Journal of Sports Sciences, 1999, 17, 231 238

EV ects of cold water immersion on the symptoms of

exercise-induced muscle damage
ROGER ESTON* and DANIEL PETERS
School of Sport, Health and Physical Education Sciences, University of Wales, Victoria Drive, Bangor,
Gwynedd LL57 2EN, UK

Accepted 29 March 1998

Cryotherapy is an eVective treatment for acute sports injury to soft tissue, although the eV ect of cryotherapy
on exercise-induced muscle damage is unclear. The aim of this study was to assess the eVects of cold water
immersion on the symptoms of exercise-induced muscle damage following strenuous eccentric exercise. After
performing a bout of damage-inducing eccentric exercise (eight sets of Wve maximal reciprocal contractions at
0.58 rad s- 1) of the elbow Xexors on an isokinetic dynamometer, 15 females aged 22.0 2.0 years (mean s)
were allocated to a control group (no treatment, n = 7) or a cryotherapy group (n = 8). Subjects in the cryotherapy group immersed their exercised arm in cold water (15C) for 15 min immediately after eccentric exercise
and then every 12 h for 15 min for a total of seven sessions. Muscle tenderness, plasma creatine kinase activity,
relaxed elbow angle, isometric strength and swelling (upper arm circumference) were measured immediately
before and for 3 days after eccentric exercise. Analysis of variance revealed signiWcant (P < 0.05) main eV ects for
time for all variables, with increases in muscle tenderness, creatine kinase activity and upper arm circumference,
and decreases in isometric strength and relaxed elbow angle. There were signiWcant interactions (P < 0.05) of
group time for relaxed elbow angle and creatine kinase activity. Relaxed elbow angle was greater and creatine
kinase activity lower for the cryotherapy group than the controls on days 2 and 3 following the eccentric exercise.
We conclude that although cold water immersion may reduce muscle stiV ness and the amount of post-exercise
damage after strenuous eccentric activity, there appears to be no eVect on the perception of tenderness and
strength loss, which is characteristic after this form of activity.
Keywords: cold water immersion, cryotherapy, muscle damage.

Introduction
Exercise-induced muscle damage, which commonly
occurs after strenuous eccentric muscle action, consists
of a dull aching pain, stiVness, tenderness and a prolonged loss of muscle strength. The symptoms tend to
develop within the Wrst 24 h post-exercise, peak between
24 and 72 h and then subside after 5 7 days (Ebbeling
and Clarkson, 1989; Armstrong, 1990; Cleak and
Eston, 1992a; Nosaka and Clarkson, 1995). As eccentric muscle actions are usually associated with a higher
force-to-activation ratio, the `loading proWle places a
high stress on the tissues involved and is most likely
a primary factor of muscle damage (Enoka, 1996).
Damage includes disruption of the sarcolemma,

* Author to whom all correspondence should be addressed. e-mail:

r.g.eston@bangor.ac.uk
0264 0414/99

1999 E & FN Spon

fragmentation of the sarcoplasmic reticulum, lesions

of the plasma membrane, cytoskeletal damage and
swollen mitochondria (Stauber, 1989; Armstrong,
1990; Friden and Leiber, 1992), although this is restricted to a relatively small proportion of Wbres within
the aVected muscle.
Additional symptoms of exercise-induced muscle
damage include swelling and a decrease in the range of
motion. Studies on the arm have observed an increase
in volume (Talag, 1973), an increase in circumference
measurements (Friden et al., 1988; Hill and Richardson,
1989; Cleak and Eston, 1992b; Howell et al., 1993;
Nosaka and Clarkson, 1997) and a decrease in active
and passive range of motion in both Xexion and extension of the elbow (Clarkson and Newham, 1994). In
addition, when the arm is hanging in a relaxed position,
the elbow angle is reduced (Cleak and Eston, 1992b;
Clarkson and Newham, 1994; Nosaka and Clarkson,
1997).

