You are on page 1of 9

ARTICLE IN PRESS

Journal of Biomechanics 39 (2006) 426–434
www.elsevier.com/locate/jbiomech
www.JBiomech.com

Effect of muscle fatigue on the compliance of the gastrocnemius
medialis tendon and aponeurosis
Lida Mademlia, Adamantios Arampatzisa,, Mark Walshb
a
Institute for Biomechanics and Orthopaedics, German Sport University Cologne, Carl-Diem- Weg 6, D-50933 Cologne, Germany
b
Department of Physical Education, Health and Sport Studies, Miami University, Oxford, Ohio, USA
Accepted 23 December 2004

Abstract

The aim of the present study was to examine whether or not the compliance of the gastrocnemius medialis (GM) tendon and
aponeurosis is influenced by submaximal fatiguing efforts. Fourteen elderly male subjects performed isometric maximal voluntary
plantarflexion contractions (MVC) on a dynamometer before and after two fatiguing protocols. The protocols consisted of: (1)
submaximal concentric isokinetic contractions (70% isokinetic MVC) at 601/s and (2) a sustained isometric contraction (40%
isometric MVC) until failure to hold the defined moment. Ultrasonography was used to determine the elongation and strain of the
GM tendon and aponeurosis. To account for the axis misalignment between ankle and dynamometer, the kinematics of the leg were
captured at 120 Hz. The maximum moment decreased from 85.9717.9 Nm prior fatigue to 79.2719 Nm after isokinetic fatigue and
to 69.9716.4 Nm after isometric fatigue. The maximal strain of the GM tendon and aponeurosis before fatigue, after isokinetic and
after isometric fatigue were 4.971.1%, 4.471.1% and 4.371.1% respectively. Neither the strain nor the elongation showed
significant differences before and after each fatiguing task at any 100 N step of the calculated tendon force. This implies that the
compliance was not altered after either the isokinetic or the isometric fatiguing task. Therefore it was concluded that the strains
during the performed submaximal fatiguing tasks, were too small to provoke any structural changes in tendon and aponeurosis.
r 2005 Elsevier Ltd. All rights reserved.

Keywords: Muscle fatigue; Tendon properties; Compliance; Human gastrocnemius medialis

1. Introduction contractions having different durations. The authors
(Kubo et al., 2001a, b) suggested that acute changes in
The decrease in compliance of tendons after cyclic tendon were involved in the observed increase in
preconditioning is a well known phenomenon from in compliance. It has also been reported that a permanent
vitro studies (Abrahams, 1967; Rigby, 1964; Schwerdt deformation or ‘‘residual’’ strain becomes apparent
et al., 1980). A common explanation for the alteration in when tendon strain exceeds 2–3% (Abrahams, 1967).
compliance is the increase in the rest length of the Therefore it can be hypothesised that changes in
tendon (‘‘residual strain’’) after preconditioning compliance of tendon and aponeurosis in vivo would
(Abrahams, 1967; Rigby, 1964). Recently also in vivo be negligible after a submaximal preconditioning, where
it has been reported that repetitive muscle contractions the limit of 2–3% strain is not surpassed.
can influence tendon compliance (Kubo et al., 2001a, b). Changes in tendon compliance during cyclic or static
Kubo et al. (2001a, b) found an increase in the preconditioning are relevant for the force generation of
compliance of the vastus lateralis tendon and aponeuro- muscle fibres (Ettema, 1996). Tendon compliance can
sis after repetitive maximal and submaximal isometric influence the length and the shortening velocity of the
contractile element and so affect its force generating
Corresponding author. Fax: +49 221 497 1598. potential due to the force–length–velocity relationship.
E-mail address: arampatzis@dshs-koeln.de (A. Arampatzis). If it is assumed e.g. that the length of the muscle–tendon

