CIfnical Scfence
(1983)
64.55-62
55
Pain and fatigue after concentric and eccentric muscle
contractions
D.
J.
NEWHAM, K. R. MILLS,
B.
M. QUIGLEY’
AND
R.
H.
T.
EDWARDS
Department
of
Human Metabolism. The Rayne Institute, University College London Medical
School,
London, and
‘Department
of
Movement Studies. University
of
Queensland, Brisbane, Australia
(Received
10
MarchlI9
July
1982; accepted
2
August 1982)
Summary
1.
Normal subjects performed
a
step test in
which the quadriceps of one leg contracted
concentrically while the contralateral musde
contracted eccentrically.
2.
Maximal voluntary force and the force:fre-
quency relationship were altered bilaterally as
a
result of the exercise, the changes being greater in
the muscle which had contracted eccentrically.
Recovery occurred over
24
h.
3.
Electromyographic studies using three sites
on each muscle showed an increase in electrical
activation during the exercise only
in
the muscle
which was contracting eccentrically. Recovery
followed
a
time course similar to that of the
contractile properties.
4.
Pain and tenderness developed only in the
muscle which had contracted eccentrically. Pain
was first noted approximately
8
h after exercise
and was maximal at approximately
48
h after
exercise, at which time force generation and
electrical activation had returned to pre-exercise
values.
5.
Eccentric contractions cause more pro-
found changes in some aspects of muscle function
than concentric contractions. These changes
cannot be explained in simple metabolic terms,
and it is suggested that they are the result of
mechanical trauma caused by the high tension
generated
in
relatively few active fibres during
eccentric contractions.
Key words: eccentric contractions, low-fre-
quency fatigue, muscle pain.
Correspondence: Professor
R.
H.
T.
Edwards,
Department
of
Human Metabolism, The Rayne
Institute, University College London Medical School,
University Street, London WClE
6JJ.
Abbreviations: EMG, electromyograph; IEMG,
integrated electromyograph.
Introduction
Muscle pain occurring
24-48
h after unac-
customed exercise is
a
phenomenon familiar to
most individuals, but the mechanisms re-
sponsible for its production are uncertain.
Asmussen
[ll
first indicated that eccentric con-
tractions (those in which the muscle is lengthened
during contraction) are particularly associated
with pain and soreness.
In
the intervening period
many workers have found this to be an interest-
ing model as it is well established that both the
metabolic cost
12-51
and the electrical activity
required to produce
a
given tension
[6-91
are less
under eccentric conditions than concentric. Komi
[
101,
however, reported the relationship between
the integrated electromyograph (IEMG) and
percentage of maximum force to be similar
whichever type of contraction was used. The fact
remains that eccentric rather than concentric
contractions predispose to delayed onset, post-
exercise pain that is not accounted for in terms of
metabolism.
The ‘torn tissue’ hypothesis of Hough
[ll]
suggested that pain resulted from structural
damage in the muscle; de Vries
[121
proposed
tonic spasms in localized motor units and both
Komi
[I31
and Asmussen
111
put forward
overstretching of the connective tissue elements
as
the cause of pain.
Repetitive, fatiguing isometric and dynamic
contractions have been shown to produce
specific, long-lasting alterations in contractile
properties of muscle such that the force:fre-
quency curve is shifted to the right and electrical
0143-5221/83/010055-08$2.00
@
1983 The Biochemical Society
and
the Medical Research Society
56
D.
J.
Newham et al.
stimulation at low frequency (1-20
Hz)
results in
decreased force generation when compared with
the fresh muscle, but the force generated by
high-frequency stimulation is relatively preserved
[
141. This type of fatigue, termed ‘low-frequency
fatigue’, has been demonstrated
in
the quadriceps,
adductor pollicis, diaphragm
[
151 and sterno-
mastoid
[
161. Although the underlying mech-
anism is not clear, it was assumed that the
amount
of
low-frequency fatigue produced in a
muscle was related to the work done by that
muscle. Recent work [17l has revealed that
eccentric contractions caused greater low-fre-
quency fatigue than concentric.
In this study, normal subjects have performed
a step test in which the quadriceps muscle of one
leg worked concentrically and the contralateral
muscle worked eccentrically. The effects of these
two types of contraction on the IEMG, voluntary
force and contractile properties
of
the muscle
have been investigated. The degree and dis-
tribution
of
tenderness over the surface
of
the
muscle has been measured
(181
in an attempt
to
define the painful tissue.
