The effects of lower body passive heating combined with mixed-method cooling during half-time on second-half intermittent sprint performance in the heat

Abstract

Purpose
This study examined the effects of combined cooling and lower body heat maintenance during half-time on second-half intermittent sprint performances.

Methods
In a repeated measures design, nine males completed four intermittent cycling trials (32.1 ± 0.3 °C and 55.3 ± 3.7% relative humidity), with either one of the following half-time recovery interventions; mixed-method cooling (ice vest, ice slushy and hand cooling; COOL), lower body passive heating (HEAT), combined HEAT and COOL (COMB) and control (CON). Peak and mean power output (PPO and MPO), rectal (Tre), estimated muscle (Tes-Mus) and skin (TSK) temperatures were monitored throughout exercise.

Results
During half-time, the decrease in Tre was substantially greater in COOL and COMB compared with CON and HEAT, whereas declines in Tes-Mus within HEAT and COMB were substantially attenuated compared with CON and COOL. The decrease in TSK was most pronounced in COOL compared with CON, HEAT and COMB. During second-half, COMB and HEAT resulted in a larger decrease in PPO and MPO during the initial stages of the second-half when compared to CON. In addition, COOL resulted in an attenuated decrease in PPO and MPO compared to COMB in the latter stages of second-half.

Conclusion
The maintenance of Tes-Mus following half-time was detrimental to prolonged intermittent sprint performance in the heat, even when used together with cooling.

Full text

Vol.:(0123456789)
1 3
European Journal of Applied Physiology
https://doi.org/10.1007/s00421-019-04177-8
ORIGINAL ARTICLE
The eects oflower body passive heating combined
withmixed‑method cooling duringhalf‑time onsecond‑half
intermittent sprint performance intheheat
JackySoo1,2· GabrielTang1,2· SaravanaPillaiArjunan3· JoelPang1· AbdulRashidAziz1· MohammedIhsan4
Received: 17 December 2018 / Accepted: 11 June 2019
© Springer-Verlag GmbH Germany, part of Springer Nature 2019
Abstract
Purpose This study examined the effects of combined cooling and lower body heat maintenance during half-time on second-
half intermittent sprint performances.
Methods In a repeated measures design, nine males completed four intermittent cycling trials (32.1 ± 0.3°C and 55.3 ± 3.7%
relative humidity), with either one of the following half-time recovery interventions; mixed-method cooling (ice vest, ice
slushy and hand cooling; COOL), lower body passive heating (HEAT), combined HEAT and COOL (COMB) and control
(CON). Peak and mean power output (PPO and MPO), rectal (Tre), estimated muscle (Tes-Mus) and skin (TSK) temperatures
were monitored throughout exercise.
Results During half-time, the decrease in Tre was substantially greater in COOL and COMB compared with CON and HEAT,
whereas declines in Tes-Mus within HEAT and COMB were substantially attenuated compared with CON and COOL. The
decrease in TSK was most pronounced in COOL compared with CON, HEAT and COMB. During second-half, COMB and
HEAT resulted in a larger decrease in PPO and MPO during the initial stages of the second-half when compared to CON. In
addition, COOL resulted in an attenuated decrease in PPO and MPO compared to COMB in the latter stages of second-half.
Conclusion The maintenance of Tes-Mus following half-time was detrimental to prolonged intermittent sprint performance
in the heat, even when used together with cooling.
Keywords Intermittent sprint performance· Mixed-method cooling· Passive heating· Half-time intervention· Team sports
Abbreviations
MPO Mean power output
COMB Combined upper body cooling and lower body
passive heating
CON Control
COOL Upper body cooling
HEAT Lower body passive heating
HR Heart rate
PPO Peak power output
SS Single sprint
RPE Rating of perceived exertion
RS Repeated sprint
TB Body temperature
TC Core temperature
Tes-Mus Estimated muscle temperature
Tm Muscle temperature
Tre Rectal temperature
TS Thermal sensation
TSK Mean skin temperature
USG Urine-specific gravity
VO2peak Peak oxygen uptake
Communicated by George Havenith.
* Jacky Soo
jackysds@gmail.com
1 Sport Physiology, Sport Science andMedicine, Singapore
Sport Institute, 3 Stadium Drive, Singapore397630,
Singapore
2 Physical Education andSports Science, Nanyang
Technological University, Singapore, Singapore
3 Physical, Sports andOutdoor Education Branch, Student
Development Curriculum Division, Ministry ofEducation,
Singapore, Singapore
4 Athlete Health andPerformance Research Centre, Aspetar
Orthopedic andSports Medicine Hospital, Doha, Qatar