232
Several studies have been undertaken to Wnd ways
to alleviate and treat the symptoms of exercise-induced
muscle damage (see Cleak and Eston, 1992a, for a
review). One such treatment modality, common to the
treatment of acute sports injury, involves cryotherapy
(ice or cold therapy). However, the eVectiveness of
this treatment is unclear due to limited research and
the variations in treatment modality, frequency of
application and duration of treatment.
The response of soft tissue to various forms of
cryotherapy and temperatures of application in both
humans and animals has been reviewed comprehensively by Meeusen and Lievens (1986). A major cause
of inXammation after acute muscle injury is an increase
in capillary permeability, which in conjunction with
an increase in extracellular protein concentration and
vasodilation, results in oedema (Schoberth, 1980).
This response varies directly with tissue temperature
(Janssen and Waaler, 1967). A decrease in tissue temperature results in a reduction in nerve conduction
velocity and activity of the muscle spindle. This
decreases the stretch reXex response and spasticity of
muscle, which reduces the pain spasm cycle and contributes to the relief of pain (Meussen and Lievens,
1986). Thus, when cryotherapy is applied at an appropriate frequency, duration and temperature to injured
muscle, it reduces the inXammatory response and
alleviates spasm and pain after muscle injury (Meeusen
and Lievens, 1986).
In the Wrst published study to investigate the eVects of
heat and cold therapy, in conjunction with stretching
exercise, on the symptoms of exercise-induced muscle
damage, Prentice (1982) observed that a combination
of cold treatment and stretching was most eVective.
However, more recent research has observed that
cryotherapy has a negligible eVect. Yackzan et al. (1984)
applied ice therapy in three separate groups (immediate,
after 24 h or after 48 h) after strenuous eccentric
exercise of the elbow Xexors. They reported no differences in the joint range of motion and concluded
that cryotherapy was not eVective for reducing the
symptoms of exercise-induced muscle damage. However, it is possible that the single cold application treatment was insuYcient to inXuence the mechanisms
responsible for reducing the joint angle in their study. In
a similar study, Isabell et al. (1992) assessed the eVects
of ice-massage alone, ice-massage with exercise and
exercise alone on range of motion of the elbow Xexors
after damaging exercise in three separate groups. No
diVerences in range of motion were observed between
the three groups. Braun and Clarkson (1989) also
observed that cold treatment applied before and during
damaging exercise of the elbow Xexors did not aVect
the relaxed arm angle. It is possible that the lack of
agreement between studies may be attributable to the

Eston and Peters

nature of the cryotherapy procedure (ice application or
ice-massage) and the timing of the applications.
Another application technique, mostly used in
research settings because of the ability to control the
temperature, is immersion in cold water. However, as far
as we are aware, this form of cryotherapy has not been
used in human studies to assess the eVects on muscle
damage resulting from exercise. The aim of this study
was to assess the eVects of cold water immersion on the
signs and symptoms of exercise-induced muscle damage
after a strenuous bout of eccentric exercise.

Methods
Fifteen female student volunteers (mean age 22.0
2.0 years) at the University of Wales, Bangor, provided
informed consent to participate in the study. After
a bout of eccentric exercise on the elbow Xexors of
the dominant arm, the participants were allocated at
random to either a cryotherapy treatment group or a
control group. The symptoms of exercise-induced
muscle damage were then compared to baseline values
over a period of 3 days.
The bout of damage-inducing exercise
The bout of eccentric exercise consisted of eight sets
of Wve maximal contractions (eccentric and concentric)
at 0.58 rad s- 1 with 60 s rest between each set on
an isokinetic dynamometer (KincomTM, Chattanooga,
Bicester, UK). This was similar to other muscle
damage-inducing protocols (Nosaka and Clarkson,
1995; Gulick et al., 1996).
Cryotherapy
The cryotherapy group submerged their exercised arm
(ensuring that the origins and insertions of the biceps
were fully submerged) into a plastic tub of water for
15 min (Meeusen and Lievens, 1986). The water
was maintained at the recommended temperature of
15 1C (by adding ice cubes, cold or hot water,
previously practised in a pilot study) to elicit a drop in
intramuscular temperature of about 7 10C (Meeusen
and Lievens, 1986). This treatment was applied
immediately post-exercise and every 12 h thereafter for
a duration of 3 days in accordance with the suggested
clinical applications of cryotherapy (Kellet, 1986).
Measurements of exercise-induced muscle damage
The criterion measures were plasma creatine kinase
activity, isometric strength of the elbow Xexors, relaxed
arm angle, local muscle tenderness and upper arm