0021-9290/$ - see front matter r 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jbiomech.2004.12.016
ARTICLE IN PRESS
L. Mademli et al. / Journal of Biomechanics 39 (2006) 426–434 427

unit remains unchanged during a certain phase of the performed. The latter three trials were used for
step cycle whilst the tendon experiences a significant correcting the measured joint moment (see below in
elongation, the muscle fibres would benefit from the moment measurement). Afterwards 10 concentric iso-
force–velocity relationship since they could work at a kinetic plantarflexion contractions at 601/s were per-
lower shortening velocity. However De Zee et al. (2000) formed in a range of motion between 851 and 1201
reported that in vitro the possible alterations of the having the knee fully extended. The first 5 contractions
mechanical properties of the tendon at low level strains were submaximal and increased progressively to achieve
(3%) are too small to influence the functionality of the the maximum moment, the last 5 contractions were
muscle tendon unit, and that tendons are able to act as maximal. During the contractions real time visual
stable force transmitters during long term submaximal feedback of the exerted moment was given and the
cyclic-loading. Similarly Ettema (1997) basing on an in maximum achieved isokinetic moment was determined.
situ study, reported that the increase in tendon stiffness Seventy percent of this maximum moment was set as
during a submaximal cyclic fatiguing protocol was very target for the isokinetic fatiguing protocol. The subjects
low and had no effect on muscle performance. However had to perform concentric isokinetic contractions until
it has not been examined, if the above findings are also failure to match the target moment. Failure was defined
valid for in vivo conditions. as the instant at which the subject could not hold the
The present study aims to examine whether or not the defined moment for more than 5 s. Between 1.5 and
compliance of the gastrocnemius medialis (GM) tendon 2 min after task failure a second isometric MVC at 901
and aponeurosis is altered after two fatiguing protocols: ankle angle was performed. From now on in the text this
(a) submaximal concentric isokinetic contractions and MVC will be referred to as MVC_2. Again two
(b) a sustained submaximal isometric contraction until submaximal dorsiflexions and a passive ankle joint
task failure. The first fatiguing protocol was chosen rotation were performed to correct the measured joint
because it is close to daily activities; the second one moment.
because it is characterised by a long duration of After 15 min recovery the second protocol was
permanent strain. Former in vivo studies have investi- started. It consisted of a sustained submaximal isometric
gated the vastus lateralis tendon using repetitive contraction at the same position as during the MVCs 1
contractions with different modes and suggested that and 2. The level of 40% of the MVC_1 was displayed on
the duration of the contraction is the most important the feedback monitor as target. The subjects had to
factor influencing the changes in compliance (Kubo match the target by exerting the corresponding moment
et al., 2001b). and maintain this isometric contraction until failure.
Again 1.5–2 min after task failure a third MVC was
performed, followed by two submaximal dorsiflexions
2. Methods and a passive ankle joint rotation for the corrections of
the measured joint moment. From now on in the text
2.1. Experimental design this MVC will be referred to as MVC_3. Intensive verbal
encouragement was given to all subjects throughout
Fourteen elderly male subjects, all active in sport, both fatiguing tasks. The whole experimental protocol is
(means7standard deviation: 65.273.6 yr; 175.775.4 cm; described in Fig. 1.
76.278.8 Kg) participated in the study after giving
informed consent to the experimental procedure accom- 2.2. Measurement of the ankle joint moment
plishing with the rules of the local scientific board. The
subjects were seated on a dynamometer (Biodex System The moments were measured using an isokinetic
3 Biodex Medical Systems Inc., USA) with the ankle in dynamometer (Biodex-System3, Biodex Medical Sys-
neutral position (tibia perpendicular to the sole, ankle tems Inc., USA). Before each MVC the axis of rotation
angle 901), the knee fully extended at 1801, and the hip of the dynamometer was carefully aligned with the axis
flexed at 1401. Velcro straps around the foot and the of rotation of the ankle joint. This was defined to be
dynamometer foot plate were used to restrict the ankle parallel to the axis of the dynamometer and passing
movement. through the midpoint of the line connecting both
The warm up period consisted of several isometric malleoli. During the contraction there was a clear shift
submaximal and two or three maximal contractions. of the two axes despite of the fixation (Arampatzis et al.,
Then a maximum voluntary isometric plantarflexion 2004a). To account for this, the resultant moments
contraction (MVC) lasting 4–5 s was performed at 901 around the ankle joint were calculated through inverse
ankle angle. From now on in the text this will be referred dynamics (Arampatzis et al., 2004a; Herzog, 1988).
to as MVC_1. Then two submaximal dorsiflexion Kinematic data were recorded using a Vicon 624 system
contractions and a passive rotation of the ankle joint with 8 cameras operating at 120 Hz. To determine the
at 51/s in a range of motion between 801 and 1351 were point of force application under the foot a flexible
ARTICLE IN PRESS
428 L. Mademli et al. / Journal of Biomechanics 39 (2006) 426–434