European Journal of Applied Physiology
1 3
Introduction
Intermittent team games are typically played for
60–90min, with temporary breaks lasting between 10 and
20min usually evident at every quarter or at the mid-way
point (i.e. half-time). During this period, players engage
in predominantly passive activities such as receiving tac-
tical instructions, attending to injuries and consuming
ergogenic aids (Russell etal. 2015b). Recent studies have
highlighted several shortcomings associated with the pas-
sive nature of recovery during half-time intervals. Notably,
this includes the decline in core (TC) and muscle tempera-
ture (Tm) (Mohr etal. 2004), which has been shown to
negatively influence sprint performances, especially dur-
ing the period of match recommencement (Edholm etal.
2015; Mohr etal. 2004). Indeed, an elevated Tm is an
important determinant of sprint performance, given that
it has been shown to improve muscle contractility (i.e.
shorter contraction and half-relaxation time) (Racinais
etal. 2016; Sargeant 1987), as well as enhance anaerobic
ATP turnover and muscle fibre conduction velocity (Gray
etal. 2006).
Previous studies have demonstrated the effectiveness of
heated or insulated clothing in preserving post-warm-up
Tm and subsequently improved sprint exercise performance
(Faulkner etal. 2012, 2013). For instance, Faulkner etal.
(2012) demonstrated that the maintenance of Tm follow-
ing warm-up using heated garments resulted in enhanced
peak power output (~ 9%) during a subsequent 30-s sprint.
Accordingly, the demonstrated importance of preserving
Tm has led to several studies exploring the use of such
strategies during half-time to minimise the decline in body
temperatures. Kilduff etal. (2013) showed that attenuat-
ing the loss in post-warm-up TC (and likely Tm) through
the use of blizzard jackets led to an improvement in peak
power output and repeated sprint (RS) performance
(6 × 40m) in rugby league players. In a subsequent study,
the use of blizzard jackets in-between two bouts of RS
sequences (6 × 40m) resulted in improved performance
within the second RS sequence, particularly within the
first two sprints (Russell etal. 2015a). However, it must
be noted that whilst heat maintenance strategies might be
beneficial upon the moments of match recommencement,
its efficacy can be contentious during prolonged intermit-
tent sprinting/running typifying actual match durations.
Given that team games typically last 60–90min, and
with several upcoming major competitions held in hot
environments, such strategies will likely exacerbate the
increase in body temperature (TB), hastening the develop-
ment of hyperthermia-related fatigue (González-Alonso
etal. 1999). In support, Skein etal. (2012) demonstrated
that whilst sprint performances during initial periods of
a prolonged intermittent sprint protocol were enhanced
following passive heating, performances during the latter
stages of the exercise protocol were considerably impaired,
with the excessive increase in TC reasoned as the contrib-
uting factor. The mechanisms underpinning the decrease in
exercise performance following hyperthermia may include
a decrease in neural drive and subsequent failure to fully
activate the exercising musculatures (i.e. central fatigue;
Girard etal. 2011; Nybo 2008), as well as an increase
in cardiovascular strain, characterised by an increase in
heart rate and skin blood flow, with concomitant decreases
in stroke volume and cardiac output (Périard etal. 2011;
Shaffrath and Adams 1984). As such, the use of passive
heat maintenance garments needs to be further deliberated
to mitigate between the neuromuscular benefits conferred
at the start of match recommencement, and the exacer-
bated increase in body temperatures that will likely ensue
when exercise is prolonged.
One possibility would be to combine cooling and heat
maintenance strategies, targeted at lowering TC whilst
maintaining the Tm of key musculatures involved in exer-
cise. Precooling is a temperature regulation strategy under-
taken to lower pre-exercise TC, consequently increasing
body heat storage capacity and ameliorating the develop-
ment of hyperthermia-induced fatigue. Strategies such as
cold water immersion, ice slurry ingestion and the use of
cooling garments are some traditional methods shown to
improve exercise performance across a variety of prolonged
continuous and intermittent modalities (Castle etal. 2006;
Ihsan etal. 2010; Siegel etal. 2010; Zimmermann etal.
2018). To improve the practicality of administering cooling
interventions during half-time, some have proposed mixed-
method modalities, which include possible combinations
with the use of cooling vests (torso), cold towels (neck and
head regions), ice slurry ingestion and hand immersion (Duf-
field etal. 2009; Minett etal. 2012). Such mixed modalities
applied prior to exercise have been shown to improve run-
ning performances (i.e. maintained RS performances and
improved self-paced running velocity) (Duffield etal. 2009;
Minett etal. 2012) which paralleled the decrease in TC and
physiological strain following cooling.
A recent study by Beaven etal. (2018) demonstrated
that combined ice slurry ingestion and passive heat main-
tenance prior to exercise improved sprinting performances
(sprint timings and fatigue index) during a 5 × 40m RS
protocol. However, we are unaware of studies that have
examined the effects of combined cooling and lower body
heat maintenance as a half-time strategy on second-half
performance during prolonged intermittent sprinting in
the heat. Such combined treatment potentially minimises
the decline in Tm, and hence muscle contractile function
following half-time intervals, yet preserves/increases
the capacity for heat storage (lower TC) essential for