233

Cryotherapy and exercise-induced muscle damage

circumference. These measurements were recorded
immediately before the damage-inducing exercise and
then every 24 h thereafter for a duration of 3 days.
For the treatment group, the measurements were performed immediately before the cryotherapy session to
avoid the initial anaesthetic eVects induced by cold
application.
Creatine kinase activity. Plasma creatine kinase activity
was determined from a Wngertip blood sample. A warm
Wngertip was cleaned with a sterile alcohol swab and
allowed to dry. Capillary puncture was made with
an Autoclix lancette and a sample of whole blood
(32 ml) was pipetted from a capillary tube onto the test
strip and analysed for creatine kinase activity using a
colorometric assay procedure (ReXotron, Boehringer
Mannheim, Lewes, UK). This system uses a plasma
separation principle which is incorporated in the
reagent carrier on the test strip. Samples with creatine
kinase activity in excess of 2000 IU l- 1 were diluted
with equal parts of distilled water according to the
manufacturers instructions.
Isometric strength. After a brief warm-up and practice
of two to three maximal repetitions, the subject performed two maximal isometric contractions of the
elbow Xexors in the standing position with the elbow
Xexed at an angle of 2.36 rad. All subjects were given
standardized instructions to exert as great a force as
possible against the lever arm of the dynamometer. The
force was displayed on the computer monitor in real
time for feedback and motivation purposes. The average
of these two scores was recorded.
Relaxed arm angle. The subjects stood with their hands
and arms down by their sides in the anatomical position.
The exercised arm was fully Xexed and then actively
extended into a relaxed resting position. The angle was
recorded using a transparent goniometer (BaselineTM,
Physiomed, Cheshire, UK). This is an indirect method
of measuring muscle stiVness and soft tissue shortening
that has been used previously (Cleak and Eston, 1992b;
Nosaka and Clarkson, 1995, 1997).
Muscle tenderness. Muscle tenderness was measured
with the subject in the sitting position and with the
exercised arm in a relaxed position with the upper
arm exposed. A computer-mediated algometer was
used (Biokinetics, Bangor, UK), which is described
elsewhere (Edwards et al., 1996). A cylindrical metal
probe with a Xat head diameter of 10 mm was attached
to the load cell of the algometer and was pushed into the
same mid-belly site of the biceps. Accurate placement of
the probe was assured by a pen mark at the mid-belly
site. The subject was asked to press the hand-held

response button when they felt the sensation change

from discomfort to pain (Edwards et al., 1996). The
reading from the algometer was then deducted from
the ceiling value of 40 N and this was taken to be the
tenderness value. This procedure was Wrst used by
Newham et al. (1983) and has been used successfully in
similar studies (Cleak and Eston, 1992b; Eston et al.,
1996). The mean score of three trials was recorded as
the subject s tenderness score.
Swelling: Upper arm circumference. An anthropometric
tape measure was used to determine the upper arm
circumference at the mid-belly of the biceps at the level
of the pen mark. The mean of three measurements was
recorded.
Treatment of data
Comparisons between the treatment and control group
were analysed by a mixed-model, two-factor (group
vs time) analysis of variance, using the SPSS (6.1)
statistical software package. Creatine kinase values were
converted to natural log values to satisfy the assumptions of homogeneity of variance necessary for the
analysis. The assumptions of homogeneity of covariance
were tested by the Mauchly sphericity test. Where this
was signiWcant, as in the case of the creatine kinase data,
the Greenhouse-Geisser epsilon was used to adjust
the degrees of freedom to increase the critical value
of the F-ratio. Tukey post-hoc tests were applied to
determine between-mean diVerences if the analysis of
variance revealed a signiWcant main eVect for time or
interaction of group time.

Results
The results for both the treatment group and the control
group are shown in Table 1.
Relaxed arm angle
The control group had a signiWcantly lower relaxed
arm angle than the cryotherapy group (mean s:
2.56 0.21 vs 2.78 0.08 rad; F1,14 = 5.99, P < 0.05).
There was also a main eVect for time (F3,39 = 12.98,
P < 0.001). The relaxed arm angle decreased from
2.82 0.09 rad at baseline to 2.67 0.16, 2.63 0.14
and 2.60 0.15 rad on days 1, 2 and 3, respectively.
There was also a group time interaction (F3,39 = 4.46,
P < 0.01) (Fig. 1). There was a larger decrease in
relaxed arm angle for the control group, which was
still 13% lower than baseline and 14% lower than the
cryotherapy group on day 3.