Fig. 1. The experimental set up consisted of a warm up period, 3 MVCs and two fatiguing tasks: (a) repetitive submaximal isokinetic concentric
plantarflexion contractions and (b) a sustained submaximal isometric plantaflexion contraction until failure to hold the predefined moments. The
isokinetic maximum was defined by the greatest moment value achieved during 5 maximal isokinetic contractions.

pressure distribution insole (pedar-system, Novel moment arm of the Achilles tendon was calculated using
GmbH, Germany) operating at 50 Hz was used. The the data provided by Maganaris et al. (1998) as function
output TTL-signal from the pedar-system was also of the ankle angle and level of contraction.
registered by the vicon unit to synchronise both systems.
The compensation of moments due to gravitational 2.3. Measurement of the EMG-activity
forces was done for all subjects. They were instructed to
completely relax the muscles of their left leg at the Autoadhesive preamplified Ag/AgCl bipolar EMG
studied position. Then the foot was passively rotated at electrodes (bandwidth 10–500 Hz, pickup surface
51/s. After three cycles the kinematics and the moments 0.8 cm2, inter-electrode distance 2 cm) were used to
were captured during the passive plantarflexion. This record the electric activity of the GM, gastrocnemius
allowed to calculate the moment due to the gravitational lateralis (GL), soleus (SOL) and TA muscles. After
forces for each angular position. preparation, the EMG signals from each muscle were
The antagonistic moment of the tibialis anterior (TA) checked online for artefacts due to mechanical causes by
during MVC was estimated by establishing a relationship passively shaking the leg. Additionally several func-
between EMG amplitude and exerted moment for the tional tests were undertaken to determine whether a
TA, whilst working as agonist (Baratta et al., 1988; good signal was obtained from each muscle. The
Mademli et al., 2004). In order to do that the EMG preparation was renewed when artefacts or a poor
activity of the TA and the corresponding moment were signal were observed. The EMG signals and the
measured at three additional trials (Mademli et al., 2004): measured moments were synchronously registered by
(a) in relaxed situation, (b) producing a dorsiflexion the Vicon system at a sampling rate of 1080 Hz. During
moment displaying an EMG-amplitude of the TA below the MVCs the root mean square (RMS) of the EMG
the maximum amplitude achieved during plantarflexion signal were calculated from the raw signal over a 2 s
and (c) a second dorsiflexion where the EMG-amplitude period at the plateau of the calculated tendon force.
was slightly above the maximum amplitude registered During the fatiguing tasks the EMG-signals, the
during the plantarflexion. The dorsiflexion moments plantarflexors moment and kinematics were recorded
measured for each of the three steps were fitted by a during the first 10 s from every minute.
linear regression curve as function of the corresponding
rectified and smoothed EMG-values of the TA. All 2.4. Measurement of tendinous tissue elongation during
measured data (kinematics, moment, centre of pressure, MVC
and EMG-signal) were interpolated using quintic splines
(Engeln-Müllges and Reutter, 1991) to achieve a A 7.5 MHz linear array ultrasound probe (Aloka SSD
common frequency (1080 Hz). 4000, 43 Hz) was used to visualise the distal tendon and
The tendon force was calculated by dividing the aponeurosis of the GM (Fig. 2). The ultrasound probe
plantarflexors moment by the tendon moment arm. The was placed above the muscle belly at about 50% of its
ARTICLE IN PRESS
L. Mademli et al. / Journal of Biomechanics 39 (2006) 426–434 429

length. A marker was placed between skin and ultra- 30
sound probe to allow the registration of any motion of ∆L measured
∆L corrected
the probe relative to the skin during the plantarflexion 25 ∆L passive
(Fig. 2). The ultrasound images were recorded on video
tape (50 Hz) for further analysis. To synchronise the
20