European Journal of Applied Physiology
1 3
maintaining prolonged exercise performance. Therefore,
the purpose of this study was to investigate the effects of
combined core cooling and lower body heat maintenance
during a half-time interval on second-half intermittent
sprint performance performed in the heat.
Methods
Participants
Nine male participants (age: 26.5 ± 4.2 years, height:
174.0 ± 5.9 cm, mass: 69.2 ± 13.8 kg, VO2peak:
42.7 ± 3.9ml.kg−1min−1) participated in the present study.
Participants were physically active and/or recreational
intermittent team sports players without any previous his-
tory of heat illness and injuries. They were informed of
the procedures and potential risks involved in this study,
and a written informed consent was subsequently obtained.
They were also instructed to refrain from strenuous physi-
cal activity, alcohol, and caffeine 24-h before all tests. This
study was approved by the Human Research Ethics Com-
mittee of the Singapore Sport Institute. This study was
conducted in Singapore during daytime, where the ambi-
ent temperature and relative humidity was ~ 30–33°C and
~ 60–80%, respectively. Weather forecast was based on an
online meteorological service (https ://www.weath er.gov.
sg/clima te-clima te-of-singa pore/).
Experimental overview
Each participant visited the laboratory on five separate
occasions, which included one pre-experimental and four
experimental sessions. During the pre-experimental ses-
sion, participants performed a graded exercise test to
determine their peak oxygen uptake (VO2peak), followed by
familiarisation to the main experimental protocol. During
the experimental sessions, participants performed an inter-
mittent sprint protocol consisting of two 33-min halves,
separated by a passive rest interval of 15min. Participants
were administered either one of four different interven-
tions in a crossover manner during the half-time interval,
before undertaking the second-half of intermittent sprint
activity. The experimental interventions included (1)
lower body passive heating (HEAT), (2) upper body cool-
ing directed at lowering TC (COOL), (3) combined upper
body cooling and lower body passive heating (COMB) or
(4) control (CON). All sessions were separated by at least
72h, conducted at the same time of the day.
Graded exercise test andfamiliarisation session
The graded exercise test and familiarisation session were
performed in standard laboratory conditions of 23.4 ± 0.6°C
and 67.4 ± 2.7% RH. Participants performed the graded
exercise test on a cycle ergometer (Lode Excalibur Sport,
Groningen, The Netherlands) and expired air was analysed
using an automated system (TrueOne 2400, Parvo Medics,
Salt Lake City, UT). The test commenced at a power output
of 100W, with 20W increments every minute thereafter
until volitional exhaustion. After a 30-min recovery period,
participants were familiarised to the intermittent sprint pro-
tocol (Wattbike Ltd., Nottingham, UK), as well to the gen-
eral procedures and equipment involved in the experimental
sessions. Participants completed four to five 4-s sprinting
efforts where they were familiarised with the sequence of
activity and the instructions of the tester. In addition, air
resistance was adjusted to determine the optimal resistance
that would achieve maximal power output. Seat and han-
dlebar positions were adjusted according to preference, and
replicated during all experimental trials. The same cycle
ergometer was used throughout the experiment, and par-
ticipants’ attire (i.e. shorts, socks and sport shoes) was kept
consistent as well.
Intermittent sprint protocol
Upon arrival to the laboratory, a urine sample was col-
lected and nude body mass was measured. Participants
then self-inserted a disposable rectal probe to a depth of
12-cm beyond the anal sphincter. Subsequently, participants
were fitted with a heart rate (HR) monitor (Polar RS400,
Polar Electro, Finland), and wireless skin temperature sen-
sors. Following instrumentation, participants proceeded
into a climate-controlled room maintained at 32.1 ± 0.3°C
and 55.3 ± 3.7% RH and performed a standardised 8-min
warm-up. The warm-up consisted of 4-min of cycling at
50% VO2peak followed by two 1-min blocks of 30s at 70%
VO2peak and 30s of passive rest. This was immediately fol-
lowed by two 4-s maximal sprints separated by a 2min of
active recovery at 35% VO2peak. Participants were then given
2min of passive rest before commencing the intermittent
sprint protocol adapted from previous research (Bishop and
Maxwell 2009; Schneiker etal. 2006). This protocol was
specifically designed to simulate the physiological demands
of match-play in team sports (Bishop and Maxwell 2009;
Schneiker etal. 2006). This test consisted of approximately
two 33-min halves, separated by a 15-min recovery period.
Specifically, each half consisted of 16 × 2-min blocks, which
included a 4-s maximal sprint, 100s of active recovery, and
20s of passive recovery (Fig.1). On two occasions during
each half (following the 8th and 16th single sprint block),
participants performed a RS sequence consisting of five 3-s