234

Eston and Peters

Table 1

Criterion measures across both groups (mean s)

Variable

Baseline

24 h

48 h

4.57 0.53
4.65 0.34

4.97 0.65
4.78 0.43

4.81 0.61
5.50 1.36

4.72 0.55
5.99 1.69

Isometric strength
Cryotherapy
101.6 34.2
Control
100.4 32.8

78.3 21.4
73.0 29.0

92.5 31.2
81.3 42.1

112.3 27.2
86.4 45.3

Relaxed arm angle (rad)

Cryotherapy
2.86 0.11
Control
2.75 0.06

2.75 0.08
2.59 0.25

2.74 0.05
2.49 0.25

2.78 0.06
2.40 0.27

Upper arm circumference (cm)

Cryotherapy
26.7 2.3
Control
29.8 2.2

27.3 2.2
30.2 2.4

27.3 2.1
30.4 2.4

27.1 2.2
30.4 2.7

Tenderness (N)
Cryotherapy
Control

22.6 7.3
26.9 3.9

24.9 6.4
26.7 5.6

22.1 5.4
21.9 6.9

Ln CK (IU l2 1)
Cryotherapy
Control

15.1 9.4
15.7 6.1

72 h

Abbreviation: CK = creatine kinase activity.

Figure 1 Relaxed arm angle of the cryotherapy group (s ) and the control group (h ) immediately before and 3 days after
strenuous eccentric exercise of the elbow Xexors (mean sx). The values for the control group were 13% lower than baseline
values and 14% lower than for the cryotherapy group by day 3 (P < 0.05).

Swelling: Upper arm circumference

Creatine kinase activity

There was a signiWcant main eVect for time (F3,39 =

11.16, P < 0.01). Post-hoc tests revealed that upper arm
circumference increased from 28.2 2.7 cm at baseline
to 28.7 2.7 cm and 28.6 2.9 cm on days 2 and 3,
respectively. However, there was no signiWcant interaction of group time.

There was a signiWcant main eVect for time (F1.3, 16.5 =

4.51, P < 0.05). Tukey post-hoc tests indicated that, by
day 3, the creatine kinase log values were signiWcantly
higher than pre-test values (5.31 1.34 vs 4.60 0.44
IU l- 1). There was also a signiWcant group time interaction (F1.3, 16.5 = 4.47, P < 0.05) (Fig. 2). Tukey tests

235

Cryotherapy and exercise-induced muscle damage

Figure 2 Natural log values for plasma creatine kinase activity of the cryotherapy group (s ) and the control group
(h ) immediately before and 3 days after strenuous eccentric exercise of the elbow Xexors (mean sx). The values for the control
group were higher than baseline values and the values for the cryotherapy group by day 3 (P < 0.05).

revealed that although there was no diVerence in

creatine kinase activity on days 0 and 1 between the two
groups, by days 2 and 3 the values in the control group
were signiWcantly higher, with the highest value for
this group being recorded on day 3. Creatine kinase
activity in the treatment group did not change signiWcantly across time.

Strength
The isometric strength values for the subjects in
this study are similar to those of a previous study
that used similar subjects and methods (Nosaka and
Clarkson, 1997). There was a signiWcant main eVect
for time on strength (F3,39 = 2.87, P < 0.05). Strength
was lower than the baseline value at day 1 (76.1 29.6
vs 101 32.2 N), returning to baseline values by
days 2 and 3. There was no signiWcant group time
interaction.

Muscle tenderness
There was a signiWcant main eVect for time on muscle
tenderness (F3,39 = 16.55, P < 0.001). Tenderness values
were signiWcantly higher than baseline values on day
1 (24.3 5.7 vs 15.4 7.8 N) and remained above
baseline values on day 2 (25.4 6.0 N) and day 3
(22.0 6.1 N). There was no signiWcant interaction of
group time on tenderness.

Discussion
The results of this study are in line with those of previous studies that examined the temporal characteristics
of exercise-induced muscle damage on the upper arm
musculature (Newham et al., 1983; Yackzan et al.,
1984; Newham and Jones, 1985; Friden et al., 1988;
Nosaka et al., 1991; Cleak and Eston, 1992b; Howell
et al., 1993; Nosaka and Clarkson, 1997). However, the
most interesting Wndings are revealed by the group
time interactions on two of the dependent variables
namely, relaxed arm angle and creatine kinase activity.
The overall decrease in relaxed arm angle in the control group by day 3 (0.35 rad) was very similar (0.38
rad) to that observed by Cleak and Eston (1992b), but
greater than (0.21 rad) that observed by Nosaka and
Clarkson (1997), who used a faster eccentric speed to
induce muscle damage (1.05 vs 0.58 rad s- 1). Several
investigators have speculated that the cause of muscle
shortening may be an abnormal increase of calcium ions
in the muscle cell (Armstrong, 1984; Newham et al.,
1985; Clarkson and Tremblay, 1988; Ebbeling and
Clarkson, 1989; Nosaka et al., 1991), which has been
attributed to a defect in the sarcoplasmic reticulum
after damaging exercise. Increased intracellular levels of
calcium have been observed in the muscles of horses
subjected to high-intensity exercise (Byrd, 1992).
The results provide support that cryotherapy may
reduce the extent to which the muscle and connective
tissue unit becomes shortened after strenuous eccentric
exercise. This is indicated by the signiWcant group