Elongation [mm]
video data (ultrasound sequences) with the data from
the Vicon system, a synchronisation box (Peak Perfor-
mance Technologies) was used. The experiment leader 15
manually triggered an analogue signal (0–5 v) which was
overlaid on the video images and simultaneously 10
captured by the Vicon system.
As mentioned above, during maximum ‘‘isometric’’ 5
plantarflexion efforts it is extremely difficult to com-
pletely prevent any joint rotation; this has a significant
0
influence on the measured elongation of the GM tendon
and aponeurosis (Magnusson et al., 2001; Muramatsu et 0 20 40 60 80 100
al., 2001). To determine the displacement of the tendon Tmax [%]
and aponeurosis due to joint rotation, the motion of the Fig. 3. Mean curves (n ¼ 14) of the measured elongation (DL
GM tendon and aponeurosis was recorded during a measured), the corrected elongation considering the joint rotation
passive motion (same as for the correction for gravita- (DL corrected), and the passive elongation due to joint rotation (DL
tional forces). This allowed the correction of the passive) of the GM tendon and aponeurosis during maximal voluntary
isometric contraction. Tmax: time to achieve maximal tendon force.
elongation measured during MVC due to joint rotation
(Fig. 3). For the analysis of the video tapes every single
frame was digitised using a video analysis software (Simi
Motion 5.0). to compare the elongation and the strain between the
The elongation and strain of the GM tendon and MVC before and after the fatiguing tasks at the same
aponeurosis during the MVC_1, MVC_2 and MVC_3 given tendon force.
were identified and analysed at the maximum calculated
tendon force and at every 100 N. This procedure allowed 2.5. Statistics

A variance analysis (Friedman test) was used to check
for any differences between the maximum values of
moment, calculated tendon force, elongation, and strain
of the GM tendon and aponeurosis during MVC_1,
MVC_2 and MVC_3. When statistically significant
differences were found the parameters were compared
using the Wilcoxon test. The same statistical procedure
was used for the comparison of the strain at every 100 N
(tendon force) between the three MVCs in order to
detect any alterations in the compliance of the tendon
and aponeurosis. The level of significance was set at
Po0:05: In the figures the data are presented as
mean71 SEM (standard error of mean), whereas in
the text and tables they are expressed as mean71 SD
(standard deviation).

3. Results

The mean time to task failure for the concentric
isokinetic fatiguing task was 117871372 s. For the
Fig. 2. Ultrasound image of the relaxed GM at about 50% of the sustained isometric fatiguing task it was 4077129 s. The
length of its muscle belly. The elongation of the GM tendon and
EMG activity of the GM, GL, SOL, and TA and the
aponeurosis was defined as the distance travelled by the cross-point
along the visualised deeper aponeurosis during the plantarflexion. The corresponding moment during the isokinetic and iso-
marker between the skin and the ultrasound probe was used to register metric fatiguing task from one subject are exemplarily
any motion of the probe relative to the skin during the plantarflexion. illustrated in Figs. 4 and 5.
ARTICLE IN PRESS
430 L. Mademli et al. / Journal of Biomechanics 39 (2006) 426–434

Fig. 4. Raw EMG signal from the GM, GL, SOL, TA, and the corresponding moment during the concentric submaximal isokinetic fatiguing task
from one subject. The first column shows data captured during the first minute of the fatiguing task, the second column during the middle interval of
the total duration and the third column during the last minute.

