ORIGINAL ARTICLE

S. Kuitunen J. Avela H. Kyro la inen C. Nicol

P.V. Komi
Acute and prolonged reduction in joint stiffness in humans
after exhausting stretch-shortening cycle exercise
Accepted: 5 June 2002 / Published online: 14 August 2002
Springer-Verlag 2002
Abstract The purpose of the present study was to
examine the acute and long-term fatigue eects of
exhausting stretch-shortening cycle (SSC) exercise on the
stiness of ankle and knee joints. Five subjects were
fatigued on a sledge apparatus by 100 maximal rebound
jumps followed by continuous submaximal jumping
until complete exhaustion. Neuromuscular fatigue ef-
fects were examined in submaximal hopping (HOP) and
in maximal drop jumps (DJ) from 35 (DJ35) and 55 cm
(DJ55) heights on a force plate. Additional force and
reex measurements were made using an ankle ergom-
eter. Jumping tests and ankle ergometer tests were car-
ried out before, immediately after, 2 h (2H), 2 days and
7 days (7D) after the SSC exercise. Kinematics, force
and electromyography (EMG) recordings were comple-
mented with inverse dynamics, which was used to cal-
culate joint moments. The quotient of changes in joint
moment divided by changes in joint angle was used as a
value of joint stiness (JS). In addition, blood lactate
concentrations and serum creatine kinase activities were
determined. The exercise induced a clear decrease in
knee joint stiness by [mean (SD)] 29 (13)% (P<0.05)
in HOP, 31 (6)% (P<0.05) in DJ35 and 34 (14)%
(P<0.05) in DJ55. A similar trend was observed in the
ankle joint stiness with signicant post-exercise reduc-
tions of 22 (8)% (P<0.05) in DJ35 and of 27 (19)%
(P<0.05) at 2H in DJ55. The subsequent recovery of JS
was slow and in some cases incomplete still at 7D.
Generally, all the EMG parameters were fully recovered
by 2H, whereas the force recovery was still incomplete at
this time. These data indicate that the immediate
reduction in JS was probably related to the eects of
both central (neural) and peripheral (metabolic) fatigue,
whereas the prolonged impairment was probably due to
peripheral fatigue (muscle damage).
Keywords Jointstiness Stretch-shortening cycle
Central fatigue Peripheral fatigue Muscle damage
Introduction
It is well known that stretching an active muscle will
increase the force in the subsequent shortening phase
when compared to a pure shortening contraction (e.g.
Cavagna 1977). This potentiation eect, which has been
shown to occur also in natural locomotion (e.g. Komi
1984), originates from the elastic recoil of tendinous
structures (Asmussen and Bonde-Petersen 1974; Alex-
ander and Bennet-Clark 1977; Cavagna 1977) and from
the myoelectric potentiation including the stretch reex
action (Dietz et al. 1979; Bosco et al. 1982, Komi and
Gollhofer 1997). It has been suggested that the eective
use of elastic energy during natural stretch-shortening
cycle (SSC) locomotion requires high tendomuscular
stiness (quotient of force change divided by length
change) (Gollhofer et al. 1984, 1992). A proper adjust-
ment in tendomuscular stiness allows the absorption of
high impact forces and the storage of elastic energy
during the active braking phase. Muscle activation prior
to the ground impact (pre-activation) has been reported
to play a major role in this regard (Asmussen and
Bonde-Petersen 1974; Gollhofer et al. 1984).
Numerous fatigue studies have shown that intense
SSC exercise can lead to acute and long-term impair-
ments of neuromuscular performance. This includes re-
ductions in the maximal activation of muscle and the
force generated in maximal voluntary contraction
(MVC) (Nicol et al. 1991a; Pullinen et al. 1997; Avela
et al. 1999) as well as in the performance of maximal
SSC (Gollhofer et al. 1987; Nicol et al. 1991b; Avela
et al. 1999). Impairments in SSC performance have been
Eur J Appl Physiol (2002) 88: 107116
DOI 10.1007/s00421-002-0669-2
S. Kuitunen (&) J. Avela H. Kyro la inen P.V. Komi
Neuromuscular Research Centre,
Department of Biology of Physical Activity,
P.O. Box 35, 40014 University of Jyva skyla , Finland
E-mail: kuitunen@maila.jyu.
Tel.: +358-14-2602058
Fax: +358-14-2602071
C. Nicol
Department of Biology of Physical Activity,
University of the Mediterranean, Marseille, France
associated with a lowered tolerance to ground impact
(Gollhofer et al. 1987; Nicol et al. 1991b; Avela et al.
1999) and a reduced muscle (Avela and Komi 1998) and
joint (Horita et al. 1996) stiness. The study of Horita
et al. (1996) demonstrated a prolonged reduction in in-
stantaneous knee joint stiness (KJS) after submaximal
SSC exercise. The authors suggested that the exercise-
induced long-lasting muscle failure resulted in changes
in the neural drive through reex modulation with as-
sociated stiness regulation. Part of the immediate and
prolonged functional impairments after SSC exercise
would be expected to result from the muscle failure
(damage) induced by the eccentric phase of the SSC
(Horita et al. 1996, 1999; Nicol et al. 1996; Avela et al.
1999). It has been shown that damage or disorganisation
of intracellular structures with their subsequent inam-
mation processes can contribute to a failure in con-
traction over several days (Bruton et al. 1998; Allen
2001; Friden and Lieber 2001).
