Content uploaded by Shona Halson
Author content
All content in this area was uploaded by Shona Halson on Aug 28, 2014
Content may be subject to copyright.
130
International Journal of Sports Physiology and Performance, 2012, 7, 130-140
© 2012 Human Kinetics, Inc.
Effect of Contrast Water Therapy Duration
on Recovery of Running Performance
Nathan G. Versey, Shona L. Halson, and Brian T. Dawson
Purpose: To investigate whether contrast water therapy (CWT) assists acute recovery from high-intensity
running and whether a dose-response relationship exists. Methods: Ten trained male runners completed 4
trials, each commencing with a 3000-m time trial, followed by 8 × 400-m intervals with 1 min of recovery.
Ten minutes postexercise, participants performed 1 of 4 recovery protocols: CWT, by alternating 1 min hot
(38°C) and 1 min cold (15°C) for 6 (CWT6), 12 (CWT12), or 18 min (CWT18), or a seated rest control trial.
The 3000-m time trial was repeated 2 h later. Results: 3000-m performance slowed from 632 ± 4 to 647 ±
4 s in control, 631 ± 4 to 642 ± 4 s in CWT6, 633 ± 4 to 648 ± 4 s in CWT12, and 631 ± 4 to 647 ± 4 s in
CWT18. Following CWT6, performance (smallest worthwhile change of 0.3%) was substantially faster than
control (87% probability, 0.8 ± 0.8% mean ± 90% condence limit), however, there was no effect for CWT12
(34%, 0.0 ± 1.0%) or CWT18 (34%, –0.1 ± 0.8%). There were no substantial differences between conditions
in exercise heart rates, or postexercise calf and thigh girths. Algometer thigh pain threshold during CWT12
was higher at all time points compared with control. Subjective measures of thermal sensation and muscle
soreness were lower in all CWT conditions at some post-water-immersion time points compared with control;
however, there were no consistent differences in whole body fatigue following CWT. Conclusions: Contrast
water therapy for 6 min assisted acute recovery from high-intensity running; however, CWT duration did not
have a dose-response effect on recovery of running performance.
Keywords: exercise, hydrotherapy, fatigue, dose response, time trial
Versey and Halson are with Performance Recovery, Australian
Institute of Sport. Dawson is with the School of Sport Science
Exercise and Health, University of Western Australia.
Elite middle and long-distance runners may regularly
become fatigued, as they typically train one or more times
per day, for 6–7 d/wk and may also compete on consecu-
tive days, often at very high intensities. In addition, many
individual (triathlon) and team sports (football, hockey)
involve large amounts of running, often at high intensi-
ties,1–3 also resulting in fatigue. To maximize exercise
performance in the short term (the subsequent training
session, or competition heat/round), athletes must recover
as fast as possible.
Contrast water therapy (CWT) is increasingly
being used by runners and other athletes to accelerate
postexercise recovery, with athletes alternating several
times between hot and cold water immersion (1–2 min in
each). Conicting literature exists on the ability of CWT
to assist acute postexercise recovery. A number of stud-
ies have reported benecial effects on performance;5–10
however, others have found no change.4,11–15 While it has
been suggested that the lack of an effect on performance
could be partly due to the methodology used,10 it may
alternatively be due to a lack of change in intramuscular
temperature.16,17 In addition, it has been suggested that
accelerating acute recovery by disrupting the mechanisms
of fatigue may blunt adaptations to training, potentially
inhibiting long-term improvements in tness.18–20 The
effect of CWT on long-term adaptations to training war-
rants further investigation; however, this is beyond the
scope of the present study. The range of CWT protocols,
and potential mechanisms by which they might affect
recovery from exercise have been discussed in recent
review articles.21,22
To our knowledge, only Coffey et al11 have inves-
tigated whether CWT could accelerate recovery from
high-intensity running (time to exhaustion at 120 and
90% of peak running speed). They found that 15 min of
CWT, active recovery or passive standing recovery had
no effect on subsequent treadmill running performance
4 h later. It was suggested that the 4 h between exercise
bouts was sufcient time to allow athletes to recover
fully, overriding any acute effects of the CWT or active
recovery interventions. To our knowledge, no other papers
have examined the acute effect of CWT on subsequent
high-intensity running performance.
Recovery from team sport exercise demands after
using CWT has been assessed by measuring running
repeat sprint ability. Hamlin et al,13,14 Ingram et al,4
and King et al23 all reported that CWT had no effect on
Contrast Water Therapy Recovery, Running 131
recovery of performance compared with active recovery
or control conditions, possibly because the body was
immersed only to waist level in cold water, and hot show-
ers were used instead of water immersion.
To our knowledge, previous work from our labora-
tory10 is the only attempt to determine the ideal duration
of CWT. We compared the effect of performing postex-
ercise CWT (alternating 1 min hot, and 1 min cold) for
6, 12, or 18 min on high-intensity cycling performance
90 min later. It was found that 6 min of CWT improved
time-trial and sprint performance, whereas 12 min only
improved sprint performance. Therefore, a dose-response
relationship was not found between CWT duration and
subsequent high-intensity cycling performance.