The maximum moment after the isokinetic fatiguing 4. Discussion
task showed a tendency to decrease from 85.9717.9 Nm
pre fatigue to 79.2719.0 Nm. This however was not The main finding of the present study is that the
statistically significant (P ¼ 0:085). After the isometric elongation and strain values of the GM tendon and
fatiguing task the maximum moment was aponeurosis before and after the fatiguing tasks did not
69.9716.4 Nm and significantly lower (Po0:05) than differ from each other at any of the 100 N steps of
the moment of both previous MVCs (Table 1). Similar tendon force (from 100 to 900 N). This implies that
to the maximum moment, the maximum calculated 1.5–2 min after task failure the compliance of the GM
tendon force after the isometric fatigue was significantly tendon and aponeurosis was not affected by the
lower than the values of the other two MVCs (Table 1). fatiguing protocols. Although it has been reported that
The maximum elongation and strain during the MVCs after submaximal preconditioning the tendon recovers
after the fatiguing tasks showed a decreasing tendency after 10 min rest (Rigby, 1964), some remaining
(P ¼ 0:062P ¼ 0:09) compared to the values found effects of the isokinetic protocol after 15 min recovery
during MVC_1 (Table 1). Comparing MVC_2 with cannot be completely discarded. However this should
MVC_3 no statistically significant differences were not affect the conclusions of the present study, since
found neither in maximum elongation nor in maximum no alteration was found in the examined tendinous
strain. properties after the fatiguing protocols. Taking the
The elongation and strain of the tendon and tendon moment arm values from the literature (Maga-
aponeurosis of the GM during the MVCs before and naris et al., 1998) could be a methodological limitation.
after each fatiguing task showed no significant differ- It is possible that interindividual anatomical differences
ences at any 100 N step of the calculated tendon force in moment arm may exist and affect the calculation of
(Table 2, Fig. 6). the tendon force. However in this study the subjects
ARTICLE IN PRESS
L. Mademli et al. / Journal of Biomechanics 39 (2006) 426–434 431

Fig. 5. Raw EMG signal from GM, GL, SOL, TA, and the corresponding moment during the submaximal isometric fatiguing task from one subject.
The first column shows data captured during the first minute of the fatiguing task, the second column during the middle interval of the total duration
and the third column during the last minute.

Table 1
Comparison of the maximum values of the examined parameters during the pre-fatigue MVC (MVC_1) with those during the MVCs after isokinetic
(MVC_2) and isometric fatigue (MVC_3) (means7 SD, n ¼ 14)

MVC_1 MVC_2 MVC_3

Moment (Nm) 85.9717.9 79.2719.0 69.9726.4a,b
Tendon Force (N) 1326.47276.5 1250.07285.0 1131.57244.4a,b
Elongation (mm) 15.173.1 13.572.8 13.372.8
Strain (%) 4.971.1 4.471.1 4.371.1

Moment: maximal exerted moment from the plantarflexor muscles during MVC; tendon Force: maximal calculated tendon force during MVC;
elongation: maximal elongation of the tendon and aponeurosis of the GM during MVC; strain: maximal strain of the tendon and aponeurosis of the
gastrocnemius medialis during MVC.
a
Significantly different to MVC_1 (Po0:05).
b
Significantly different to MVC_2 (Po0:05).

were compared with themselves before and after reduction of the force generating capacity of the muscle
each fatiguing task. Thus this error would be systematic due to fatigue. After the isokinetic fatiguing task these
and therefore should not influence the presented values were about 93% of the maximum value measured
findings. during the MVC_1 (Table 1). After the isometric
The decrease in maximum moment and calculated fatiguing task the maximum calculated tendon force
tendon force after the fatiguing tasks confirmed the and moment were significantly lower (Po0:05) and
ARTICLE IN PRESS
432 L. Mademli et al. / Journal of Biomechanics 39 (2006) 426–434

Table 2
Means7standard deviations of the elongation of the GM tendon and aponeurosis at every 100 N of the calculated tendon force during the MVCs
before fatigue (MVC_1), after isokinetic fatigue (MVC_2), and after isometric fatigue (MVC_3)

Tendon force [N] MVC_1 Elongation [mm] MVC_2 Elongation [mm] MVC_3 Elongation [mm]

0 0.370.1 0.470.1 0.370.1
100 1.070.3 1.370.2 1.370.2
200 2.170.5 2.270.3 2.570.2
300 3.170.6 3.270.4 3.770.2
400 4.270.6 4.070.5 4.970.3
500 5.370.7 5.170.6 5.970.3
600 6.470.7 6.070.6 7.070.4
700 7.670.6 7.170.6 8.170.4
800 8.870.5 8.070.7 9.370.4
900 10.370.6 9.670.8 10.770.5

The table ends at 900 N, which corresponds to the maximum common force achieved by all subjects during the MVCs (n ¼ 14).