European Journal of Applied Physiology
1 3
sprints with 18s of recovery between successive sprints.
During the last 5s of each passive recovery period, partici-
pants were instructed to assume a “ready” position, and a
3-s countdown was given prior to the start of each sprint.
Cycling was performed in a seated position throughout
the trial. Strong verbal encouragement was provided for
every sprint effort. Participants were given 100ml of water
(29.0 ± 1.6°C) to ingest at the 18th min during each halves
for all conditions. Upon completion of the trial, another
urine sample was collected. Participants then towelled
themselves dry, and nude body mass was measured. Peak
(PPO) and mean (MPO) power output (W) for every sprint
effort was retrieved using the Wattbike software, and sub-
sequently downloaded for offline analysis. The second-half
single sprint (SS) data were subsequently grouped into rep-
resentative SETS 1 (SS 1–4), 2 (SS 5–8), 3 (SS 9–12), and
4 (SS 13–16). Similarly, the first and second RS sequence
(five 3-s sprints) during second-half was grouped as SETS
1 and 2, respectively.
Half‑time interval
Immediately following the first-half, participants proceeded
out of the climate room, and were administered with either
HEAT, COOL, COMB or CON in standard laboratory condi-
tions of 21.1 ± 0.3°C and 71.4 ± 1.9% RH. The HEAT trial
involved participants wrapping a blizzard blanket (Blizzard
Survival Blanket, Blizzard Protection Systems Ltd, UK)
around their lower body that covered the gluteus, quadriceps,
hamstrings and gastrocnemius while being seated. The blizzard
blanket was made from ReflexCell™ material. The material
provides insulation by trapping warm still air, and its reflective
surface acts to block radiated heat, while its elastic proper-
ties cause the material to clinch the body reducing convection
(Allen etal. 2010). The COOL trial involved a combination of
cooling modalities which included hand cooling, wearing an
ice vest and consumption of an ice slushy (Duffield etal. 2009;
Ihsan etal. 2010). Specifically, participants were instrumented
with an ice vest (Artic Heat, Brisbane, Australia) covering the
torso region. The ice vest was made from an inner polyester
fabric and micromesh outer consisting of pockets filled with
crystals. Prior to use, the ice vest was immersed in water to
activate the crystals into a viscous gel and subsequently frozen
at approximately − 20°C. Hand cooling was undertaken on the
participants’ non-dominant hand, immersed to the wrist level
in a water bath maintained at ~ 9.0°C (Minett etal. 2012).
In addition, participants consumed 3.4gkg−1 body mass of
ice slushy (0.7 ± 0.6°C) made using a commercially avail-
able ice slushy machine. Servings were provided in portions
of 75–100g at 4–5-min intervals over the 15-min half-time
interval, ensuring the stipulated amount of ice to be consumed
is reached. The COMB trial involved the combined adminis-
tration of HEAT and COOL interventions. During the CON
trial, participants rested passively in a seated position over the
15-min interval. In addition, participants in both the HEAT
and CON trial consumed 3.4gkg−1 body mass of tap water
(20.2 ± 0.2°C) served at similar proportions and timings in all
other conditions. Participants returned to the climate room at
the 14th min, and assumed position on the cycle ergometer in
preparation for the second-half. Physiological measurements
were also recorded prior to the start of the second-half.
Temperature measurements andcalculations
Participants’ Tre was monitored throughout via a data logger
(Cole Parmer Thermistor Thermometer 8502-12, Cole Par-
mer Instrument Co., USA) connected to a disposable rectal
probe (YSI401, Yellow Springs Instruments, Yellow Springs,
OH, USA). Temperature sensors (iButton, DS1922L, Maxim
Integrated Products Inc, Sunnyvale, CA, USA) were attached
to the chest (mid-point of the pectoralis major at the under-
arm level; Tchest), arm (mid-belly of the lateral tricep; Tarm),
thigh (lateral vastus lateralis, 15cm superior to the patella;
Tthigh), and calf (mid-belly of the lateral gastrocnemius; Tcalf)
on the right-hand side of the body using an adhesive dressing
(Tegaderm, 3M Healthcare, USA), and mean skin temperature
(TSK) was calculated using the following equation (Ramana-
than 1964):
Fig. 1 Schematic representation of one half of the intermittent-sprint
protocol. The protocol consisted of 16 single sprints (4 s) interspersed
by 100 s of active recovery, followed by 20 s of passive recovery.
Twenty seconds following the 8th and 16th single sprint, participants
performed a repeated sprint sequence consisting of five 3 s sprints
with 21 s of passive recovery in-between