Methods
Subjects
Four
healthy, normal subjects performed the
experiments, three males and one female, the age
range being 3
1-45
years (mean 36-25 years).
Step test
Subjects performed a step test for
I5
or
20
min, using a
46
cm step. The stepping pattern
was designed
so
that the quadriceps
of
one leg
contracted concentrically (stepping up) through-
out the test, while the contralateral muscle
contracted eccentrically (stepping down). A rate
of
15 cycleslmin was used and
an
electronic
metronome provided audible timing clicks,
so
that each stepping phase lasted 1
s.
During the
exercise period a total height of 103.5
m
was
ascended. Subjects were encouraged to fully
control each eccentric contraction and as far as
possible to maintain
a
constant stepping rate.
Force measurements
The force produced by electrically stimulated
and maximal voluntary isometric contractions
was measured by using previously described
techniques [191.
The force :frequency characteristics
of
the
muscle were monitored by percutaneous stimu-
lation at 1
Hz
(for
5
s)
and 10,20,50 and
100
Hz
(for
2
s)
by using square wave pulses
of
50
ps.
Maximal voluntary force measurements and
electrical stimulation of the quadriceps muscles of
both legs were carried out before exercise, then
2,
10
and 30 min and
1,
5,
24 and 48 h after
exercise.
Electromyograp hy
Areas over rectus femoris, vasti medialis and
lateralis on each leg were prepared by abrasion
and alcohol swabs to lower the skin resistance to
less than
5
kohm. These areas were marked
so
that identical sites would be used
on
subsequent
testing. Silver/silver chloride cup electrodes were
filled with electrode jelly and taped in place.
Unipolar recordings were made from these sites
and amplified with reference to an electrode
placed over the lower lumbar spine in the midline.
Signals were amplified (S.E. Labs, type 4901) and
band pass filtered between
0.2
Hz
and 10 kHz,
and displayed
on
a
U.V.
oscillograph and re-
corded
on
light-sensitive paper. The six raw
signals were integrated over 300 ms periods and
similarly displayed. Recordings were made from
these sites at intervals during stepping.
In
order to investigate any changes in acti-
vation patterns as a result
of
the exercise,
electrical activity of the three muscles
on
each leg
was recorded during active, submaximal con-
tractions during knee extension from 90° to full
extension, which was held for 2
s,
with
a
3 kg
weight attached to the foot. These recordings
were made before and at intervals after exercise.
To
study the relationship between muscular
activity and joint angle during stepping and the
submaximal knee extension tests, electronic
goniometers were used.
A
rotary potentiometer
with
a
linear response was mounted as the pivot
between two long Perspex arms. The gonio-
meters were placed laterally on each leg with the
potentiometer sited over the fulcrum of the knee
joint and the Perspex arms taped in place along
the femur and fibula. A signal, proportional to the
knee angle, was displayed on the
U.V.
oscillo-
scope with the EMG and simultaneously re-
corded. Rate of knee extension was kept as
constant as possible during the test by displaying
to the subject a signal proportional to angular
velocity.
Measurement
of
severity and distribution
of
muscle tenderness
A polythene sheet marked with a grid of
intercepts
2
cm apart, to be used as test sites, was
Muscle pain after exercise
57
wrapped around the thigh, the skin of which was
marked to ensure constant positioning in sub-
sequent tests. A round-ended, wooden probe
(2
cm diameter) was attached to a strain gauge and
the amplified force signal was displayed on a
U.V.
oscillograph. At each test site, a gradually
increasing force was applied up to a maximum of
40
N. The subject was asked to indicate verbally
when the sensation of pressure changed to one of
discomfort, whereupon the probe was immedi-
ately withdrawn. If no indication was given at
a
deflection on the oscillograph proportional to
40
N,
tenderness was considered not to be present at
that site. Each site was tested in a defined order,
enabling a record to be made of the degree of
tenderness over the whole surface of the muscle.
From these records maps were drawn showing
the degree and distribution of muscle tenderness.