6 becomes apparent in relaxed tendons (Abrahams, 1967).
MVC_1
Although in previous studies it has been reported, that
MVC_2
5 MVC_3 when this limit is exceeded an irreversible structural
change takes place (Abrahams, 1967), more recent
4
studies found no significant alterations in biological
tissue stiffness after preconditioning with superimposed
Strain [%]

sinusoidal vibration of small amplitude (Ettema, 1998).
3
The study of Ettema (1998) distinguishes between plastic
deformation and slow viscous behaviour of a tissue and
2 states that the pure elastic behaviour is independent of
any long term changes in the tissue.
1 Thornton et al. (1997, 2002) reported a mechanism to
prevent ligament damage due to creep at low stresses.
0 They found that a progressive collagen fibre recruitment
0 200 400 600 800 1000 1200 1400 can reduce the stress on the initially loaded fibres. The
Tendon Force [N] increase in the amount of the collagen fibres distributes
the stress over a greater area. Thus the stress placed on
Fig. 6. Means and SEM of the strain of the GM tendon and the initial loaded fibres will be reduced. This same
aponeurosis at every 100 N and at maximum of the calculated tendon
force during the MVCs before fatigue (MVC_1), after isokinetic
mechanism might prevent the GM tendon and apo-
fatigue (MVC_2), and after isometric fatigue (MVC_3). The curve ends neurosis to experience any permanent structural changes
at 900 N, which corresponds to the maximum common force achieved during the examined fatiguing tasks. Consequently the
by all subjects during the MVCs (n ¼ 14). capacity of the tendon and aponeurosis to resist force
remained unaffected after submaximal fatigue.
Opposite to this study other in vivo studies found an
about 84% of the maximum value measured during the increase in tendon compliance after fatigue (Kubo et al.,
MVC_1 (Table 1). 2001a, b). In order to explain the altered tendon
The isokinetic fatiguing task consisted of submaximal compliance, they suggested that the tendon suffered
concentric isokinetic plantarflexions at 70% of the acute changes during the repeated muscle contractions
maximal isokinetic contraction with an angular velocity (Kubo et al., 2001a, b). This apparent discrepancy can
of 601/s. Seventy percent of the maximal isokinetic be due to the different protocols and methods used in
contraction corresponds to nearly 50% of the isometric both studies. Kubo et al. (2001a, b) investigated the
MVC_1. At this tendon force the corresponding strain vastus lateralis using repetitive contractions at higher
of the GM tendon and aponeurosis is about 2.4%. The loads. Moreover Kubo et al. (2001a, b) did not consider
second fatiguing task consisted of a sustained isometric the joint rotation occurring during an ‘‘isometric’’
contraction at 40% of the MVC_1, where the strain of contraction in their methodology. This happens
the GM tendon and aponeurosis was about 1.9%. None despite of using external fixations and can influence
of the strain values did exceed the limit of 2–3%, which the estimation of the mechanical properties of the
is the point, beyond which a permanent deformation tendon and the aponeurosis (Arampatzis et al., 2004b;
ARTICLE IN PRESS
L. Mademli et al. / Journal of Biomechanics 39 (2006) 426–434 433

Bojsen-Møller et al., 2003). The joint rotation might 350 GM GL SOL
affect the comparison of the compliance between two
MVCs. This latter explanation is supported by the fact, 300
that when comparing our measured elongation values at
250