European Journal of Applied Physiology
1 3
Body temperature was calculated using the following equa-
tion (Burton 1935):
Quadriceps Tm was estimated using recent methods devel-
oped by Flouris etal. (2015). Briefly, an insulation disk
(iDISK) was placed directly on top of a temperature sensor
on the left lateral mid-thigh (vastus lateralis) at a distance
of 15cm superior to the patella, and secured with adhesive
dressing. The iDisk was made by cutting a circle measuring
50mm in diameter and 4.8mm in thickness from an open
cell foam neoprene material. Changes in Tm during exercise
and post-exercise recovery were calculated by the following
equations, respectively (Flouris etal. 2015):
where iDISK is the current temperature of the vastus later-
alis as measured by the temperature sensor; iDISKlag4 is the
difference in temperature between the current value and the
temperature 4min before.
Perceptual ratings
Rating of perceived exertion (RPE) and thermal sensation
(TS) were recorded before the start of the intermittent sprint
test and 9, 17, 25 and 33min into each half. Rating of per-
ceived exertion was assessed using Borg’s 6-to-20 scale
(Borg 1982). Thermal sensation was assessed using a 9-point
scale ranging from 0 (unbearably cold) to 8 (unbearably hot)
(Toner etal. 1986).
Hydration anduid loss
Urine samples collected before and after the trial were used
to determine urine-specific gravity (USG) using a clinical
refractometer (Atago UG-1, Atago Co. Ltd, Japan). Changes
in nude body mass were used to assess fluid loss or gain
during the trial.
Statistical analysis
All within- and between-group comparisons were under-
taken using magnitude-based inferences (Hopkins 2006).
Changes in all variables were analysed in raw units rela-
tive to the smallest worthwhile change (SWC) with 90%
(a)
TSK
=
0.3
×
Tchest
+
0.3
×
Tarm
+
0.2
×
Tthigh
+
0.2
×
Tcalf.
(b)
TB
=
0.65
×
TC
+
0.35
×
TSK .
(c)
Estimated muscle temperature (
Tes−Mus
)during exercise
=iDISK ×0.599−iDISK
lag4
×0.311 +15.63,
(d)
T
es−Mus
during post
−
exercise recovery
=iDISK ×0.657−iDISK
lag4
×0.538 +
13.283,
confidence intervals. The SWC (0.2 × between-subject SD)
for PPO and MPO during SS and RS, as well as temperature
responses (Tre, TSK, TB, Tes-Mus) and other measures (i.e. HR,
RPE, TS) were each determined from first-half (i.e. mean of
all SWC within each of the four conditions). The SWC for
changes in all temperature responses during half-time was
determined from the CON group. Quantitative chances of
higher or lower differences were evaluated qualitatively as:
< 1%, almost certainly not; 1–5%, very unlikely; 25–75%,
possible; 75–95%, likely; 95–88%, very likely; > 99% almost
certain. In addition, differences were considered unsubstan-
tial/unclear if the probability of the difference being sub-
stantially greater or lower were both > 5% (Hopkins etal.
2009). The magnitude-based inference statistics are also
supplemented with repeated measures analysis of variance,
where p values for all main effects are reported.
Results
Main effects for time, condition and interaction for changes
in PPO during SS were p = 0.003, p = 0.672 and p = 0.413,
respectively. Mean changes in PPO (90% CI) with quantita-
tive chances are presented in Fig.2. There were substan-
tial decreases in PPO within all experimental conditions in
the second-half (SETS 1, 3 and 4) compared with the first
(Fig.2a–d). Between conditions, COOL resulted in a smaller
decrement in PPO compared with CON at SET 4 (Fig.2h),
and compared with COMB at SET 3 (Fig.2g). In addition,
COMB resulted in a greater decrease in PPO compared with
CON at SET 1 (Fig.2e).
Main effects for time, condition and interaction for
changes in MPO during SS were p = 0.002, p = 0.484 and
p = 0.175, respectively. Mean changes in MPO (90% CI)
with quantitative chances are presented in Fig.3. Substan-
tial decreases in MPO during SS were observed within all
experimental conditions, particularly during SET 1, 3 and 4
(Fig.3a–d). Between conditions, COOL resulted in smaller
decrements in MPO compared with CON (SET 4, Fig.3h),
HEAT (SET 4, Fig.3h) and COMB (SETS 1, 3 and 4,
Fig.3e, g, h). The COMB treatment also resulted in a lower
MPO compared with CON (SETS 1, Fig.3e).
Time, condition and interaction effects for changes in
PPO during RS were p < 0.001, p = 0.273 and p = 0.384,
respectively. Substantial decreases in PPO were evident
within all experimental conditions within the second, com-
pared to the preceding half during RS (Fig.4a–d). Between
conditions, the decrease in PPO was substantially smaller
in COOL compared with CON (SET 2, Fig.4f) and COMB
(SET 1 and SET 2, Fig.4e, f). Time, condition and interac-
tion effects for changes in MPO during RS were p < 0.001,
p = 0.076 and p = 0.093, respectively. Substantial decre-
ments in MPO were observed in the second-half within all