The present study was designed to replicate our pre-
vious investigation,10 using high-intensity running instead
of cycling. Running involves eccentric muscle contrac-
tions, which can cause signicant muscle damage.24 In
contrast, less muscle damage is generally evident after
predominantly concentric exercise such as cycling,25
thus allowing any effect of CWT on recovery from dif-
ferent modes of exercise to be compared. The primary
purpose of this study was to determine if CWT assists
acute, same-day recovery from high-intensity running
performance. In addition, the study aimed to determine
if a dose-response relationship exists between CWT
duration and subsequent high-intensity running perfor-
mance. It was hypothesized that CWT would assist acute,
same-day postexercise recovery, therefore improving
subsequent high-intensity running performance, and that
performing CWT for a longer duration would increase
recovery benets, as running generally promotes more
muscle damage than cycling.
Methods
Subjects
Ten trained male long-distance runners (mean ± SD,
age: 36.8 ± 9.2 y, height: 1.79 ± 0.08 m, body mass:
70.8 ± 8.3 kg, VO2max: 64.3 ± 5.0 mL · kg–1 · min–1, sum
of seven skinfolds: 49.3 ± 12.0 mm) volunteered for the
study. The participants typically covered >60 km/wk
in training, and were not regular users of hydrotherapy
recovery techniques. Participants were informed of the
risks and provided written informed consent. The study
was approved by the Australian Institute of Sport (AIS)
Research Ethics Committee in the spirit of the Helsinki
Declaration, and conducted according to the protocols
recommended by Harris and Atkinson.26
Experimental Design
This study used a study design, interventions, and out-
come measures similar to that of Versey et al,10 except
participants performed high-intensity running instead of
cycling. The study used a post-only crossover design with
time trial 1 and wind speed as covariates. Participants
completed 2 familiarizations to minimize any learning
or training effects, followed by 4 trials in randomized,
counterbalanced order separated by a minimum of 4 d.
Trials differed only in the recovery protocol and each
participant performed their trials at the same time of day.
Procedures
Before the study, participants performed a VO2max test
on a custom-built motorized treadmill (AIS, Canberra,
Australia) as described in Robertson et al,27 with run-
ning speeds adjusted to suit the participants (submaxi-
mal speed range: 13–17 km/h). A participant’s sum of
7 skinfolds was measured by a certied ISAK level 2
anthropometrist.28
During the 24 h before each trial, participants
repeated the same training consisting of up to 45 min of
low-intensity exercise to ensure they commenced each
trial with a similar (low) level of fatigue. Participants
also completed a food diary to ensure energy and uid
consumptions were similar on each occasion, and did not
consume caffeine or alcohol during the 24 h before each
trial. At 90 min before each trial, participants consumed
standardized food (2 g of carbohydrate per kilogram of
body mass) and uid (850 mL) to limit substrate deple-
tion and dehydration as sources of fatigue.
Each trial (Figure 1) commenced with exercise
bout 1 (Ex 1), whereby participants initially performed
a 15-min self-paced warm-up, including 3 × 100 m
efforts at predicted 3000-m time-trial pace. They then
completed a maximal 3000-m running time trial (time
trial 1) on an International Association of Athletics Fed-
erations standard 400-m athletics track (AIS, Canberra,
Australia) to measure performance. Five minutes later,
participants commenced a set of 8 × 400 m intervals, with
1 min of recovery between efforts, aiming to complete the
intervals in the fastest cumulative time possible. This is a
typical training set performed by middle and long-distance
runners and was designed to increase fatigue levels. A 7-min
self-paced warm-down was then performed in all conditions.
Two hours after completing Ex 1, the warm-up, 3000-m
time trial (time trial 2), and warm-down were repeated
in exercise bout 2 (Ex 2) to determine the affect of the
recovery interventions on performance. The 2 h break
was designed to be short enough to ensure a decrease
in performance in the control trial, while simulating the
twice a day training frequently performed by middle and
long-distance runners. At 90 min before Ex 2, participants
again consumed the standardized food and uid intake.
To avoid pacing, participants did not receive their
splits, nal times, or heart rate data until after each trial.
In addition, participants’ start times were separated by 10
s to decrease the likelihood of pacing off other runners.
The same researcher conducted all trials and provided
strong, consistent verbal encouragement throughout. To
minimize the effect of environmental conditions (tem-
perature: 14.9 ± 4.9°C, humidity: 50 ± 16%, wind speed:
0.4 ± 0.8 m/s) on performance, trials were rescheduled
when hot or windy conditions were predicted. Despite
this, unexpected low wind speeds were present in some
132 Versey, Halson, and Dawson
trials. Trials were performed in winter and spring,
resulting in a relatively large standard deviation in air
temperature and ad libitum water consumption (571 ±
477 mL) during each trial.