Although the accuracy of localization by muscle
nociceptors is not well defined, the fact that the
receptive areas of mechanical nociceptors are
spot-like
[201
and also that subjects are well able
to localize the sites of contusions and needle
biopsies in muscle suggests that the degree of
localization is adequate to indicate the sites of
muscle trauma.
Results
No
significant difference was found between the
15 and
20
min exercise periods, therefore the
following are combined results of both periods.
Force changes
(a) Maximal voluntary force. Maximal volun-
tary force was reduced in both legs aRer exercise,
the reduction being significant only in the muscle
which had contracted eccentrically when pre-
exercise values were compared with those at
2
and 10 min after exercise
(P
<
0401). Force did
not recover to pre-exercise values until
24
h after
exercise.
(b) Stimulated forces. As an index of low-
frequency fatigue, we have used the ratio of the
forces produced by a low stimulation frequency
(10
Hz)
to
the force produced by a high
stimulation frequency
(50
Hz), expressed as a
percentage (T10/50%).
A significant decrease in T10/50% was found
in the quadriceps of both legs
(P
<
0.001); the
fall was more marked in the muscle which had
contracted eccentrically in the exercise period
(Fig. 1). The difference of the T10/50% between
the two muscles was significant at
2
min
(P
<
0.02
>0.01) 10 min
(P
<
0.025
>
0.002)
after exercise and most highly significant 1 h after
Concentric
Frdl
Fatigued
Eccentric
Fresh
Fatigued
StimuWs
I
1
~IO~~~~
Time($
0
5
7
9
1113
FIG.
1.
Force generation in response to electrical
stimulation at
1,
10,
20,
50
and
100
Hz
in the
quadriceps before and
10
min after
a
20
min period
of
stepping in which one muscle contracted concentri-
cally and the other eccentrically. Female subject,
32
years.
exercise
(P
<
0.001) as the muscle which had
contracted concentrically began to recover.
Twenty-four hours after exercise there was no
significant difference between the two muscles,
although when compared with the pre-exercise
values, the T10/50% had not fully recovered
(Fig.
2).
Electromyography
In both of the two subjects studied no
significant increase was seen in the IEMG of the
concentrically contracting muscle during the
stepping period. In contrast, a progressive in-
crease in the IEMG of all sites monitored was
found in the eccentrically contracting muscle
throughout the exercise period (Fig.
3).
Similar changes in electrical activation were
seen during the submaximal knee extension test
when pre- and post-exercise data were compared
(Fig.
4);
only the muscles which had contracted
eccentrically showed increased electrical acti-
vation for the generation of
a
given muscular
tension.
There was no significant change in the ratio
01
the contributions of the three muscle components
to that of the total measured quadriceps activity,
58
D.
J.
Newham et al.
100
-
-
m
.-
I
.-
.B
80
c
0
c
60
s
-
m
e,
2
2
40
.
-
:
20
2
U
0
lh 5h
'24
h
Time after exercise
FIG.
2.
Changes in the relationship between force generated by
10
and
50
Hz
stimulation
(expressed
as
a percentage of pre-exercise values) after
15
or
20
min stepping. Mean values
k
SEM
are shown for seven subjects.
160
Concentric
120
I
I
I
1
I
I
0
5 10 15
20
0
5
10
15
20
Duration of exercise
(min)
I
I
J
FIG.
3.
Integrated electromyogram
(IEMG)
from three sites
of
both quadriceps recorded at
5
min
intervals during a
20
min period
of
stepping. Each point is the mean of three consecutive
concentric
or
eccentric contractions and is expressed as a percentage of the pre-exercise value.
W,
Rectus fernoris;., vastus medialis;
6,
vastus lateralis. Male subject,
45
years.
in either leg, during stepping, in the immediate
post-exercise phase
or
during the period when the
leg was painful. The simultaneous recording of
joint angle and IEMG during stepping (Fig.
5)
in
the concentrically contracting muscle showed
that the main peak of electrical activity occurs
during the stepping up phase, with a smaller burst
as the opposite muscle lowered the body weight
to the ground, to be supported by the former.
The eccentrically contracting muscle showed
two peaks in each cycle which were approxi-
mately similar in amplitude
to
each other, and
shorter in duration than the main peak in the
concentrically contracting muscle. One peak
occurred during the eccentric contraction itself,
and the other at the time when the same leg was
taking part
of
the body weight after the opposite
leg had raised the body up on the step.