EMGRMS [ADU]
every 100 N calculated tendon force without considering
the joint rotation we find significant differences in 200
elongation between MVC_1 and MVC_2 above 300 N.
The present study shows that tendon compliance was 150
not altered after the two presented fatiguing protocols.
100
It seems that tendons are capable to resist submaximal
contractions without any alteration in their mechanical 50
properties, which would have an impact on their
muscle’s function. This is in agreement with earlier MVC_1 MVC_2 MVC_3
reports from in vitro (De Zee et al., 2000) and in situ
studies (Ettema, 1997). However Ettema (1997) found Fig. 7. Means and SEM of the RMS of the EMG signal from the GM,
that the compliance of the rat GM tendon shows a small GL and SOL during the MVC before fatigue MVC_1, after isokinetic
fatigue (MVC_2), and after isometric fatigue (MVC_3) (n ¼ 14).
but statistically significant decrease after submaximal
repetitive contractions when using electro stimulation.
The differences in tendon elongation prior and after
fatigue in the study of Ettema (1997) were about 7-8% went any changes, we compared the RMS values of the
and were too small to influence muscle function EMG-activity from the different muscles during the
(Ettema, 1997). The discrepancy with the results of the three MVCs. We found similar relative EMG-activities
present study might be due to the different accuracies of in GM, GL and SOL (Fig. 7). This suggests that the
both applied methods. Such small changes in elongation relative contribution of each muscle was not substan-
can hardly be detected in vivo by means of ultrasound tially altered and therefore this issue should have no
imaging. At our study 8% change in elongation would influence on the presented results.
correspond to about 0.8 mm at 900 N calculated tendon In conclusion no differences in strain and elongation
force. This is within the accuracy of the ultrasonography of the GM tendon and aponeurosis were found at any
(0.7  0.4 mm resolution of the ultrasound images). given calculated tendon force. This implies that the
However some factors influencing the finding of the compliance was not altered after the isokinetic and
present study may be: (a) changes in the initial length of isometric fatiguing task. It can therefore be suggested
the tendon and aponeurosis and (b) different relative that the strains during the performed submaximal
contributions of the GM, GL and SOL to the tendon fatiguing tasks, are too small to provoke any structural
force during the three MVCs. Regarding the first changes in the GM tendon and aponeurosis, which are
possible factor, it is well known from the literature that presumably subjected to high loads during daily
changes in the initial length of the tendon influence its activities.
compliance (Rigby, 1964; Schwerdt et al., 1980). It can
be hypothesised that after the performed fatiguing tasks
creep could alter the initial length of the GM tendon and References
aponeurosis. In the present study it was not possible to
measure directly the changes in the initial length, Abrahams, M., 1967. Mechanical behaviour of tendon in vitro. A
because we could not identify the same point at rest preliminary report. Medical and Biological Engineering and
Computing 5, 433–443.
before and after the fatiguing tasks on the ultrasound
Arampatzis, A., Morey-Klapsing, G., Karamanidis, K., De Monte, G.,
images. Instead of this the fascicle length and the Stafilidis, S., Bruggemann, G.P., 2004a. Differences between
pennation angle of the GM at rest were estimated at the measured and resultant joint moments during isometric contrac-
same position prior and after fatigue. The mean- tions at the ankle joint. Journal of Biomechanics, in Press.
s7standard deviation of the pennation angle and the Arampatzis, A., Karamanidis, K., De Monte, G., Stafilidis, S., Morey-
fascicle length before fatigue were 21.071.91 and Klapsing, G., Bruggemann, G.P., 2004b. Differences between
measured and resultant joint moments during voluntary and
52.178.6 mm respectively and did not differ from those artificially elicited isometric knee extension contractions. Clinical
after each fatiguing task (after isokinetic fatigue: Biomechanics 19, 277–283.
21.172.11 pennation angle and 52.677.9 mm fascicle Baratta, R., Solomonow, M., Zhou, B., Letson, D., Chuinard, R.,
length; after isometric fatigue: 20.773.61 and D’Ambrosia, R., 1988. Muscular coactivation. The role of the
53.979.6 mm, respectively). This supports that there antagonist musculature in maintaining knee stability. The Amer-
ican Journal of Sports Medicine 16, 113–122.
were no considerable changes in initial length of the GM Bojsen-Møller, J., Hansen, P., Aagaard, P., Kjaer, M., Magnusson,
tendon and aponeurosis. In order to find out if the S.P., 2003. Measuring mechanical properties of the vastus lateralis
relative contributions of the examined muscles under- tendon-aponeurosis complex in vivo by ultrasound imaging.
ARTICLE IN PRESS
434 L. Mademli et al. / Journal of Biomechanics 39 (2006) 426–434