Recovery Interventions
Ten minutes after Ex 1, participants performed 1 of 4
recovery protocols: CWT for 6 (CWT6), 12 (CWT12),
or 18 (CWT18) min in duration, or a seated rest control
trial (temperature: 21.7 ± 2.7°C, humidity: 43.0 ± 9.3%).
The CWT alternated between hot (38.4 ± 0.3°C) and
cold water (14.6 ± 0.5°C) (T106 thermometer, Testo
AG, Germany) every minute with a 5-s changeover.
Participants immersed their entire body (seated, exclud-
ing head and neck) in the pools. The CWT protocol was
based on previous research that found it to be effective
at accelerating recovery of exercise performance.6,9,10,20
In addition, it maximized movement between extreme
temperatures and involved equal durations of hot and
cold water immersion. Following CWT, participants sat
at rest (temperature: 22.1 ± 2.4°C, humidity: 41.0 ± 7.4%)
until Ex 2 and did not perform any additional organized
recovery strategies, including stretching, massage, nap-
ping, or compression garments.
Recovery Assessment
Running Performance. Running performances in
the 3000-m time trial 2 were the primary measure of
recovery. Split and nal times were measured using a
stopwatch (S141, Seiko, Japan). Times for each of the
400-m intervals were added to provide a total time. The
typical errors were 1.1 and 1.2% for the time trial and
interval total times, respectively.
Heart Rate. Heart rate was logged every 5 s by a Polar
heart rate monitor (S810i, Polar, Finland). Average heart
rate for each time trial and peak heart rate at the end of
the interval set were determined.
Leg Girths. Calf and mid-thigh girths were measured
on the right leg according to the International Standards
for Anthropometric Assessment28 as an indicator of leg
edema. The positions of the tape (W606PM, Lufkin,
USA) were marked on the leg to ensure the same locations
were measured throughout each trial, therefore improving
reliability. Typical errors for calf and thigh girth were
each 0.1%. Calf and thigh girths were measured pre
Ex 1, immediately, 30, 60, and 90 min after Ex 1, and
immediately after Ex 2.
Algometer Measures. Pressure-to-pain threshold was
measured as an indicator of quadriceps muscle soreness
at the front-thigh skinfold site with the participant supine,
and calf muscle soreness at the height of the medial-calf
skinfold site 3 cm lateral to the midline of the calf with
the participant prone.28 Both measures were made on
the right leg using a calibrated algometer (Somedic,
Sweden) according to the manufacturer’s instructions.
A at, circular probe with a surface area of 2 cm2 and
covered with a 2-mm-thick piece of rubber was pressed
perpendicularly against the skin. A single operator aimed
to maintain a rate of increase in pressure at 30 kPa/s, and
feedback on the rate of increase was provided by the
Figure 1 — Schematic of conditions indicating when exercise bouts 1 (Ex 1) and 2 (Ex 2); contrast water therapy of 6 (CWT6), 12
(CWT12), or 18 (CWT18) min in duration; and seated rest occurred. *Indicates standardized food and uid consumption.
Contrast Water Therapy Recovery, Running 133
digital display on the algometer. However, the operator
was blinded to the absolute pressure value until the
participants had pressed a switch when their sensation
changed from pressure to pain. Measures were made in
duplicate 10 s apart, with typical errors for calf and thigh
pressure-to-pain threshold of 6.5 and 9.7% respectively.
Pressure-to-pain threshold differed between duplicate
measures for the calf and thigh by 1.0 (P = .485) and
–2.3% (P = .272) respectively. Algometer measures were
made before Ex 1; immediately, 30, 60, and 90 min after
Ex 1; and immediately after Ex 2.
Subjective Measures. Rating of perceived exertion
(RPE), effort, and motivation were recorded after each
time trial. The RPE was on a scale of 6 (no exertion) to
20 (maximal exertion);29 effort and motivation were rated
out of 100% (maximal).
Thermal sensation, whole body fatigue, and muscle
soreness were measured before Ex 1; immediately, 30,
45, 60, and 90 min after Ex 1; and immediately before
and after Ex 2. A Likert scale measured thermal sensa-
tion (0 = unbearably cold, 8 = unbearably hot).30 Borg’s29
CR10 scale was adapted to measure whole body fatigue
and muscle soreness31 (0 = nothing at all, 10 = extremely
high). Before assessing muscle soreness, participants
performed a standardized half squat so they experienced
the same movement/sensation on each occasion.9
After completing all trials, participants ranked the
experimental conditions from 1 (most) to 4 (least) accord-
ing to how likely they were to use them to assist with
recovery in a similar competitive situation.
Statistical Analysis
A post-only analysis was conducted because this tech-
nique produced narrower condence limits for effect sizes
than a pre–post analysis. Time trial 1 was incorporated
as a covariate to take into account participants’ baseline
ability for each trial. Wind speed during exercise was also
used as a covariate to adjust for any impact on time-trial
time. Analysis was conducted using a modied statisti-
cal spreadsheet.32 The performance, heart rate, girth, and
algometer data were log transformed before analysis to
reduce nonuniformity of error. Algometer data was the
average of duplicate measures.