On the other hand, the muscle damage and inam-
mation cause some biochemical substances (e.g. cyto-
kines) to be released (Armstrong et al. 1991). Some of
these substances, as well as intramuscular increases in
pressure and temperature, are known to stimulate
groups III and IV muscle aerents (Kniki et al. 1978;
Kaufman and Rybicki 1987; Mense and Meyer 1988;
Rotto and Kaufman 1988). The stimulation of these
muscle aerents has been suggested to lead to presy-
naptic inhibition of Ia aerent terminals resulting in a
reduced motoneuronal drive to muscles (Duchateau and
Hainaut 1993). Intense SSC exercise can also lead to
increased muscle acidosis and the accumulation of me-
tabolites, which may impair the contractile function di-
rectly through disturbances in Ca
++
dynamics (Green
1997; Allen and Westerblad 2001) or indirectly through
groups III and IV muscle aerents (see above) (Mense
1977; Synoway et al. 1993; Darques et al. 1998). There is,
however, evidence indicating that neither pH (Wester-
blad et al. 2002) nor inorganic phosphate (Coupland
et al. 2001) have large direct eects on force production.
It has also been suggested that maintained voluntary
contraction can lead to decreased facilitation of the
a-motoneuron pool (disfacilitation) through reduced
activity in the c-loop and thus play a role in the decline in
neural drive (Bongiovanni and Hagbarth 1990; Mace-
eld et al. 1991). In addition to these alterations in reex
pathways insucient supraspinal drive (Brasil-Neto et al.
1994) may also be present during the fatiguing exercise.
The purpose of the present study was to examine the
acute and delayed eects of exhausting SSC exercise on
the regulation of KJS and ankle joint stiness (AJS).
Methods
Subjects
Five recreationally active subjects [mean (SD)] [age 24.4 (3.4) years,
height 1.77 (0.07) m and body mass 76.4 (11.9) kg] participated in
the study after giving their written informed consent. The study was
conducted according to the Declaration of Helsinki and was ap-
proved by the Ethics Committee of the University of Jyva skyla .
Experiment design
Fatigue was induced by an exhausting SSC exercise. The eects of
fatigue were investigated through reex and force measurements of
the plantarexor muscles as well as in jumping tests. These mea-
surements were carried out before (BEF) and immediately after
(AFT) as well as 2 h (2H), 2 days (2D) and 7 days (7D) after the
SSC exercise.
Exhausting SSC exercise
Subjects were fatigued until exhaustion by performing rebound
jumps on a special sledge ergometer (Kyro la inen and Komi 1995)
inclined at 24.9 from the horizontal. They were seated on the
sledge, and after a familiarization phase they were dropped from
dierent starting heights to determine the optimal dropping height
that led to their highest rebound performance (for details see
Kyro la inen and Komi 1995; Nicol et al. 1996). The fatigue protocol
included rst 100 single maximal drop jumps (DJ) from the optimal
dropping height that were repeated every 5 s. This series of jumps
was immediately followed by a continuous rebounding exercise (see
also Nicol et al. 1996) to a submaximal height representing 70% of
their maximal initial rebound performance. This exercise was
stopped when the subjects were so exhausted that they were unable
to jump at all. In this SSC exercise, the lower limbs were supported
by a rope which was attached to the ceiling at one end and to the
subjects ankles by a soft belt at the other. This procedure was done
to minimize the metabolic fatigue of the quadriceps muscle group
during the airborne phase of the leg action. The supportive system
did not restrict the normal range of movement during the jumps.
Reex and torque measurements
Passive stretch reexes (PSR), maximal H-reex and M-wave re-
sponses of soleus (SOL) muscle, as well as plantarexor torque
during isometric MVC and low frequency (LF) electrical stimula-
tion (20 Hz) were measured from the left leg. The subjects were
seated in an ergometer chair with their left thighs clamped and their
foot mounted on the motor-driven platform (for details see Nicol
et al. 1996). Special care was taken to align the rotation axis of the
ankle joint to correspond to that of the ankle ergometer platform.
In PSR, the ankle ergometer was controlled by a computer to in-
duce ten consecutive dorsiexions of 0.175 rad amplitude and at a
mean stretching velocity of 1.9 rads
1
. Bipolar surface electrodes
(Beckman miniature skin electrodes 650437, Ill., USA) with a xed
inter-electrode distance of 20 mm were placed on the lateral gas-
trocnemius (GA) and SOL of the left leg according to recommen-
dations of the Surface Electromyography for the Non-Invasive
Assessment of Muscles-project (SENIAM 1999). The electromyo-
gram (EMG) signals were amplied and recorded telemetrically
(ME3000P, MEGA, Kuopio, Finland; cut-o frequency 360
Hz/3 dB).
Standard methodology was used to record the H-reexes
(Hugon 1973). After preparation of the skin, the stimulation elec-
trode (pregelled Ag-AgCl electrodes, Niko, Denmark) was placed
over the tibial nerve in the popliteal fossa. The position of the
stimulating electrode was tested rst in the standing position and
then checked in the experimental position to ensure constant re-
cording conditions. The intensity of the electrical stimulus was
adjusted in every test unit to elicit maximal M- and H-responses in
the SOL muscle. Single rectangular pulses of 1 ms duration
(stimulus frequency 0.2 Hz) were delivered by an evoked-potential
measuring system (MEB-5304 K, Nihon Kohden, Japan) when
testing M-waves and H-reexes.
The LF stimulation of the triceps surae muscle was produced by
two stimulating electrodes (5 cm5 cm; StimRode, Axelgaard
Manufacturing Co., Calif., USA) placed percutaneously on the
108
proximal head of the medial and lateral GA muscles. One anode
was located distally on the muscle-tendon region on to the SOL.