236
time interaction for the relaxed arm angle measurement.
It can be seen from Fig. 1 that the cryotherapy group
experienced less of a decrease in relaxed arm angle
(increased range of motion) than the control group.
Studies on humans and animals have shown that
local cooling aV ects the motor neuron pool, the muscle
spindle and its aVerents, the skin aVerents, the myoneural junction and the extrafusal Wbres of the muscle
(see Meeusen and Lievens, 1986, for a review). The
progressive fall in conduction velocity of the motor and
sensory nerves exposed to cold application is in proportion to the temperature of the tissues, and continues
until the nervous tissue temperature decreases to 10 15C (Fox, 1961). The rate of Wring from muscle spindle aVerents decreases when the whole muscle is cooled.
This eVect is immediate and results in a reduction in
spasticity.
Our results lend some support to a previous study
(Prentice, 1982), which examined the use of heat and
cold therapy in conjunction with stretching to determine what combination would elicit the greatest
amounts of relaxation within muscles that were in a
state of exercise-induced damage. Electromyographic
activity indicated that the application of cold followed
by stretching was the most successful combination.
It was proposed that cold application reduced the level
of electrical activity of the muscle spindle by increasing
the threshold stimulus for Wring and decreasing the
aVerent Wring rate. However, the results are not in
line with those of previous studies (Yackzan et al.,
1984; Braun and Clarkson, 1989; Isabell et al., 1992)
that examined the eVects of cold therapy on muscle
spasticity following exercise-induced muscle damage.
This may be a result of the variation in timing and
nature of cold applications.
These factors may also explain why our results diVer
from those of Isabell et al. (1992), who observed that
cryotherapy, applied after damaging eccentric exercise,
had no eVect on plasma creatine kinase activity.
Although the overall mean values (the main eVect for
group) indicated that creatine kinase activity increased
above baseline values on days 2 and 3, this eVect appears
to be a phenomenon of the group time interaction.
It can be seen from Fig. 2 that creatine kinase activity
continued to increase in the control group, whereas
in the cryotherapy group there was no increase above
baseline values.
It is unclear what mechanism can account for the
diVerence in creatine kinase activity between the two
groups after the bout of damaging exercise. Several lines
of explanation might be proposed: there was a reduced
eZux of creatine kinase from the muscle into the
lymphatic system, a reduction in the amount of postexercise damage as a result of cold application in
the treatment group, or an increase in creatine kinase

Eston and Peters

clearance rates in the treatment group. We believe the
last of these explanations not to be plausible, as this
would depend on an enhanced blood Xow that might
increase eZux of creatine kinase from the muscle. The
cold water immersion technique used here was designed
to constrict the capillaries, reduce capillary permeability
and reduce blood Xow to the damaged muscle.
It is possible that the diVerence in creatine kinase
response between the two groups may be associated
with the eVect of cryotherapy on the permeability of
the blood and lymph vessels. As indicated earlier, one
of the major aspects of the inXammatory response
resulting from muscle injury is the increase in the permeability of the vessel wall, which, in combination with
an augmented extravascular and extracellular protein
concentration, results in oedema formation (Schoberth,
1980). As creatine kinase diVuses into the lymph vessel,
it is possible that a reduced permeability of the vessel
walls induced by the cold treatment reduced the rate of
creatine kinase eZux from the muscle. It is also possible
that the cold treatment reduced lymph Xow, as this is
directly aVected by factors which aVect the permeability
of the capillary (Byrne and Levy, 1993). It has been
shown experimentally that cooling the intestinal tissue
of dogs in vivo to 4.5C reduces lymph Xow to about
33% of baseline levels (Ackerman, 1975).
A further explanation may be that the rate of postexercise damage to the muscle tissue was reduced by
the cold treatment, which may have resulted in less
creatine kinase eZux in the cryotherapy group. It is
worth noting that the intramuscular cooling eVect is
maintained for several hours after removal of the surface
cooling (Meeusen and Lievens, 1986). However, if
this was the case, then we might expect the statistical
analysis to reveal a signiWcant group time interaction
on strength. Thus, it is somewhat diYcult to substantiate this hypothesis, as both groups experienced a similar
amount of strength loss after 24 h. However, it is interesting to note that the return of strength to baseline
levels was faster in the cryotherapy group (111% vs 86%
of baseline values by day 3), although this diVerence was
not statistically signiWcant.
As previously indicated, there were signiWcant increments in the amount of swelling experienced by both
groups throughout the 72 h of measurements. This is
attributed to an inXammatory response that involves
movement of Xuid, plasma proteins and leukocytes into
the damaged tissues (Smith, 1991). The time-course of
swelling was similar for both groups.
As both relaxed arm angle and upper arm circumference demonstrated a main eVect for time in this study,
we considered it appropriate to assess the relationship between the relaxed arm angle and the amount of
swelling, particularly as swelling may be a contributing
factor to stiVness of the muscle (Clarkson and Newham,