The submaximal extension test showed an
increase in IEMG at
all
knee-joint angles be-
Muscle pain after exercise
59
300
250
u
>
-
!A
.-
200
8
r
n
L
150
c
0
E
!i
100
50
Eccentric
300
250
200
150
100
50
I
I
1
0123456
Concentric
I
I
I
I I
I
0123456
Time after exercise
(h)
FIG.
4.
Integrated electromyograph (IEMG) from three sites of both quadriceps during a
submaximal knee extension held
for
2
s,
after a
20
min period of stepping. Each value is expressed
as a percentage
of
the pre-exercise value. See Fig.
3
for explanation of symbols. Male subject,
45
years.
Concentric
t
t
t
V.
lateralis
/
Eccentric
V.
lateralis
\
4s
FIG.
5.
Simultaneously recorded knee angles and
IEMG from vastus lateralis of both legs during
stepping.
tween
0
and
90°,
in addition to the amount of
electrical excitation required to maintain full
extension over a
2
s
period (Fig.
6).
At
no
time throughout the testing period was
there evidence of spontaneous electrical activity
when the muscle was at rest.
Muscle pain and tenderness
Subjective pain was reported by all subjects
in
Concentric
Before exercise
-,
angle
900
I,
V.
lateralis
..
R.
femoris
A
V.
medialis
30
min after exercise
,
:
V.
lateralis
I,
R.
fernoris
I,
V.
medialis
2s
-
Eccentric
I,
-
I.
FIG.
6.
Integrated EMG from three sites on both
quadriceps during knee extension before and
30
min
after stepping. Male subject,
45
years.
60
24
h
I.
\
Force at which tenderness reported
(Nl
p:
I
36-40
D.
J.
Newham et al.
dA
h
Distal
28-35
2&27
12-19
4
cm
FIG.
7.
Distribution and severity
of
tenderness
24,48
and
55
h
after eccentric contractions
in
the
right
quadriceps. Female subject,
32
years.
the quadriceps muscle which had contracted
eccentrically in the,step test.
No
pain was noticed
in
the quadriceps of the opposite leg, although
two subjects reported pain in the calf muscles
of
that leg. The pain was first apparent approxi-
mately
8-10
h after exercise and reached
maximal intensity between
24
and
48
h after
exercise in different subjects. It was also found to
be uncomfortable to descend stairs and especi-
ally to contract the muscle isometrically in a fully
shortened position, although isometric contrac-
tion in the mid-length position was less painful.
Strain gauge measurements showed that
initially tenderness was primarily located at the
distal, medial and lateral parts
of
the quadriceps
and along its lateral margin, with relative sparing
of the central and proximal medial regions. At
peak intensity the tenderness was more diffuse,
but reflected the same pattern. As tenderness
diminished
a
more clear regional localization was
again seen
as
in the early stages
of
pain (Fig.
7).
Both soreness and tenderness had disappeared in
all subjects by the fourth day after exercise.
All subjects experienced
a
feeling of weakness
and instability in the immediate post-exercise
period only in the muscle which had contracted
eccentrically. This sensation was noted at the end
of
exercise and lasted
for
approximately
2
h. It
was particularly noticeable on performing ec-
centric contractions, i.e. descending stairs.
Discussion
Despite the relatively low energy cost
of
eccentric
contractions, they are capable
of
causing more
profound changes in some aspects
of
muscle
function, especially the force
:
frequency curve,
than concentric contractions, which clearly can-
not be explained in simple metabolic terms.
The fact that greater tension per muscle fibre is
generated under eccentric contraction conditions
[21,
221
provides
a
situation where relatively few
fibres are recruited and are producing relatively
large forces. In this situation the uneven mech-
anical stresses produced in the muscle and its
attachments can be imagined
to
predispose to
physical damage as with the weakest link in a
chain.
Mechanical damage
to
the sarcoplasmic reti-
culum resulting in less calcium release for each
excitatory action potential has been suggested as
the cause
of
low-frequency fatigue
[23],
and
if
this is the case it is consistent with our results.