Magnitude-based inferences (90% condence limits
[CL]) were used to assess effects between trials33 using
the following qualitative probabilities: <1%: almost
certainly not, <5%: very unlikely, <25%: unlikely/prob-
ably not, 25–75%: possibly/possibly not, >75%: likely/
probably, >95%: very likely, >99% almost certainly. A
substantial effect was set at >75%. Effects were unclear
if the 90% CL spanned both substantially positive and
negative values.
Performance in the 3000-m time trials was designed
to simulate performance in competitive high-intensity
3000-m running races and be similar to other middle and
long-distance events. Hopkins et al34 calculated the small-
est worthwhile change that could inuence performance
in a competitive event as 0.3 times the within-athlete
between-race variability in performance. It is reasonable
to assume the individual variability in race performances
is similar to the individual variability in the time trials
performed in the present study; therefore, we used the
error of measurement to simulate the between-race vari-
ability when calculating the smallest worthwhile change
in performance. To obtain the error of measurement,
changes in time-trial 2 performances between conditions
were calculated and the SD determined, then divided by
root 2.35 To decrease the potential impact of individual
responses on the estimate of variability, the mean of the
smallest worthwhile changes from the following com-
parisons was calculated: Control versus CWT6, CWT6
versus CWT12, and CWT12 versus CWT18.
Results
Running Performance
Smallest worthwhile change in 3000-m time-trial perfor-
mance was calculated as 0.3%. Descriptive statistics for
3000-m time trials and the 400-m interval set are shown
in Table 1. Performance in time trial 1 and the interval set
did not differ between conditions and slowed from time
trial 1 to 2 in all conditions. Time-trial 2 performance
for each condition (compared with control) are shown in
Figure 2, performance following CWT6 was substantially
faster than control (87% probability, 0.8 ± 0.8% mean
± 90% CL); however, CWT12 (34%, 0.0 ± 1.0%) and
CWT18 (34%, –0.1 ± 0.8%) had no effect.
Heart Rate
Descriptive heart rate data are shown in Table 1. No
substantial differences existed between conditions in
time-trial average heart rate during Ex 1 or Ex 2, or in
interval peak heart rate. All effects were trivial or unclear
with an effect size <0.20.
Leg Girths
Calf and thigh girths for each condition (compared with
control) are shown in Figure 3. No substantial differences
between conditions were found at any time point, with
all effects being trivial (effect size <0.20).
Algometer Measures
Calf and thigh pressure to pain thresholds for each
condition (compared with control) are shown in Figure
4 (larger values represent a higher pain threshold), data
for CWT6 and CWT18 are based on nine participants.
Thigh pressure-to-pain threshold was higher in CWT12
compared with control at all time points.
Subjective Measures
The RPE was not substantially different between condi-
tions after time trial 1. After time trial 2, RPE was sub-
stantially lower following CWT6 (75%, –0.41 ± 0.54)
134 Versey, Halson, and Dawson
and CWT18 (88%, –0.56 ± 0.53) compared with control.
Following CWT12, RPE was lower than control, with a
small effect size (66%, –0.36 ± 0.70).
No substantial differences existed between condi-
tions in effort during time trial 1 or 2. However, effort
during time trial 2 was lower during CWT18 compared
with control, with a small effect size (68%, –0.38 ± 0.73).
Motivation during time trial 1 was substantially
lower in CWT6 (75%, -0.48 ± 0.73) compared with
control. Motivation during time trial 2 was substantially
lower in CWT6 (75%, –0.41 ± 0.57) and CWT18 (79%,
–0.65 ± 0.98) compared with control.
Thermal sensation, whole body fatigue, and muscle
soreness for each condition compared with control are
shown in Figure 5. Thermal sensation and muscle sore-
ness were lower in all CWT conditions compared with
control at some time points following water immersion
until time trial 2. No consistent differences existed
between conditions in whole body fatigue following
CWT.
Only 7 participants provided a preference order
for the recovery interventions. Three chose CWT6, 1
chose CWT12 and 3 chose CWT18 as their most pre-
ferred recovery intervention. One chose CWT6, 5 chose
CWT12, and 1 chose CWT18 as their second preference.
The least preferred intervention was control (n = 6) and
CWT18 (n = 1).
Discussion
The present study investigated whether postexercise CWT
assisted subsequent high-intensity running performance
2 h later, and if a dose-response relationship existed. The
main ndings were that recovery of 3000-m running
time-trial performance was substantially faster follow-
ing 6 min of CWT compared with no intervention, and
that 12 and 18 min of CWT did not accelerate recovery
of performance (Figure 2). As a result, CWT duration
did not have a dose-response effect on acute recovery of
high-intensity running performance.