237

Cryotherapy and exercise-induced muscle damage

1994). The relationship was analysed across the four
separate days for both groups (i.e. 60 paired comparisons). Pearson product moment correlation analysis
revealed a highly signiWcant r-value of - 0.54 (P <
0.001). Thus, in this study, swelling of the upper arm
was inversely associated with the relaxed arm angle.
One of the consequences of increased water within
the extracellular matrix (e.g. connective tissue) is
increased pressure on pain receptors in muscle tissue
(Stauber, 1989). As the time-course of swelling was
similar in both groups, this may account for the similar
increases in tenderness measurements across time. Peak
tenderness occurred at around 48 h post-exercise,
which is in line with previous studies (Ebbeling and
Clarkson, 1989; Cleak and Eston, 1992b; Eston et al.,
1996). Nevertheless, the relaxed elbow angle and
creatine kinase activity values provide evidence of less
damage in the cryotherapy group, which might logically
be expected to be associated with reduced levels of
tenderness in this group. It is possible that the apparent
similarity in tenderness measurements between the two
groups may, in part, be attributable to the method of
assessing tenderness in this study at the mid-belly site
only. Muscle tenderness is more pronounced at the
musculotendinous junction after damaging eccentric
exercise (Cleak and Eston, 1992b; Clarkson and
Newham, 1994; Eston et al., 1994; Baker et al., 1997).
Selection of the method was based on observations
in our laboratory (Baker et al., 1997). Here, tenderness
was measured at nine sites on the quadriceps femoris
before and after eccentric exercise. Tenderness at sites
on the mid-belly of the muscle were less variable than at
the proximal and distal musculotendinous junctions.
Although our Wndings generally contradict previous
observations (Yackzan et al., 1984; Isabell et al., 1992;
Gulick et al., 1996), it is possible that the repeated
treatments applied in this study, which followed
clinically prescribed procedures, might provide a valuable insight into the eVects that cryotherapy may have
on some of the functional consequences of exerciseinduced muscle damage. This disagreement may be
due to the variation in application methods, application
time and temperature of the methods used. The form
of cryotherapy used in this study was essentially different from that in previous studies, which used direct
application of ice, an ice pack or ice-massage on the
skin. Immersion in cold water allows for the possibility
of greater temperature control. Also, it may not be quite
as uncomfortable for the subject.
On a cautionary note, research has shown that limb
blood Xow is decreased by immersion in water baths of
temperatures between 42 and 14C, but that it increases
at temperatures below 10C (see Meeusen and Lievens,
1986, for a review). A decrease in tissue temperature
below 25C for lengthy periods dilates the muscle vessels.

Excessive cryotherapy therefore has an adverse eVect

owing to the increased haemorrhage and inXammatory
response (Barcroft and Edholm, 1943; Kellet, 1986;
Meeusen and Lievens, 1986). In a recent study, histological evidence from the muscle tissue of rats indicated
that cryotherapy treatment exacerbated damage to the
myoWbrils (as measured by electron microscopy) after
intense exercise (Fu et al., 1997). There is evidence,
therefore, to suggest that whatever form cryotherapy
takes, recommended guidelines must be followed.
In conclusion, although the Wndings from this study
lend some support to the repeated application of cryotherapy after strenuous eccentric exercise, further research using cryotherapy procedures is recommended.
It appears that this form of cryotherapy decreases the
degree of shortening of the muscle or the connective
tissue after eccentric exercise. Evidence is also presented
to indicate a possible reduction in damage to the muscle
tissue. This may reduce the possibility of re-rupture and
may allow earlier mobilization of the muscle, which is
an important aspect in the healing of muscle injuries
(Hurme et al., 1993). However, despite the apparent
eVects of cryotherapy on exercise-induced muscle
damage, it does not appear to alter the normal temporal
pattern of tenderness, swelling and strength loss.