Komi
&
Rusko
[241
reported that with
isokinetic exercise at the forearm flexors ec-
centric contractions cause a greater reduction in
Muscle pain after exercise
61
maximal voluntary force than concentric con-
tractions, and in contrast to our findings found
similar IEMG changes with both types of
contraction. They concluded that the differences
in force changes were due to extreme mechanical
loading of the elastic components during ec-
centric contractions. This mechanical stress and
trauma may well be the explanation of the
reduction in maximal voluntary force, increase in
electrical activation for
a
given muscle tension
and profound low-frequency fatigue, changes
which were all more' marked in the muscle which
had contracted eccentrically.
The distribution of tenderness revealed that the
muscle bellies are relatively spared, and the areas
of musculo-tendinous attachment are the main
sites of pain and tenderness. These findings are in
agreement with the conclusions of Asmussen that
the cause of pain is due to over-stretching of
elastic non-contractile tissues, which was sup-
ported by Abraham
[251,
who reported rises in
hydroxyproline
:
creatinine ratios at peak muscle
soreness with no changes as
a
result of con-
centric contractions. The same author was able to
correlate myoglobin release with exercise inten-
sity, but not with soreness, and this argues
against the theory that the muscle itself is not the
sole tissuq responsible for this type of muscle
pain.
In agreement with other workers we were not
able to detect any evidence of localized muscle
spasm during pain as reported by de Vries. The
relative contribution of rectus femoris, the medial
and lateral vasti to the total measured electrical
activity of the muscle did not significantly alter
during either short-term fatigue or delayed onset
pain, thus providing no evidence of changes in
recruitment patterns with fatigue
or
inhibition of
painful areas.
An interesting, but poorly defined, pheno-
menon is the feeling of weakness and instability
experienced for
a
few hours immediately after
exercise only in the muscle which had contracted
eccentrically. Further work (unpublished) has
indicated that this sensation is an indication of
pain to follow, and is presumably a reflection of
profound low-frequency fatigue with inappro-
priate forces being generated by the relatively low
normal physiological firing frequency
126,271.
In conclusion, eccentric muscle contractions
have marked effects, initially on the contractile
properties and force generating capabilities of
muscle, and result in pain of delayed onset. These
findings are accountable
for
in terms of the high
forces generated by relatively few muscle fibres
and the transmission of these uneven forces to the
non-contractile tissues with resultant mechanical
damage, and are not related to the metabolic
energy cost of the contractions. The different time
courses of these phenomena may reflect an
inflammatory process in the musculo-tendinous
areas andlor the different turnover rate of the two
tissues.
Acknowledgments
We thank Professor
J.
P Moss
(U.C.H.
Dental
School) for the use of his EMG apparatus, and
Mr D. St Andrew, who designed both the EMG
and integrator circuits as well
as
the gonio-
meters. The support of the Wellcome Trust and
Muscular Dystrophy Group of Great Britain is
gratefully acknowledged.
References
1
I
I
ASMUSSEN, E. (1956) Observations
on
experimental muscular
soreness.
Acta Rheumatologica Scandinavica,
2,109-1 16.
121 ABBOTT, B.C., BIQLAND, B.
&
RITCHIE, J.M. (1952)
The
physiological cost of negative work.
Journal
of
Physiology
(London),
II7.380-390.
131 KNWQEN, H.G., BONDE-PETERSON, F.
&
KLAUSEN, K.
(1971) Oxygen uptake and heart rate response to exercise
performed with concentric and eccentric muscle contractions.
Medicine and Science
in
Sports.
3, 1-5.
141 ASMUSSEN, E. (1952) Positive and negative muscular work.
Acta Physiologica Scandinavica,
28.364-382.
151
DAVIES, C.T.M.
&
BARNES, C. (1972) Negative (eccentric)
work.
VI.
Physiological responses to walking uphill and
downhill
on
a motor driven treadmill.
Ergonomics,
IS,
121-131.
161 BIGLAND, B.
&
LIPPOLD, 0.CJ. (1954) The relation between
force, velocity and integrated electrical activity in human
muscles.
Journal
of
Physiology (London),
123,214-224.
171 BASMNIAN, J.R. (1967)
Muscles Alive: Their Funcfion
Revealed by Electromyography,
2nd edn, p. 221. Williams and
Wilkins, Baltimore.
181
ASMUSSEN, E. (1953) Positive and negative work.
Acfa
Physiologica Scandinavica.
28.364-382.