The stimulation intensity was set to a level corresponding to three
times the motor threshold determined before every test occasion.
Plantarexion MVC was measured three times and the level of
the maximal voluntary neural activation was checked by superim-
posed double twitch stimuli applied to the tibial nerve. In each of
these tests, the EMG and force signals were amplied and stored
simultaneously in a computer for further analysis.
Jumping tests
Approximately 5 min after the reex and torque measurements, the
subjects performed the rst continuous submaximal hopping (HOP)
at their preferred frequency and then ve to seven maximal DJ from
0.35m (DJ35) and 0.55m (DJ55) on a force plate. The subjects were
instructed to keep their hands on their waists and to minimize the
angular displacements of the joints of the lower extremity. Vertical
ground reaction force (GRF) was measured using a force platform
[Raute Precision Oy, Lahti, Finland; natural frequency 180 (10) Hz].
Vertical GRF and EMG (detailed methodology given above in Re-
ex and torque measurements) of SOL, GA, tibialis anterior (TA),
vastus lateralis (VL) and biceps femoris (BF) were collected at a
sampling frequency of 1 kHz. Joint markers were placedonthe distal
head of the fth metatarsal bone, heel, lateral malleolus, lateral ep-
icondyle of the femur, greater trochanter and shoulder for kinematic
recordings at 200 framess
1
in the sagittal plane.
Blood samples
Capillary blood samples from the ngertip were taken to determine
blood lactate concentration ([La

]
b
) BEF and AFT the SSC exer-
cise as well as 3 min, 8 min and 2H post-exercise. The [La

]
b
was
analysed enzymatically using a commercial Lactate enzymatic kit
(Biochemica Boehringer, Mannheim, Germany). Furthermore, ve-
nous blood samples were taken from the ulnar vein BEF, and AFT,
as well as 2H, 2D, 3D and 7D after the SSC exercise for deter-
mining serum creatine kinase (CK) activity. The CK activity was
determined using a CK enzymatic test kit (Biochemica Boehringer).
Data treatment
In the reex tests, the rst PSR response was excluded from the
analysis to minimize the eect of muscle thixotropy (Proske et al.
1993). Thus, nine PSR responses were averaged for each subject to
determine the peak-to-peak EMG amplitude of SOL. Mean values
of three highest peak-to-peak amplitudes of maximal H-reex and
maximal M-wave responses of SOL were calculated to determine
the quotient H/M. In the isometric strength test, the MVC having
the highest peak torque was chosen for the analysis and the average
EMG (aEMG) was calculated for a 500 ms window around peak
torque. The average torque during the plateau phase of LF stim-
ulation was also determined. The activation level (AL) was then
indirectly assessed according to Strojnik and Komi (1998) by
comparing the increase in torque induced by the superimposed
double twitch (T
b
) over the maximal torque in the MVC (T
max
) in
proportion to the torque in passive double twitch (T
DTW
), as fol-
lows: AL=100D(T
b
/T
max
)/T
DTW
100 [originally formulated by
Merton (1954)].
In the jumping tests, jumping performance was determined by
vertical take-o velocity (v
0
) of the centre of gravity that was cal-
culated from the ight time (t
f
) according to the equation: v
0
=1/2
t
f
g (g=9.81 ms
2
) (Bosco et al. 1981). From the video recordings,
body segment coordinates were digitized (Motus software, Peak
Performance Technologies, Inc., Denver, USA), ltered (Butter-
worth lter, cut-o frequency 20 Hz) and transferred to the com-
puter system (SGI O2 R5000, Silicon Graphics, Inc., Calif. USA)
for further analysis. The scaled coordinates and GRF were syn-
chronized and joint moments were calculated at 200 Hz (Belli et al.
2002). Anthropometric data, provided by the standards of Demster
(1955), were used to determine inertia and mass of the segments.
Four to eight successful trials were averaged in each jumping
condition. The average AJS and KJS was calculated as a change in
the joint moment (DM) divided by the change in joint angle (Dh) in
the braking phase (Fig. 1) (Kuitunen et al. 2002).
Statistical analysis
Means and standard deviations (SD) were calculated and the non-
parametric Wilcoxon test was used to compare the signicance of
absolute changes from the pre-exercise values. In addition, Pearson
correlation coecients were calculated to determine the relation-
ships between selected variables.
Results
The exhausting SSC exercise lasted on average 27 min
[8 min 20 s for the 100 maximal rebound jumps and
Fig. 1. Knee joint moment
joint angle relationship in max-
imal drop jumps from heights
of 35 cm during the ground
contact. Joint stiness (JS) for
both ankle and knee joints was
calculated as an average value
in the braking phase by dividing
the change in joint moment
(DM
braking
) by the correspond-
ing change in joint angle
(Dh
braking
). The direction of
positive joint moments and in-
dications of the joint angles are
given in the inset to the gure
109
18 min 40 s (13 min 10 s) for the subsequent continuous
jumping at 70% intensity]. A clear reduction in jumping
performance was observed after the exercise. The v
0
decreased immediately after the exercise by 13 (14)%
(P<0.05) in submaximal HOP. Similarly to HOP, v
0
decreased in maximal DJ by 10 (9)% (n.s.) in DJ35 and
8 (7)% (n.s.) in DJ55 (Fig. 2). Due to a minor increase
(2%3%) in post-exercise v
0
in DJ35 and DJ55 in one
subject, no statistical signicances were achieved in DJ.
