelekin.2004.01.003
27. Kubo K, Tsunoda N, Kanehisa H, Fukunaga T.
Activation of agonist and antagonist muscles at
different joint angles during maximal isometric
efforts. Eur J Appl Physiol. 2004;91:349-352.
http://dx.doi.org/10.1007/s00421-003-1025-x
28. Lankhorst GJ, Van de Stadt RJ, Van der Korst
JK. The relationships of functional capacity,
pain, and isometric and isokinetic torque in osteoarthrosis of the knee. Scand J Rehabil Med.
1985;17:167-172.
29. Lewek MD, Rudolph KS, Snyder-Mackler L.
Quadriceps femoris muscle weakness and
activation failure in patients with symptomatic
knee osteoarthritis. J Orthop Res. 2004;22:110115. http://dx.doi.org/10.1016/S07360266(03)00154-2
30. Macaluso A, Nimmo MA, Foster JE, Cockburn
M, McMillan NC, De Vito G. Contractile muscle
volume and agonist-antagonist coactivation
account for differences in torque between young
and older women. Muscle Nerve. 2002;25:858863. http://dx.doi.org/10.1002/mus.10113
31. McNair PJ, Marshall RN, Maguire K. Swelling
of the knee joint: effects of exercise on quadriceps muscle strength. Arch Phys Med Rehabil.
1996;77:896-899.
32. Mizner RL, Petterson SC, Stevens JE, Vandenborne K, Snyder-Mackler L. Early quadriceps
strength loss after total knee arthroplasty. The
contributions of muscle atrophy and failure
of voluntary muscle activation. J Bone Joint
Surg Am. 2005;87:1047-1053. http://dx.doi.
org/10.2106/JBJS.D.01992
33. Mizner RL, Stevens JE, Snyder-Mackler L. Voluntary activation and decreased force production
of the quadriceps femoris muscle after total
knee arthroplasty. Phys Ther. 2003;83:359-365.
34. Mohamed O, Perry J, Hislop H. Relationship
between wire EMG activity, muscle length, and
torque of the hamstrings. Clin Biomech (Bristol,
Avon). 2002;17:569-579.
35. Mundermann A, Dyrby CO, Hurwitz DE, Sharma
L, Andriacchi TP. Potential strategies to reduce
medial compartment loading in patients with
knee osteoarthritis of varying severity: reduced
walking speed. Arthritis Rheum. 2004;50:11721178. http://dx.doi.org/10.1002/art.20132
36. Nordesjo LO, Nordgren B, Wigren A, Kolstad K.
Isometric strength and endurance in patients
with severe rheumatoid arthritis or osteoarthrosis in the knee joints. A comparative study in
healthy men and women. Scand J Rheumatol.
1983;12:152-156.
37. O'Reilly SC, Jones A, Muir KR, Doherty M.
Quadriceps weakness in knee osteoarthritis: the
effect on pain and disability. Ann Rheum Dis.
1998;57:588-594.
38. Paillard T, Noe F, Passelergue P, Dupui P.
Electrical stimulation superimposed onto
voluntary muscular contraction. Sports Med.
2005;35:951-966.
39. Pap G, Machner A, Awiszus F. Strength and
voluntary activation of the quadriceps femoris
muscle at different severities of osteoarthritic
knee joint damage. J Orthop Res. 2004;22:96103. http://dx.doi.org/10.1016/S07360266(03)00128-1
40. Petterson SC, Barrance P, Buchanan T,
Binder-Macleod S, Snyder-Mackler L. Mechanisms underlying quadriceps weakness in
knee osteoarthritis. Med Sci Sports Exerc.
2008;40:422-427. http://dx.doi.org/10.1249/
MSS.0b013e31815ef285
41. Radin EL, Yang KH, Riegger C, Kish VL, O'Connor
JJ. Relationship between lower limb dynamics
and knee joint pain. J Orthop Res. 1991;9:398405. http://dx.doi.org/10.1002/jor.1100090312
42. Roos EM, Lohmander LS. The Knee injury
and Osteoarthritis Outcome Score (KOOS):
from joint injury to osteoarthritis. Health
Qual Life Outcomes. 2003;1:64. http://dx.doi.
org/10.1186/1477-7525-1-64
43. Shield A, Zhou S. Assessing voluntary muscle
activation with the twitch interpolation technique. Sports Med. 2004;34:253-267.
44. Slemenda C, Brandt KD, Heilman DK, et al.
Quadriceps weakness and osteoarthritis of the
knee. Ann Intern Med. 1997;127:97-104.
45. Stackhouse SK, Dean JC, Lee SC, Binder-
46.
47.
48.
49.
50.
51.
52.
53.
MacLeod SA. Measurement of central
activation failure of the quadriceps femoris
in healthy adults. Muscle Nerve. 2000;23:17061712. http://dx.doi.org/10.1002/10974598(200011)23:11<1706::AIDMUS6>3.0.CO;2-B
Stackhouse SK, Stevens JE, Johnson CD,
Snyder-Mackler L, Binder-Macleod SA. Predictability of maximum voluntary isometric knee
extension force from submaximal contractions
in older adults. Muscle Nerve. 2003;27:40-45.
http://dx.doi.org/10.1002/mus.10278
Stackhouse SK, Stevens JE, Lee SC, Pearce KM,
Snyder-Mackler L, Binder-Macleod SA. Maximum voluntary activation in nonfatigued and
fatigued muscle of young and elderly individuals. Phys Ther. 2001;81:1102-1109.
Stevens JE, Mizner RL, Snyder-Mackler L. Quadriceps strength and volitional activation before
and after total knee arthroplasty for osteoarthritis. J Orthop Res. 2003;21:775-779. http://
dx.doi.org/10.1016/S0736-0266(03)00052-4
Sturnieks DL, Besier TF, Hamer PW, et al. Knee
strength and knee adduction moments following
arthroscopic partial meniscectomy. Med Sci
Sports Exerc. 2008;40:991-997. http://dx.doi.
org/10.1249/MSS.0b013e318167812a
Tan J, Balci N, Sepici V, Gener FA. Isokinetic and
isometric strength in osteoarthrosis of the knee.
A comparative study with healthy women. Am J
Phys Med Rehabil. 1995;74:364-369.
Tanner JM. Fallacy of per-weight and per-surface
area standards, and their relation to spurious
correlation. J Appl Physiol. 1949;2:1-15.
Torry MR, Decker MJ, Viola RW, O'Connor DD,
Steadman JR. Intra-articular knee joint effusion
induces quadriceps avoidance gait patterns.
Clin Biomech (Bristol, Avon). 2000;15:147-159.
Toth MJ, Goran MI, Ades PA, Howard DB, Poehlman ET. Examination of data normalization procedures for expressing peak VO2 data. J Appl
Physiol. 1993;75:2288-2292.
@
CEH;?D<EHC7J?ED
WWW.JOSPT.ORG
FIND Author Instructions & Tools on the Journal’s Website
JOSPT’s instructions to authors are available at www.jospt.org by clicking
“AUTHOR TOOLS & INSTRUCTIONS” in the upper right-hand column of the
home page, or by visiting “INFORMATION FOR AUTHORS”, located in the site’s
navigation bar in the left-hand column. The Journal’s editors have
assembled a list of useful tools and links for authors as well as reviewers.
journal of orthopaedic & sports physical therapy | volume 39 | number 11 | november 2009 |
815