191 BIGLAND-RITCHIE, B.
&
WOODS,
J.J. (1976) Integrated EMG
and
0,
uptake during positive and negative work.
Journal
of
Physiology (London),
260,267-277.
1101 KOMI, P.V. (1973) Relationship between muscular tension,
EMG and velocity of contraction under concentric and
eccentric conditions.
In:
New Developments in Electromyo-
graphy and Clinical Neurophysiology,
pp. 596-606.
Ed.
Desmedt, J. Karger, Easel.
I
1 11 HOUQH, T. (1902) Ergographic studies in muscular
soreness.
American Journal
of
Physiology,
7.76-92.
I121 DE
VRlEs,
H.A. (1966) Quantitative electromyographic
investigation of the..spasm theory of muscle pain.
American
Journal
of
Physical Medicine,
45.119-134.
1131 KOMI,
P.V.
&
BUSKIRK. E.R. (1972) ElTect of eccentric and
concentric muscle conditioning
on
tension and electrical
activity of human muscles.
Ergonomics,
IS.
4 17-434.
I141 EDWARDS, R.H.T., Hru. D.K.
&
JONES, D.A. (1977)Fatigue
of
long
duration in human skeletal muscle alter exercise.
Journal
of
Physiology (London),
212,769-778.
1151 MOXHAM, J.. MORRIS, A.J.R.,
SPIRO.
S.,
EDWARDS, R.H.T.
&
GREEN,
M. (1980) The contractile properties and fatigue of the
diaphragm in man.
Clinical Science,
S8,6~.
I161 MOXHAM, J.,
WILES,
C.M., NEWHAM, D.
&
EDWARDS, R.H.T.
(1980) Stemomastoid muscle function and fatigue in man.
Clinical Science,
59.463-468.
Greater low frequency fatigue produced by eccentric than
concentric muscle contractions.
Journal
of
Physiology
(London),
317.17~.
1171 EDWARDS, R.H.T., MILLS, K.R.
&
NEWHAM, DJ. (1981)
62
D.
J.
hlewham et al.
1181
EDWARDS, R.H.T., MILLS, K.R.
&
NEWHAM,
D.J.
(1981)
Measurement of severity and distribution of experimental
muscle tenderness.
JournalofPhysiology (London),
317.1P-2P.
1191
EDWARDS. R.H.T.. Y0u~ci.A.. HOSKING, G.P.
&
JONES,
D.A.
(1977)
Human skeletal muscle function: description of tests
and normal values.
Clinical Science and Molecular Medicine,
52,283-290.
1201
KUMAZAWA,
T.
&
MIZUMUNA, K.
(1977)
Thin fibre receptors
responding to mechanical, chemical and thermal stimulation
in
the skeletal muscle of the dog.
Journal
of
Physiology (London),
273,179-194.
1211
KATZ, B.
(1939)
The relation between force and speed in mus-
cular contraction.
Journal
of
Physiology (London),
96,46-64.
1221
CURTIN, N.
&
DAvIes,
R.E.
(1973)
Chemical and mechanical
changes during stretching of activated frog skeletal muscle.
Cold Spring Harbour Symposia
on
Quantitative Biology,
37,
6
19-626.
1231
JONES, D.A.
(1981)
Muscle fatigue due to changes beyond
the neuromuscular junction.
In:
Human Muscle Fatigue:
Physiological Mechanisms,
pp.
178-196.
Ed. Porter, R.
&
Whelan,
J.
Ciba Symposium
82.
Pitman Medical Press,
London.
1241
KOMI, P.V.
&
RUSKO.
0.
(1974)
Quantitative evaluation
of
mechanical and electrical changes during fatigue loading of
eccentric and concentric work.
Scandinavian Journal
of
Rehabilitation Medicine,
Suppl.
3, 121-126.
1251
ABRAHAM, W.M.
(1977)
Factors in delayed muscle soreness.
Medicine and Science in Sports,
9,
11-26.
1261
CLAMANN, H.P.
(1970)
Activity of single motor units during
isometric tension.
Neurology,
20,254-260.
1271
MILNER-BROWN, H.S.,
Smm,
R.B.
&
YEMM, R.
(1973)
The
contractile properties of human motor units during voluntary
isometric contractions.
Journal
of
Physiology (London),
228,
28.5-306.