Coffey et al11 investigated whether CWT accelerates
recovery from high-intensity running performance, they
performed two pairs of treadmill time-to-exhaustion tests
at 120 and 90% of peak running speed, with 4 h between
each pair of tests. Following the rst pair of tests partici-
pants performed 15 min of CWT (alternating 60 s in 10°C,
and 120 s in 42°C water), active recovery (running at 40%
of peak running speed) or passive standing recovery. They
found that CWT had no effect on subsequent treadmill
running performance compared with the active and pas-
sive recovery conditions; however, they suggested that
Table 1 Descriptive Statistics (Mean ± Percent Coefficient of Variation) for
Performance and Heart Rate Data in Control, CWT6, CWT12, and CWT18 Conditions
Variable Control CWT6 CWT12 CWT18
Time (s)
3000-m time trial 1 632 ± 4a631 ± 4a633 ± 4a631 ± 4a
400-m intervals (total time) 617 ± 4 615 ± 4 620 ± 4 622 ± 4
3000-m time trial 2 647 ± 4 642 ± 4b648 ± 4 647 ± 4
Heart Rate (bpm)
3000-m time trial 1 (average) 165 ± 5 166 ± 5 166 ± 5 166 ± 4
400-m intervals (peak) 171 ± 4 171 ± 4 171 ± 4 170 ± 4
3000-m time trial 2 (average) 166 ± 5 167 ± 5 166 ± 5 166 ± 5
Abbreviations: bpm = beats per minute.
Contrast water therapy for 6 (CWT6), 12 (CWT12), or 18 (CWT18) min in duration.
a Substantial difference between time trial 1 and 2 within a condition.
b Substantial difference in time trial 2 compared with control
Figure 2 — Percent change in 3000-m time-trial performance
for contrast water therapy of 6 (CWT6), 12 (CWT12), and 18
(CWT18) min duration compared with the control trial. Data
are means ± 90% condence intervals. Shaded boxes indicate
the smallest worthwhile change in performance.
Contrast Water Therapy Recovery, Running 135
4 h was sufcient time to allow participants to recover
fully in all conditions, overriding any acute effects of the
CWT recovery intervention. The present study also used
a strenuous exercise protocol, but a shorter (2 h) break
between exercise bouts. A decrease in 3000-m running
performance (–2.4%) was observed from time trial 1 to
2 in the control condition, and a reduced magnitude of
decline in performance was observed following CWT6.
The effect of CWT on recovery of team sport
demands has been assessed in a number of studies by
measuring running repeat sprint ability.13,14 Hamlin et
al examined the acute affect of performing postexercise
CWT for 6 min (alternating 1 min in 8–10°C water, and
1 min in a 38°C shower) on repeat sprint tests (10 × 40
m sprints on 30 s) separated by 60 min in developmental-
level rugby players,13 and repeat sprint tests (6 × 5, 10 and
15 m shuttles on 30 s) separated by 2 h in developmental-
level netballers.14 In contrast to the ndings of the present
study, these studies13,14 reported that 6 min of CWT did
not affect subsequent repeat sprint test performance. The
contrasting ndings could be due to the different exercise
protocols, as repeat sprint test performance36 requires
different energy system contributions to middle-distance
running,37 or the different CWT protocols performed in
the respective studies. In the present study, participants
performed full body water immersion; however, Hamlin
et al13,14 immersed participants only to hip level in cold
water and used hot showers on the legs. Full body immer-
sion will likely produce a greater physiological response
than half body immersion or showers due to the greater
proportion of the body in contact with water, and the
effects of hydrostatic pressure on the body.10
Figure 3 — Calf (a) and thigh girth (b) for contrast water therapy of 6 (CWT6), 12 (CWT12), and 18 (CWT18) min in duration
compared with control (dotted line). Data are the mean ± 90% CL. Ex 1 = Exercise bout 1, Ex 2 = Exercise bout 2. Shaded boxes
indicate the CWT durations. There were no substantial differences between conditions.
136 Versey, Halson, and Dawson
The present study replicated previous work from
our laboratory;10 however, participants performed high-
intensity running instead of cycling. We previously
reported that CWT6 improved 5-min cycling time-trial
performance by 1.5 ± 2.1% compared with control,
whereas CWT12 and CWT18 had no effect. The decrease
in cycling time-trial average power in the control trial
(1.8%) was comparable to the increase in running time-
trial duration in the present study (2.4%). Cycling sprint
total work was also greater after CWT6 (3.0 ± 3.1%) and
CWT12 (4.3 ± 3.4%) compared with control.10 Compar-
ing these results to the running time trials performed in
the present study suggests that CWT6 is the duration
most likely to assist recovery of performance and that
high-intensity cycling is likely to benet by a greater
magnitude than running (0.8 ± 0.8%) within a 2-h time
frame. We hypothesized that CWT of greater durations
would assist recovery from running more than cycling,
due to the greater muscle damage generally caused by
the eccentric muscle contractions involved in running,24
compared with the predominantly concentric contrac-
tions in cycling.25
The results of this and our previous study10 both
suggest that postexercise CWT18 was too long to assist
exercise performance 2 h later, and therefore a dose-
response relationship does not exist between CWT dura-
tion and subsequent high-intensity running or cycling
performance. We had anticipated that all CWT conditions
would assist recovery of exercise performance and that
a dose-response relationship might exist. In the present
Figure 4 — Calf (a) and thigh pressure-to-pain threshold (b) for contrast water therapy of 6 (CWT6) (n = 9), 12 (CWT12), and 18
(CWT18) (n = 9) min in duration compared with control (dotted line). Data are means ± 90% CL. Ex 1 = Exercise bout 1, Ex 2 =
Exercise bout 2. Shaded boxes indicate the CWT durations. Substantial effects for CWT trials compared with control are indicated
by 12 or 18.