European Journal of Applied Physiology
1 3
Fig. 2 Changes in peak power output (PPO) during single sprints (SS) within CON (a), COOL (b), HEAT (c) and COMB (d), as well as between conditions during SET 1 (e), SET 2 (f), SET
3 (g) and SET 4 (h). Changes are presented in raw units (90% confidence intervals) relative to the smallest worthwhile change (i.e., shaded region). Quantitative chances (%) of increase/trivial/
decrease are indicated in parentheses and qualitatively interpreted as: > 25–75%, possibly; > 75–95%, likely; > 95–99%, very likely; > 99%, almost certainly

European Journal of Applied Physiology
1 3
Fig. 3 Changes in mean power output (MPO) during single sprints (SS) within CON (a), COOL (b), HEAT (c) and COMB (d), as well as between conditions during SET 1 (e), SET 2 (f), SET
3 (g) and SET 4 (h). Changes are presented in raw units (90% confidence intervals) relative to the smallest worthwhile change (i.e., shaded region). Quantitative chances (%) of increase/trivial/
decrease are indicated in parentheses and qualitatively interpreted as: > 25–75%, possibly; > 75–95%, likely; > 95–99%, very likely; > 99%, almost certainly

European Journal of Applied Physiology
1 3
Fig. 4 Changes in peak (PPO) and mean power output (MPO) during repeated sprints (RS) within CON (a), COOL (b), HEAT (c) and COMB (d). Between condition changes for PPO during
RS for SET 1 (e) and SET 2 (f) and for MPO for SET 1 (g) and SET 2 (h). Changes are presented in raw units (90% confidence intervals) relative to the smallest worthwhile change (i.e., shaded
region). Quantitative chances (%) of increase/trivial/decrease are indicated in parentheses and qualitatively interpreted as: > 25–75%, possibly; > 75–95%, likely; > 95–99%, very likely; > 99%,
almost certainly

European Journal of Applied Physiology
1 3
experimental condition except COOL (Fig.4a–d). Between
conditions, COOL resulted in a smaller decrement in APO
compared with CON and COMB in SET and SET 2 (Fig.4g,
h).
Differences in Tre at the end of the first-half were
unclear between conditions (CON: 38.6 ± 0.6°C; COOL:
38.6 ± 0.4°C; HEAT: 38.6 ± 0.6°C; COMB: 38.8 ± 0.4°C,
p = 0.383). Main effects for time, condition and interac-
tion for changes in Tre during half-time were p < 0.001,
p = 0.400 and p = 0.075, respectively (Fig.5a). At the end
of the 15min half-time period, the decrease in Tre within
COOL and COMB was greater compared with CON and
HEAT. Differences in TSK at the end of the first-half were