The v
0
recovered gradually thereafter until 7D in DJ,
whereas in HOP a second decline of 11 (8)% (P<0.05)
in 7D was observed. In addition, the v
0
was inversely
related to the preferred HOP frequency throughout the
experiment to which it had a highly signicant correla-
tion (r=0.87, P<0.01). The HOP frequency was
9 (8)% (n.s.) higher at 7D with a value of 1.90 (0.15) Hz
compared to the pre-exercise value of 1.76 (0.08) Hz.
The exhausting SSC exercise induced a dramatic im-
mediate post-exercise decrease and delayed recovery in
joint stiness (JS; Fig. 3, upper graphs). The observed
decrease was larger in the KJS than in AJS. These ob-
servations were of greater magnitude in maximal DJ
compared to submaximal HOP. The immediate post-
exercise reductions in KJS were 29 (13)% (P<0.05) in
HOP, 31 (6)% (P<0.05) in DJ35 and 34 (14)%
(P<0.05) in DJ55. No recovery in the KJS had taken
place by the 2H measurement. A similar trend was also
observed in AJS with signicantly reduced post-exercise
values of 22 (8)% (P<0.05) in DJ35 and 25%27%
(18%20%) from 2H to 7D in DJ55. Thus the delayed
recovery of the KJS and the AJS did not occur in phase.
Fig. 2. Relative changes (D) in vertical take-o velocity in jumping
tests [submaximal hopping HOP, drop jumps from 35 and 55 cm
(DJ35, DJ55, respectively)] after the stretch shortening cycle
exercise. *P<0.05 Signicantly dierent from the pre-exercise
value. BEF Before, AFT immediately after, and 2H 2 hours, 2D 2
days, 7D 7 days after
Fig. 3. Relative changes in knee (DKSJ; upper left panel) and ankle
joint stiness (DASJ; upper right panel) in dierent jumping tests
(HOP, DJ35, DJ55). Lower panels represent the relative changes in
joint angles (Dh) and joint moments (DM) for the knee (left panel)
and ankle (right panel). Please note the dierent scales for ankle and
knee in the lower panels. *P<0.05 Signicantly dierent from the
pre-exercise value. For other denitions see Fig. 2
110
The reductions in the KJS and AJS resulted from an
increase in Dh along with a decrease in DM in the
braking phase (Fig. 3, lower graphs). It can also be seen
in Fig. 3 that Dh for the knee followed well the changes
in the KJS, whereas the AJS was more closely related to
DM. In addition, a dramatic and prolonged post-exercise
increase occurred in the angular displacement of the hip
in the braking phase especially in DJ55 (Fig. 4). This
increase in hip exion followed well the respective
changes in knee exion (Fig. 3, lower graph on left) thus
revealing a modied jumping strategy until 7D in DJ55.
The EMG analysis of the dierent types of jumps
revealed a consistent trend towards a higher preactivity
with increasing load (DJ55>DJ35>HOP) and for a
lower activation during the push-o phase in HOP. The
post-fatigue analysis revealed general trends of an acute
reduction in the EMG activity of the triceps surae
muscle group (Fig. 5, upper two graphs) and a bimodal
pattern in the quadriceps muscle group (Fig. 5, lower
graph). Comparison of the respective changes in EMG
activity and JS showed that the relative immediate post-
exercise changes in AJS and KJS were both positively
related to the respective changes in the pre-activation of
the triceps surae (r=0.92, P<0.05) and VL (r=0.91,
P<0.05) muscles (Fig. 6). In addition, the TA and BF
muscles showed trends of decrease and recovery similar
to those of the agonist muscles.
In Fig. 7 it can be seen that a marked and immediate
impairment and subsequent bimodal recovery occurred
in the plantarexor muscles just after the SSC exercise,
both in muscle activation and in the capacity for muscle
force production. The aEMG in MVC showed a re-
duction of 28 (16)% (P<0.05) in SOL, and 26 (10)%
(P<0.05) in GA, respectively. This was followed by a
subsequent recovery to the pre-exercise level at 2H with
a secondary decline of 14 (5)% (P<0.05) in GA at 2D.
The plantarexion torque decreased immediately after
the exercise by 30 (20)% (P<0.05) in the MVC test from
784 (91) to 560 (213) Nm and by 45 (16)% (P<0.05) in
the LF test from 39.0 (8.6) to 22.2 (10.3) Nm. Recovery
of the LF torque was found to be particularly slow. As
shown by Fig. 8, the immediate relative post-exercise
changes in MVC torque were signicantly correlated to
the respective changes in the AL (r=0.96; P<0.01) so
that the subjects who had the largest decline in MVC
torque demonstrated also the largest decit in AL.
With regard to H/M and PSR of the SOL muscle, no
signicant post-exercise changes were observed. It
should be mentioned, however, that four out of the ve
subjects showed clear post-exercise declines of 39 (30)%
and 34 (15)% in H/M and PSR, respectively. These two
reex parameters had recovered by 2H similarly to the
changes in aEMG during the jumping tests.
[La

]
b
and CK
Exhausting SSC exercise induced a signicant increase in
[La

]
b
from the value at rest of 2.7 (0.6) mmoll
1
to an
immediate post-exercise value of 12.5 (3.5) mmoll
1
(P<0.05) recovering thereafter to 3.5 (1.0) mmoll
1
by
2H. Serum CK activity showed an expected individual
response with a 38 (15)% increase from 83 (46) Ul
1
to
112 (54) Ul
1
(P<0.05) during the exercise, with a de-
layed peak value of 691 (580) Ul
1
(P<0.05). It can also
be seen from Fig. 9 that the relative changes in CK at
2H post-exercise were negatively correlated with the
respective changes in KJS and AJS, showing a clear
decrease in JS in the subjects, who already had high CK
values.