137
Figure 5 — Thermal sensation (a), whole body fatigue (b), and muscle soreness (c) for contrast water therapy of 6 (CWT6), 12
(CWT12), and 18 (CWT18) min in duration compared with control (dotted line). Data are means ± 90% CL. Ex 1 = Exercise bout
1, Ex 2 = Exercise bout 2. Shaded boxes indicate the CWT durations. Substantial effects for CWT trials compared with control are
indicated by 6, 12, or 18.
138 Versey, Halson, and Dawson
study, CWT12 and CW18 did not assist recovery of run-
ning performance, possibly because their cooling effects
were too large for the cold environmental conditions (14.9
± 4.9°C) experienced by participants during exercise,
potentially hindering their ability to warm up in Ex 2.
Participants generally reported lower thermal sensations
following all CWT immersions until Ex 2 compared
with control. Following Ex 2 thermal sensation was still
lower in CWT18 (Figure 5). An elevation in internal body
temperatures is recognized as an important component of
warming up for exercise and the subsequent improvement
in performance.38
Our previous investigation10 also found that core
temperatures before Ex 2 were substantially lower fol-
lowing CWT12 (–0.19 ± 0.14°C) and CWT18 (–0.21 ±
0.10°C) compared with control. Core temperature was
not measured in the present study because it was not
practical or comfortable for rectal temperature to be
recorded while running in the eld. Participants here
probably experienced changes in core temperature
similar to those of our previous investigation10 because
the same CWT protocols were performed; however,
due to the colder outdoor environmental conditions,
this may have occurred at a lower absolute core
temperature. If this study was repeated in warmer envi-
ronmental conditions, CWT12 and CWT18 might also
assist in the recovery of running performance. It should
be noted that Vaile et al6 reported no difference in core
temperature values between control and CWT (alternat-
ing 1 min in 38°C, and 1 min in 15°C water for 14 min)
in the 15 min following CWT, but the reason for these
differing ndings is unclear.
Previous studies reporting that CWT assists postex-
ercise recovery have used immersion durations of 146,9
and 15 min,5 and measured performance for up to 4 d
postexercise. Other studies have performed CWT for
shorter durations, but as discussed previously, 2 used
hot showers instead of pools, and failed to use full body
immersion,13,14 potentially decreasing any benecial
effects of CWT, while the third did not have a control
condition.12 The combined results of the present study
and the previously discussed literature, suggests that
full body CWT immersion durations of 6–15 min are
most appropriate for assisting postexercise recovery of
performance.
Heart rate, leg girths, and pressure-to-pain thresholds
were measured here to help explain changes in exercise
performance. Heart rate did not differ between conditions
at any time point (Table 1), indicating the improved per-
formance in CWT6 was not due to participants sustain-
ing a higher heart rate during exercise. Likewise, there
were no differences between conditions in leg girths
at any time point (Figure 3), suggesting that CWT did
not alter calf or thigh swelling in response to fatiguing
exercise. In contrast, Vaile et al5 reported a decreased
thigh volume immediately and 48 h following eccentric
leg-press exercise and CWT. However, the contrasting
ndings could be because the leg-press protocol was
specifically designed to induce muscle damage and
delayed onset muscle soreness, possibly causing greater
uid shifts and inammation than the running protocol
in the present study.
Calf and thigh pressure-to-pain thresholds were no
different in CWT6 compared with control (Figure 4);
therefore, the faster time-trial performance in CWT6 was
probably not due to an increased pressure-to-pain thresh-
old. In addition, the increased thigh pressure-to-pain
threshold following CWT12 did not appear to improve
subsequent running performance. The mechanisms by
which CWT affects postexercise recovery are unclear and
warrant further investigation to optimize CWT protocols
for athletes.
Three of 7 participants selected CWT6 as the
recovery condition they were most likely to use in a
competitive situation, despite CWT6 resulting in the
greatest performance recovery. The authors believe this
was largely because 3 other participants selected CWT18
based on the theory of “more is better.” The participants
were asked to provide these rankings based on their
experiences during the study; however, their existing
beliefs are likely to have inuenced their responses. The
participants existing beliefs could also explain why 6 of
7 participants would prefer to use CWT of any duration
over no intervention.