Acknowledgements
We thank Ron Maughan, Priscilla Clarkson, Mike Gleeson
and the anonymous referees for their helpful comments. We
also thank Declan Connolly for his advice in the early stages of
the project.

References
Ackerman, N.B. (1975). The inXuences of changes in temperature on intestinal lymph Xow and relationship to
operations for carcinoma of the intestine. Surgery,
Gynecology and Obstetrics, 140, 885 888.
Armstrong, R.B. (1984). Mechanisms of exercise-induced
delayed onset muscle soreness: A brief review. Medicine and
Science in Sports and Exercise, 16, 529 538.
Armstrong, R.B. (1990). Initial events in exercise induced
muscular injury. Medicine and Science in Sports and Exercise,
22, 429 435.
Baker, S.B., Kelly, N. and Eston, R.G. (1997). Pressure pain
tolerance at diVerent sites on the quadriceps femoris
muscle prior to and following eccentric exercise. European
Journal of Pain, 1, 229 233.
Barcroft, H. and Edholm, K. (1943). The eVects of temperature on blood Xow and deep temperature on the human
forearm. Journal of Physiology, 102, 5 20.
Braun, B. and Clarkson, P.M. (1989). EVect of cold treatment
during eccentric exercise. Medicine and Science in Sports and
Exercise, 21 (suppl.), S32.

238
Byrd, S.K. (1992). Alterations in the sarcoplasmic reticulum:
A possible link to exercise-induced muscle damage.
Medicine and Science in Sports and Exercise, 24, 531 536.
Byrne, R.M. and Levy, M.N. (1993). Physiology. St. Louis,
MO: Mosby Year Book.
Clarkson, P.M. and Newham, D.J. (1994). Associations
between muscle soreness, damage and fatigue. In Fatigue:
Neural and Muscular Mechanisms (edited by S.C. Gandevia,
R.M. Enoka, A.J. McComas, D.G. Stuart and C.K.
Thomas), pp. 457 469. New York: Plenum Press.
Clarkson, P.M. and Tremblay, I. (1988). Exercise induced
muscle damage, repair and adaptation in humans. Journal
of Applied Physiology, 65, 1 6.
Cleak, M.J. and Eston, R.G. (1992a). Delayed onset muscle
soreness: Mechanisms and management. Journal of Sports
Sciences, 10, 325 341.
Cleak, M.J. and Eston, R.G. (1992b). Muscle soreness,
swelling, stiV ness and strength loss after intense eccentric
exercise. British Journal of Sports Medicine, 26, 267 272.
Ebbeling, C.B. and Clarkson, P.M. (1989). Exercise induced
muscle damage and adaptation. Sports Medicine, 7, 207 234.
Edwards, T., Baker, S. and Eston, R.G. (1996). A method of
detecting the muscle pain threshold using an objective
software-mediated technique. Perceptual and Motor Skills,
82, 955 960.
Enoka, R.M. (1996). Eccentric contractions require unique
activation strategies by the nervous system. Journal of
Applied Physiology, 81, 2339 2346.
Eston, R.G., Critchley, N. and Baltzopoulos, V. (1994). Delayed onset muscle soreness, strength loss characteristics
and creatine kinase activity following uphill and downhill
running. Journal of Sports Sciences, 12, 135.
Eston, R.G., Finney, S., Baker, S. and Baltzopoulos, V. (1996).
Muscle tenderness and peak torque changes after downhill running following a prior bout of isokinetic eccentric
exercise. Journal of Sports Sciences, 14, 291 299.
Fox, R.H. (1961). Local cooling in man. B ritish Medical
Bulletin, 17, 14 18.
Friden, J. and Leiber, R.L. (1992). Structural and mechanical
basis of exercise-induced muscle injury. Medicine and
Science in Sports and Exercise, 24, 521 530.
Friden, J., Seger, J. and Ekblom, B. (1988). Sublethal muscle
Wbre injuries after high tension anaerobic exercise.
European Jour nal of Applied Physiology, 57, 360 368.
Fu, F.H., Cen, H.W. and Eston, R.G. (1997). The eVects
of cryotherapy on muscle damage in rats subjected to
endurance training. Scandinavian Journal of Medicine and
Science in Sports, 7, 358 362.
Gulick, D.T., Kimura, I.F., Sitler, M., Paolone, A. and Kelly,
J.D. (1996). Various treatment techniques on signs and
symptoms of muscle soreness. Journal of Athletic Training,
31, 145 152.
Hill, D.W. and Richardson, J.D. (1989). EVectiveness of 10%
trolamine salicylate cream on muscular soreness induced
by a reproducible program of weight training. Journal of
Orthopaedic and Sports Physical Therapy, 11, 19 23.
Howell, J.N., Chlebourn, G. and Conatser, R. (1993). Muscle
stiV ness, strength loss, swelling and soreness following