Discussion
JS and jumping performance
The exhausting SSC exercise of the present study in-
duced a clear immediate and prolonged decrease in
jumping performance as well as in KJS and AJS. The
observed reduction in jumping performance (8%14%)
Fig. 4. Relative changes in hip
angular displacement (Dh
HIP
) in
jumping tests (HOP, DJ35,
DJ55) after the stretch shorten-
ing cycle exercise. *P<0.05
Signicantly dierent from the
pre-exercise value. For other
denitions see Fig. 2
111
was slightly larger than the 4%9% post-exercise
declines reported after either intense short-term SSC
exercises (Horita et al. 1996, 1999; Nicol et al. 1996) or
long-lasting ones (Avela and Komi 1998; Avela et al.
1999). As reported by Horita et al. (1996), however, the
recovery in jumping performance and JS did not follow
each other. The v
0
showed a rapid partial recovery by
2H (Fig. 2), whereas both KJS and AJS recovered over
a longer time period (Fig. 4). A faster recovery and a
smaller initial decline in v
0
, compared to KJS and AJS,
are likely to be partly due to an increased contribution
of the hip extensor muscles. As shown in Fig. 4,
a dramatic post-exercise increase in hip angular
Fig. 5. Relative changes (D%) in pre-activation, braking phase
activation and push-o phase activation of soleus (SOL), gastroc-
nemius (GA) and vastus lateralis (VL) muscles in dierent jumping
tests. *P<0.05, Signicantly dierent from the pre-exercise value.
For other denitions see Fig. 2
Fig. 6. Left: relationship between changes (D%) in triceps surae
(soleus+gastrocnemius, SOL+GA) muscle preactivity (pre-act)
and ankle joint stiness (AJS) immediately after the fatiguing
stretch shortening cycle exercise. Right: the corresponding rela-
tionship between changes in vastus lateralis (VL) preactivity and
knee joint stiness (KJS). Points represents individual mean
changes of all jumping tests (HOP, DJ35 and DJ55). For other
denitions see Fig. 2
112
displacement occurred in the jumping tests. This was
most evident in DJ55, where the angular displacement
was still twice the pre-exercise value at 7D. It must be
noted that the present fatigue task on the sledge ap-
paratus loaded mainly the knee and ankle extensor
muscles with presumably smaller loading on hip ex-
tensors due to the xed position of the upper body. It
is likely that in the jumping tests on a force plate the
subjects tried to minimize the fatigue-induced reduction
in v
0
by increasing the contribution of (probably less
fatigued) hip extensor muscles. The slower recovery
observed in JS compared to v
0
was also expected to
have resulted from the fact that larger and more pro-
longed functional decrements occurred in the braking
than in the push-o phase. The JS was calculated only
for the braking phase whereas the v
0
reected the
overall performance with more emphasis on the push-
o phase. These results are in line with the observation
of Avela et al. (1999) that a complete recovery of the v
0
took place in maximal DJ 6D after SSC exercise, while
signicant force impairments were still observed in the
braking phase.
When considering the SSC induced changes in JS, the
immediate post-exercise reduction was more dramatic in
KJS than in AJS. As shown in Fig. 6, the acute reduc-
tions in JS resulted partly from a decrease in pre-acti-
vation of both plantarexor and knee extensor muscles.
This observation is in line with the reported importance
of pre-activation in muscle and JS regulation in the SSC
locomotion (Asmussen and Bonde-Petersen 1974;
Gollhofer et al. 1984; Horita et al. 1996; Avela et al.
1996). The immediate reduction in JS was maintained
until 2H in KJS and up to 2D in AJS, and was followed
by a slow recovery. In some cases, the recovery of JS was
incomplete even 7D post-exercise. These results are in
line with the earlier observation of a long-lasting im-
pairment in KJS after exhausting SSC exercise (Horita
et al. 1996). In addition, our data demonstrated that
changes in AJS and KJS may not be identical regarding
either the magnitude of the post-exercise reduction or
the time course of recovery.
An unexpected second decline in v
0
was observed in
submaximal HOP at 7D (Fig. 1), whereas DJ35 and
DJ55 showed a gradual recovery up to 7D. As shown in
Fig. 8. Relationship between the relative (%) immediate post-
exercise changes (D) in maximal voluntary contraction (MVC)
torque and activation level (AL) of plantarexor muscles
Fig. 9. The correlation between the relative changes (D) in knee
and ankle joint stiness [JS, individual mean changes of HOP,
DJ35 and DJ55] and creatine kinase (CK) activity at 2 hours (2H)
after the stretch shortening cycle exercise. BEF Before. For other
denitions see Fig. 2
Fig. 7. Upper: relative changes (D%) in electromyogram (EMG) of
the soleus (SOL) and gastrocnemius (GA) muscles and plantar-
exor torque during the maximal voluntary contraction (MVC).
Lower: the relative changes in plantarexor torque during low
frequency (LF) electrical stimulation. *P<0.05 Signicantly dier-
ent from the pre-exercise value. For other denitions see Fig. 2
113
Fig. 5, this resulted partly from a decreased muscle ac-
tivation of the triceps surae and quadriceps muscle
groups in the push-o phase in HOP. Since all the
measured neural, mechanical and metabolic parameters
showed complete or at least progressive recovery at 7D,
it is very unlikely that the observed second decline in v
0
in HOP was due to exercise induced fatigue eects. More
likely the subjects had learned to optimize the mechan-
ical eciency of the HOP movement during the present
experiment. In line with this, the HOP frequency had
increased by 9 (8)% [from a BEF value of 1.76 (0.08) to
a 7D value of 1.90 (0.15) Hz, n.s.] and thus became
closer to the previously reported most economical fre-
quency value of approximately 2.1 Hz (Melvill Jones
and Watt 1971; Farley et al. 1991). This potential per-
formance optimization, by increasing the HOP fre-
quency, may explain the observed decrease in v
0
in HOP
at 7D. This hypothesis is supported by a high inverse
relationship (r=0.87, P<0.01) between the HOP fre-
quency and the v
0
, i.e. the higher the HOP frequency the
lower the v
0
.