The possibility of a placebo effect in the present
study is acknowledged. However, in studies of this type,
it is virtually impossible to blind the participants and
researchers to the water temperatures and immersion
durations. Effort and motivation during the 3000-m time
trials were measured, as they are factors that can inuence
exercise performance. However, here the participants
reported few differences in effort and motivation between
conditions; therefore, any differences in performance are
unlikely to be due to these factors.
Practical Applications
and Conclusions
The ndings of the present study suggest that performing
postexercise CWT for 6 min might accelerate subsequent
3000-m time-trial running performance 2 h later; how-
ever, a dose-response relationship does not exist between
CWT duration and recovery of running performance.
Performing CWT for 12 and 18 min did not accelerate
recovery when exercising outdoors in cold environmental
conditions. The effect of performing CWT on subsequent
high-intensity running performance in warmer envi-
ronmental conditions is unknown and requires further
investigation.
Acknowledgments
The authors acknowledge the generous funding assistance of
the Australian Institute of Sport and the University of Western
Australia. In addition, Philo Saunders (Australian Institute of
Sport) is acknowledged for his valuable assistance with study
design.
Contrast Water Therapy Recovery, Running 139
References
1. Bradley P, Sheldon W, Wooster B, Olsen P, Boanas P,
Krustrup P. High-intensity running in English FA Premier
League soccer matches. J Sports Sci. 2009;27:159–168.
PubMed doi:10.1080/02640410802512775
2. Hausswirth C, Le Meur L, Bieuzen F, Brisswalter J,
Bernard T. Pacing strategy during the initial phase of the
run in triathlon: Inuence on overall performance. Eur J
Appl Physiol. 2010;108:1115–1123. PubMed doi:10.1007/
s00421-009-1322-0
3. Spencer M, Rechichi C, Lawrence S, Dawson B, Bishop
D, Goodman C. Time-motion analysis of elite field
hockey during several games in succession: A tournament
scenario. J Sci Med Sport. 2005;8:382–391. PubMed
doi:10.1016/S1440-2440(05)80053-2
4. Ingram J, Dawson B, Goodman C, Wallman K, Beilby
J. Effect of water immersion methods on post-exercise
recovery from simulated team sport exercise. J Sci Med Sport.
2009;12:417–421. PubMed doi:10.1016/j.jsams.2007.12.011
5. Vaile J, Gill N, Blazevich A. The effect of contrast water
therapy on symptoms of delayed onset muscle soreness.
J Strength Cond Res. 2007;21:697–702. PubMed
6. Vaile J, Halson S, Gill N, Dawson B. Effect of hydro-
therapy on recovery from fatigue. Int J Sports Med.
2008;29:539–544. PubMed doi:10.1055/s-2007-989267
7. Pournot H, Bieuzen F, Dufeld R, Lepretre P-M, Coz-
zolino C, Hausswirth C. Short term effects of various
water immersions on recovery from exhaustive intermit-
tent exercise. Eur J Appl Physiol. 2011;111:1287–1295.
PubMed doi:10.1007/s00421-010-1754-6
8. Buchheit M, Horobeanu C, Mendez-Villanueva A, Simp-
son B, Bourdon P. Effects of age and spa treatment on
match running performance over two consecutive games
in highly trained young soccer players. J Sports Sci.
2011;2011 Epub.
9. Vaile J, Halson S, Gill N, Dawson B. Effect of hydro-
therapy on signs and symptoms of delayed onset muscle
soreness. Eur J Appl Physiol. 2008;102:447–455. PubMed
doi:10.1007/s00421-007-0605-6
10. Versey N, Halson S, Dawson B. Effect of contrast water
therapy duration on recovery of cycling performance: A
dose–response study. Eur J Appl Physiol. 2011;111:37–46.
PubMed doi:10.1007/s00421-010-1614-4
11. Coffey V, Leveritt M, Gill N. Effect of recovery modal-
ity on 4-hour repeated treadmill running performance
and changes in physiological variables. J Sci Med Sport.
2004;7:1–10. PubMed doi:10.1016/S1440-2440(04)80038-0
12. Kinugasa T, Kilding AE. A comparison of post-match
recovery strategies in youth soccer players. J Strength
Cond Res. 2009;23:1402–1407. PubMed doi:10.1519/
JSC.0b013e3181a0226a
13. Hamlin MJ. The effect of contrast temperature water
therapy on repeated sprint performance. J Sci Med
Sport. 2007;10:398–402. PubMed doi:10.1016/j.
jsams.2007.01.002
14. Hamlin MJ. The effect of recovery modality on blood lac-
tate removal and subsequent repetitive sprint performance
in netball players. N Z J Sports Med. 2007;34:12–17.