Scandinavian Journal of Medicine and Science in Sports 13, on the estimation of the antagonistic moment during maximal
259–265. plantarflexion. Journal of Electromyography and Kinesiology 14,
De Zee, M., Bojsen-Moller, F., Voigt, M., 2000. Dynamic viscoelastic 591–597.
behavior of lower extremity tendons during simulated running. Maganaris, C., Baltzopoulos, V., Sargeant, A., 1998. Changes in
Journal of Applied Physiology 89, 1352–1359. Achilles tendon moment arm from rest to maximum isometric
Engeln-Müllges, G., Reutter, F., 1991. Formelsammlung zur Numer- plantarflexion: in vivo observations in man. Journal of Physiology
ischen Mathematik mit Turbo Pascal-Programmen. Wissenschafts- 510.3, 977–985.
verlag, Mannheim/Wien/Zürich. Magnusson, S.P., Aagaard, P., Dyhre-Poulsen, P., Kjaer, M., 2001.
Ettema, G.J., 1996. Mechanical efficiency and efficiency of storage and Load–displacement properties of the human triceps surae apo-
release of series elastic energy in skeletal muscle during stretch- neurosis in vivo. Journal of Physiology 531, 277–288.
shorten cycles. Journal of Experimental Biology 199, 1983–1997. Muramatsu, T., Muraoka, T., Takeshita, D., Kawakami, Y., Hirano,
Ettema, G.J., 1997. Mechanical behaviour of rat skeletal muscle during Y., Fukunaga, T., 2001. Mechanical properties of tendon and
fatiguing stretch-shortening cycles. Experimental Physiology 82, aponeurosis of human gastrocnemius muscle in vivo. Journal of
107–119. Applied Physiology 90, 1671–1678.
Ettema, G.J., 1998. A new method to measure elastic properties of Rigby, B., 1964. Effect of cyclic extension on the physical properties of
plastic-viscoelastic connective tissue. Medical Engineering and tendon collagen and its possible relation to biological ageing of
Physics 20, 308–314. collagen. Nature 202, 1072–1074.
Herzog, W., 1988. The relation between the resultant moments at a Schwerdt, H., Constantinesco, A., Chambron, J., 1980. Dynamic
joint and the moments measured by an isokinetic dynamometer. viscoelastic behaviour of the human tendon in vitro. Journal of
Journal of Biomechanics 21, 5–12. Biomechanics 13, 913–922.
Kubo, K., Kanehisa, H., Kawakami, Y., Fukunaga, T., 2001a. Effects Thornton, G.M., Oliynyk, A., Frank, C.B., Shrive, N.G., 1997.
of repeated muscle contractions on the tendon structures in Ligament creep cannot be predicted from stress relaxation at low
humans. European Journal of Applied Physiology 84, 162–166. stress: a biomechanical study of the rabbit medial collateral
Kubo, K., Kanehisa, H., Kawakami, Y., Fukunaga, T., 2001b. ligament. Journal of Orthopaedic Research 15, 652–656.
Influences of repetitive muscle contractions with different modes Thornton, G.M., Shrive, N.G., Frank, C.B., 2002. Ligament creep
on tendon elasticity in vivo. Journal of Applied Physiology 91, recruits fibres at low stresses and can lead to modulus-reducing
277–282. fibre damage at higher creep stresses: a study in rabbit medial
Mademli, L., Arampatzis, A., Morey-Klapsing, G., Bruggemann, collateral ligament model. Journal of Orthopaedic Research 20,
G.P., 2004. Effect of ankle joint position and electrode placement 967–974.