Eston and Peters

exercise-induced injury in humans. Journal of Physiology
(Lond.), 464, 183 196.
Hurme, T., Ratanen, J. and Kalimo, H. (1993). EV ects of
early cryotherapy in experimental skeletal muscle injury.
Scandinavian Journal of Medicine and Science in Sports, 3,
46 51.
Isabell, W.K., Durrant, E., Myrer, W. and Anderson, S.
(1992). The eVects of ice massage, ice massage with
exercise, and exercise on the prevention and treatment of
delayed onset muscle soreness. Journal of Athletic Training,
27, 208 217.
Janssen, C.W. and Waaler, E. (1967). Body temperature,
antibody formation and inXammatory response. Acta
Pathologica, Microbiologica et Immunologica Scandinavica,
69, 557 566.
Kellet, J. (1986). Acute soft tissue injuries: A review of the
literature. Medicine and Science in Sports and Exercise, 18,
489 500.
Meeusen, R. and Lievens, I. (1986). The use of cryotherapy
in sports injuries. Sports Medicine, 3, 398 414.
Newham, D.J. and Jones, D.A. (1985). Intramuscular pressure
in the painful human biceps. Clinical Science, 69 (suppl.), 27.
Newham, D.J., Mills, K.R., Quigley, B.M. and Edwards,
R.H.T. (1983). Pain and fatigue after concentric and
eccentric contractions. Clinical Science, 64, 55 62.
Newham, D.J., Jones, D.A., Tolfrey, S.E. and Edwards, R.H.T.
(1985). Skeletal muscle 99m technetium pyrophosphate
uptake, enzyme release and muscle pain following stepping
exercise. European Journal of Applied Physiology, 55, 106 122.
Nosaka, K. and Clarkson, P.M. (1995). Muscle damage
following repeated bouts of high force eccentric exercise.
Medicine and Science in Sports and Exercise, 27, 1263 1269.
Nosaka, K. and Clarkson, P.M. (1997). InXuence of previous
concentric exercise on eccentric exercise-induced muscle
damage. Journal of Sports Sciences, 15, 477 483.
Nosaka, K., Clarkson, P.M., McGuiggin, M.E. and Byrne,
J.M. (1991). Time course of muscle adaptation after
high force eccentric exercise. European Journal of Applied
Physiology, 63, 70 76.
Prentice, W.E. (1982). An electromyographic analysis of the
eV ectiveness of heat or cold and stretching for inducing
relaxation in injured muscle. Journal of Orthopaedic and
Sports Physical Therapy, 3, 133 140.
Schoberth, H. (1980). Eisbehandlung bei Sportverletzungen.
Physiotherapie, 71, 814 818.
Smith, L. (1991). Acute inXammation the underlying
mechanism in delayed onset of muscle soreness? Medicine
and Science in Sports and Exercise, 23, 542 551.
Stauber, W.T. (1989). Eccentric action of muscles: Physiology,
injury and adaptation. In Exercise and Sport Sciences
Reviews, Vol. 17 (edited by K.B. Kandolph), pp. 157 185.
Baltimore, MD: Williams & Wilkins.
Talag, T. (1973). Residual muscle soreness as inXuenced by
concentric, eccentric and static contractions. Research
Quarterly, 44, 458 469.
Yackzan, L., Adams, C. and Francis, K.T. (1984). The eVects
of ice massage on DOMS. American Journal of Sports
Medicine, 12, 159 165.