15. Dawson B, Gow S, Modra S, Bishop D, Stewart G.
Effects of immediate post-game recovery procedures on
muscle soreness, power and exibility levels over the next
48 hours. J Sci Med Sport. 2005;8:210–221. PubMed
doi:10.1016/S1440-2440(05)80012-X
16. Myrer JW. Contrast therapy and intramuscular temperature
in the human leg. J Athl Train. 1994;29:318–322. PubMed
17. Higgins D, Kaminski TW. Contrast therapy does not
cause uctuations in human gastrocnemius intramuscular
temperature. J Athl Train. 1998;33:336–340. PubMed
18. Yamane M, Teruya H, Nakano M, Ogai R, Ohnishi N,
Kosaka M. Post-exercise leg and forearm exor muscle
cooling in humans attenuates endurance and resistance
training effects on muscle performance and on circulatory
adaptation. Eur J Appl Physiol. 2006;96:572–580. PubMed
doi:10.1007/s00421-005-0095-3
19. Howatson G, Goodall S, van Someren K. The infuence
of cold water immersions on adaptation following a
single bout of damaging exercise. Eur J Appl Physiol.
2009;105:615–621. PubMed doi:10.1007/s00421-008-
0941-1
20. Higgins T, Heazlewood I, Climstein M. A random control
trial of contrast baths and ice baths for recovery during
competition in U/20 rugby union. J Strength Cond Res.
2011;25:1046–1051. PubMed
21. Cochrane D. Alternating hot and cold water immersion for
athlete recovery: A review. Phys Ther Sport. 2004;5:26–32.
doi:10.1016/j.ptsp.2003.10.002
22. Hing WA, White SG, Bouaaphone A, Lee P. Con-
trast therapy: A systematic review. Phys Ther Sport.
2008;9:148–161. PubMed doi:10.1016/j.ptsp.2008.06.001
23. King M, Dufeld R. The effects of recovery interven-
tions on consecutive days of intermittent sprint exercise.
J Strength Cond Res. 2009;23:1795–1802. PubMed
doi:10.1519/JSC.0b013e3181b3f81f
24. Connolly D, Sayers S, McHugh M. Treatment and preven-
tion of delayed onset muscle soreness. J Strength Cond
Res. 2003;17:197–208. PubMed
25. Green H. Mechanisms of muscle damage in intense
exercise. J Sports Sci. 1997;15:247–256. PubMed
doi:10.1080/026404197367254
26. Harris D, Atkinson G. International journal of sports
medicine – Ethical standards in sport and exercise science
research. Int J Sports Med. 2009;30:701–702. PubMed
doi:10.1055/s-0029-1237378
27. Robertson E, Saunders P, Pyne D, Aughey R, Anson J,
Gore C. Reproducibility of performance changes to simu-
lated live high/train low altitude. Med Sci Sports Exerc.
2010;42:394–401. PubMed
28. Marfell-Jones M, Olds T, Stewart A, Carter L. Interna-
tional Standards for Anthropometric Assessment. Potchef-
stroom, NM: International Society for the Advancement
of Kinanthropometry; 2006.
29. Borg G. Borg’s Perceived Exertion and Pain Scales.
Champaign, IL: Human Kinetics; 1998.
30. Young AJ, Sawka MN, Epstein Y, Decristofano B, Pandolf
KB. Cooling different body surfaces during upper and
lower body exercise. J Appl Physiol. 1987;63:1218–1223.
PubMed
140 Versey, Halson, and Dawson
31. Cleak M, Eston R. Muscle soreness, swelling, stiffness
and strength loss after intense eccentric exercise. Br J
Sports Med. 1992;26:267–272. PubMed doi:10.1136/
bjsm.26.4.267
32. Hopkins W. Spreadsheets for analysis of controlled trials,
with adjustment for a subject characteristic. Sportscience
[serial on the Internet]. 2006;10: Available from: http://
sportsci.org/2006/wghcontrial.htm.
33. Batterham A, Hopkins W. Making meaningful infer-
ences about magnitudes. Int J Sports Physiol Perform.
2006;1:50–57. PubMed
34. Hopkins WG, Hawley JA, Burke LM. Design and
analysis of research on sport performance enhance-
ment. Med Sci Sports Exerc. 1999;31:472–485. PubMed
doi:10.1097/00005768-199903000-00018
35. Hopkins WG. Measures of reliability in sports medi-
cine and science. Sports Med. 2000;30:1–15. PubMed
doi:10.2165/00007256-200030010-00001
36. Spencer M, Bishop D, Dawson B, Goodman C. Physi-
ological and metabolic responses of repeated-sprint
activities: Specic to eld-based team sports. Sports Med.
2005;35:1025–1044. PubMed doi:10.2165/00007256-
200535120-00003
37. Duffield R, Dawson B, Goodman C. Energy system
contribution to 1500- and 3000-metre track run-
ning. J Sports Sci. 2005;23:993–1002. PubMed
doi:10.1080/02640410400021963
38. Fradkin A, Zazryn T, Smoliga J. Effects of warming-up
on physical performance: A systematic review with meta-
analysis. J Strength Cond Res. 2010;24:140–148. PubMed
doi:10.1519/JSC.0b013e3181c643a0