Acute adaptation mechanisms
It is suggested that the immediate reduction in force
output after the present SSC exercise originated partly
from a reduced neural drive, thus suggesting central
fatigue. This can be seen from the immediate parallel
changes in activation level and MVC torque. Bigland-
Ritchie and Woods (1984) have suggested, however, that
central fatigue may not necessarily be a factor limiting
force production in intermittent submaximal exercise
using the lower limb muscles and lasting up to 20 min.
On the other hand, the authors stated that in maximal
contractions of durations of over 60 s, a declining motor
drive may well limit the force generation (Bigland-
Ritchie and Woods 1984). The fatiguing exercise of the
present study lasted on average 27 min including both
maximal and submaximal intermittent contractions. By
the last stages of the exhausting submaximal rebounding
on the sledge, the intensity of exercise had become
maximal despite the gradual decline in rebound height.
It is thus to be expected that the long duration and the
maximal intensity reached at the end of the present SSC
exercise might have induced a central fatigue. Thist
could partly explain the immediate post-exercise reduc-
tions in aEMG in the jumping and MVC tests and,
consequently, part of the reduction in force. This does
not exclude the contribution of pure metabolic fatigue
(see following discussion).
When referring to the underlying mechanisms of the
neural changes observed, a possible contribution from
supraspinal fatigue cannot be ruled out (Brasil-Neto
et al. 1994). Peripheral inuences via dierent reex
pathways have also been reported to take place after
fatiguing SSC exercises (Horita et al. 1996; Nicol et al.
1996; Avela and Komi 1998; Avela et al. 1999). In the
present study, no signicant changes were observed ei-
ther in PSR response or in H/M. However, we cannot
exclude the possibility of reduced post-exercise reex
sensitivity since four of the subjects showed a decline in
PSR and H/M.
In addition, it is very likely that part of the observed
immediate post-exercise reduction in MVC and jumping
performances might also have resulted from the failure
of contraction itself, due to the exercise-induced muscle
damage and metabolic fatigue (see Nicol and Komi
2002). Exercises involving eccentric muscle actions are
known to induce morphological changes in the intra-
cellular proteins (e.g. disorganisation or damage) (Allen
2001; Fride n and Lieber 2001) as well as a disruption of
the sarcolemma with a subsequent increase in intracel-
lular Ca
++
(Armstrong et al. 1991; Bruton et al. 1998),
thus attenuating the force-production of cross-bridge
cycles. The dramatic post-exercise decrease in the LF
torque observed, which demonstrated impairment in
excitation-contraction coupling, may have resulted from
muscle damage (Edwards et al. 1977; Jones 1981;
Takekura et al. 2001). In addition, the elevated post-
exercise [La

]
b
values in the present study were likely to
have been associated with accumulation of other meta-
bolic by-products (e.g. hydrogen ions) that can impair
the cross-bridge function (e.g. Green 1997).
Long-term adaptation mechanisms
Muscle damage and the ensuing inammation processes
(e.g. Armstrong et al. 1991) may also explain the delayed
and prolonged recovery of force parameters and KJS
and AJS in the present study. It is known that the de-
gradation of intracellular proteins begins a few hours
after the initial injury and proceeds for several days (e.g.
Armstrong et al. 1991). This may well explain the slow
recovery of the LF torque that remained reduced until
2D after the exercise. Similarly to Horita et al. (1999), a
signicant correlation was found between the changes in
CK activity and JS 2H post-exercise (Fig. 8) that would
support the inuence of muscle damage on the reduced
JS after exhausting SSC exercise.
Contrary to the prolonged impairment in force-pro-
duction, most of the EMGparameters hadfully recovered
2H post-exercise. This emphasizes the contraction failure
as an origin of the fatigue in the prolonged reduction in
force andin JS observed. On the other hand, the MVCtest
demonstrated a secondary decline in the EMG of SOL
and GA at 2D. This bimodal recovery pattern (Faulkner
et al. 1993) has frequently been observed after exercises
involving eccentric and/or SSC muscle actions (MacIn-
tyre et al. 1996; Nicol et al. 1996; Avela et al. 1999). This is
thought toresult partly froma presynaptic inhibitionof Ia
terminals through the sensitization of groups III and IV
muscle aerents in the case of muscle inammation (Nicol
et al. 1996). This is usually attributed to intramuscular
increases in pressure, temperature and in some biochem-
ical substances (Kniki et al. 1978; Kaufman and Rybicki
1987; Mense and Meyer 1988).
114
It must also be emphasized that the measurements of
neuromuscular responses to SSC exercise were carried
out primarily on plantarexor muscles, whereas the
changes in the neural and mechanical parameters (in-
cluding JS) measured in the jumping tests were more
considerable in the knee extensor muscles and in the
knee joint.
The results of the present study showed that ex-
hausting SSC exercise induced acute and prolonged JS
reduction that was more pronounced in the knee than in
the ankle joint. The immediate post-exercise decline in
JS was associated with both insucient neural drive to
muscles (central fatigue) and with contraction failure
(peripheral fatigue) through metabolic fatigue and
muscle damage. The fast recovery of the EMG with a
long-lasting impaired force production implies that the
prolonged reduction in JS was probably related to a
contraction failure that may have been caused by muscle
damage. However, any interpretation of the results of
the present study has to be made with caution as only a
small number of subjects was used.
Acknowledgements The authors would like to acknowledge the
assistance of Ms. Sirpa Roivas and Mr. Markku Ruuskanen
(technical preparations), Ms. Pirkko Puttonen and Ms. Marja-Liisa
Romppanen (data analysis) and Mr. Risto Puurtinen (biochemical
analysis). This study was supported in part by a grant (121/722/
2000) from the Ministry of Education, Finland.
References
Alexander R, Bennet-Clark H (1977) Storage of elastic strain en-
ergy in muscle and other tissues. Nature 265:114117
Allen DG (2001) Eccentric muscle damage: mechanisms of early
reduction of force. Acta Physiol Scand 171:311319
Allen DG, Westerblad H (2001) Role of phosphate and calcium
stores in muscle fatigue. J Physiol (Lond) 536:657665
Armstrong RB, Warren GL, Warren JA (1991) Mechanisms of
exercise-induced muscle ber injury. Sports Med 12:184207
Asmussen E, Bonde-Petersen F (1974) Storage of elastic energy in
skeletal muscles in man. Acta Physiol Scand 91:385392
Avela J, Komi PV (1998) Reduced stretch reex sensitivity and
muscle stiness after long-lasting stretch-shortening cycle ex-
ercise in humans. Eur J Appl Physiol 78:403410
Avela J, Santos PM, Komi PV (1996) Eects of dierently induced
stretch loads on neuromuscular control in drop jump exercise.
Eur J Appl Physiol 72:553562
Avela J, Kyro la inen H, Komi PV, Rama D (1999) Reduced reex
sensitivity persist several days after long-lasting stretch-short-
ening cycle exercise. J Appl Physiol 86:12921300
Belli A, Kyro la inen H, Komi PV (2002) Moment and power of
lower limb joints in running. Int J Sports Med 23:136141
Bigland-Ritchie B, Woods JJ (1984) Changes in muscle contractile
properties and neural control during human muscular fatigue.
Muscle Nerve 7:691699
Bongiovanni LG, Hagbarth K-E (1990) Tonic vibration reexes
elicited during fatigue from maximal voluntary contractions in
man. J Physiol (Lond) 423:114
Bosco C, Komi PV, Ito A (1981) Prestretch potentiation of human
skeletal muscle during ballistic movement. Acta Physiol Scand
111:135140
Bosco C, Tarkka I, Komi PV (1982) Eect of elastic energy and
myoelectric potentiation of triceps surae during stretch-short-
ening cycle exercise. Int J Sports Med 3:137140
Brasil-Neto JP, Cohen LG, Hallet M (1994) Central fatigue as
revealed by post exercise decrement of motor evoked potentials.
Muscle and Nerve 17:713719
Bruton JD, La nnergren J, Westerblad H (1998) Mechanisms un-
derlying the slow recovery of force after fatigue: importance of
intracellular calcium. Acta Physiol Scand 162:285293
Cavagna GA (1977) Storage and utilization of elastic energy in
skeletal muscle. Exerc Sports Sci Rev 5:89129
Coupland M, Puchert E, Ranatuga K (2001) Temperature depen-
dence of active tension in mammalian (rabbit psoas) muscle -
bers: eect of inorganic phosphate. J Physiol (Lond) 536:879891
Darques JL, Decherchi P, Jammes Y (1998) Mechanisms of fa-
tigue-induced activation of group IV muscle aerents: the roles
played by lactic acid and inammatory mediators. Neurosci
Lett 257:109112
Demster WT (1955) Space requirements of the seated operator
(WADC TR-55159). Aerospace Medical Research Laboratory
(NTIS No. AD-87892), Wright-Patterson Air Force Base, Ohio
Dietz V, Schmidtbleicher D, Noth J (1979) Neuronal mechanisms
of human locomotion. J Neurophysiol 42:12121222
Duchateau J, Hainaut K (1993) Behaviour of short and long la-
tency reexes in fatigued human muscles. J Physiol (Lond)
471:787799
Edwards RG, Hill DK, Jones DA (1977) Fatigue of long duration
in human skeletal muscle after exercise. J Physiol (Lond)
272:769778
Farley C, Blickhan R, Saito J, Taylor CR (1991) Hopping fre-
quency in humans: a test how springs set stride frequency in
bouncing gaits. J Appl Physiol 71:21272132
Faulkner JA, Brooks SV, Opiteck JA (1993) Injury to skeletal
muscle bers during contractions: conditions of occurrence and
prevention. Phys Ther 73:911921
Fride n J, Lieber RL (2001) Eccentric exercise-induced injuries to
contractile and cytoskeletal muscle bre components. Acta
Physiol Scand 171:321326
Gollhofer A, Schmidtbleicher D, Dietz V (1984) Regulation
of muscle stiness in human locomotion. Int J Sports Med
5:1922
Gollhofer A, Komi PV, Miyashita M, Aura O (1987) Fatigue
during stretch-shortening cycle exercises: changes in mechanical
performance of human skeletal muscle. Int J Sports Med 8:71
78
Gollhofer A, Strojnik V, Rapp W, Schweizer L (1992) Behaviour of
triceps surae muscle-tendon complex in dierent jump condi-
tions. Eur J Appl Physiol 64:283291
Green HJ (1997) Mechanisms of muscle fatigue in intense exercise.
J Sports Sci 15:247256
Horita T, Komi PV, Nicol C, Kyro la inen H (1996) Stretch short-
ening cycle fatigue: interactions among joint stiness, reex,
and muscle mechanical performance in the drop jump. Eur J
Appl Physiol 73:393403
Horita T, Komi PV, Nicol C, Kyro lainen H (1999) Eect of ex-
hausting stretch-shortening cycle exercise on the time course of
mechanical behaviour in the drop jump: possible role of muscle
damage. Eur J Appl Physiol 79:160167
Hugon M (1973) Methodology of the Homann reex in man. In:
Desmedt JE (ed) New developments in electromyography and
clinical neurophysiology. Karger, Basel, pp 277293
Jones DA (1981) Muscle fatigue due to changes beyond the neu-
romuscular junction. In: Porter R, Whelan J (eds) Human
muscle fatigue: physiological mechanisms (Ciba Foundation
Symposium 82). Pitman Medical, London, pp 178196
Kaufman MP, Rybicki KJ (1987) Discharge properties of group III
and IV muscle aerents: their responses to mechanical and
metabolic stimuli. Circ Res [Suppl] 61:160165
Kniki KD, Mense S, Schmidt RF (1978) Responses of group IV
aerent units from skeletal muscle to stretch, contraction and
chemical stimulation. Exp Brain Res 31:511522
Komi PV (1984) Physiological and biomechanical correlates of
muscle function: eects of muscle structure and stretch-short-
ening cycle on force and speed. Exerc Sports Sci Rev 12:81121
115
Komi PV, Gollhofer A (1997) Stretch reexes can have an im-
portant role in force enhancement during SSC exercise. J Appl
Biomech 13:451460
Kuitunen S, Komi PV, Kyro la inen H (2002) Knee and ankle joint
stiness in sprint running. Med Sci Sports Exerc 34:166173
Kyro la inen H, Komi PV (1995) The function of neuromuscular
system in maximal stretch-shortening cycle exercise: compari-
son between power- and endurance-trained athletes. J Electro-
myogr Kinesiol 5:1525
Maceeld G, Hagbarth K-E, Gorman R, Gandevia SC, Burke D
(1991) Decline in spindle support to a-motoneurones during
sustained voluntary contractions. J Physiol (Lond) 440:497512
MacIntyre DL, Reid WD, Lyster DM, Szasz IJ, McKenzie DC
(1996) Presence of WBC, decreased strength, and delayed
soreness in muscle after eccentric exercise. J Appl Physiol
80:10061013
Melvill Jones G, Watt DGD (1971) Observation on the control of
stepping and hopping movements in man. J Physiol (Lond)
219:729737
Mense S (1977) Nervous outow from skeletal muscle following
chemical noxious stimulation. J Physiol (Lond) 267:7588
Mense S, Meyer H (1988) Bradykinin-induced modulation of the
response behaviour of dierent types of feline group III and IV
muscle receptors. J Physiol (Lond) 398:4963
Merton PA (1954) Voluntary strength and fatigue. J Physiol
(Lond) 123:553564
Nicol C, Komi PV (2002) Stretch-shortening cycle fatigue and its
inuence on force and power production. In: Komi PV (ed)
Strength and power in sport. Blackwell Scientic Publications,
Oxford, in press
Nicol C, Komi PV, Marconnet P (1991a) Fatigue eects of mara-
thon running on neuromuscular performance. II. Changes in
force, integrated electromyographic activity and endurance ca-
pacity. Scand J Med Sci Sports 1:1824
Nicol C, Komi PV, Marconnet P (1991b) Fatigue eects of mar-
athon running on neuromuscular performance. I. Changes in
muscle force and stiness characteristics. Scand J Med Sci
Sports 1:1017
Nicol C, Komi PV, Horita T, Kyro la inen H, Takala TES (1996)
Reduced stretch-reex sensitivity after exhausting stretch-
shortening cycle exercise. Eur J Appl Physiol 72:401409
Proske U, Morgan DL, Gregory JE (1993) Thixotropy in skeletal
muscle and in muscle spindles: a review. Prog Neurophysiol
41:705721
Pullinen T, Leynaert M, Komi PV (1997) Neuromuscular function
after marathon. XVIth International Society of Biology Con-
gress, University of Tokyo, Tokyo, (abstracts) p 174
Rotto DM, Kaufman MP (1988) Eect of metabolic products of
muscular contraction on discharge of group III and IV aer-
ents. J Appl Physiol 64:23062313
SENIAM (1999) European recommendations for surface electr-
omyography. In: Hermens H, Freriks B, Merletti R, Stegeman
D, Blok J, Rau G, Disselhorst-Klug C, Ha gg D (eds) ISBN 90
75452152
Strojnik V, Komi PV (1998) Neuromuscular fatigue after maximal
stretch-shortening cycle exercise. J Appl Physiol 84:344350
Synoway LI, Hill JM, Pickar JG, Kaufman MP (1993) Eects of
contraction and lactic acid on the discharge of group III muscle
aerents in cat. J Neurophysiol 69:10531059
Takekura H, Fujinami N, Nishizawa T, Ogasawara H, Kasuga N
(2001) Eccentric exercise-induced changes in the membrane
systems involved in excitation-contraction coupling in rat
skeletal muscle. J Physiol (Lond) 533:571583
Westerblad H, Allen D, La nnergren J (2002) Muscle fatigue: lactic
acid or inorganic phosphate the major cause? News Physiol Sci
17:1721
116