unclear between conditions (CON: 36.5 ± 0.9°C; COOL:
36.7 ± 0.7°C; HEAT: 36.6 ± 0.6°C; COMB: 36.5 ± 0.7°C,
p = 0.446), whilst time, condition and interaction effects for
changes during half-time were all p < 0.001. Specifically,
while a decrease in TSK was evident in all experimental
conditions during half-time (Fig.5b), the decrease within
COOL was the most pronounced, and was greater com-
pared with CON (− 1.2 ± 0.7 °C), HEAT (− 1.8 ± 0.6°C)
and COMB (− 1.5 ± 0.7°C). Moreover, the decline in TSK
in both HEAT (0.6 ± 0.6°C) and COMB (0.3 ± 0.7 °C)
was smaller compared with CON. Body temperature at the
end of the first-half was similar between conditions (CON:
37.7 ± 0.3°C; COOL: 37.9 ± 0.3°C; HEAT: 37.8 ± 0.4°C;
COMB: 37.9 ± 0.3°C, p = 0.922). Main effect for time was
p = 0.001, while condition and interaction effects were
p < 0.001 for changes in TB during half-time. Similar to
changes in TSK, the decline in TB was most pronounced
in COOL compared with CON (− 0.6 ± 0.2 °C), HEAT
(− 0.7 ± 0.2°C) and COMB (− 0.3 ± 0.2 °C) (Fig.5c). In
addition, the decline in TB within HEAT was substantially
lower compared with CON (− 0.3 ± 0.2°C) and COMB
(− 0.3 ± 0.2°C). Estimated muscle temperature was simi-
lar at the end of the first-half (CON: 38.3 ± 0.3°C; COOL:
38.3 ± 0.2°C; HEAT: 38.3 ± 0.3°C; COMB: 38.4 ± 0.3°C,
p = 0.211). Main effects for time, condition and interaction
for changes in Tes-Mus were all p < 0.001. The decline within
HEAT and COMB was substantially attenuated compared
with COOL and CON (Fig.5d). Moreover, the decrease in
Tes-Mus was substantially greater in COMB compared with
HEAT towards the end of the half-time period (~ 0.2°C).
Changes in all temperature measurements during the sec-
ond-half are presented in Fig.6. Time, condition and interac-
tion effects for changes in Tre were p < 0.001, p = 0.796 and
p = 0.726, respectively, where changes in Tre were similar
between conditions upon commencement and throughout
the second-half (Fig.6a). Main effects for time, condition
and interaction in TSK (Fig.6b) during the second-half were
p < 0.001, p = 0.013 and p < 0.001, respectively. Specifically,
TSK in COOL was lower compared with CON (0.8–0.9°C),
Fig. 5 Changes in rectal temperature (Tre, a), mean skin temperature
(TSK, b), mean body temperature (TB, c) and estimated vastus later-
alis muscle temperature (Tes-Mus , d) during half-time interval. Error
bars have been omitted to improve visual clarity. Between group
differences are denoted by symbols; *CON vs. COMB, #CON vs.
COOL, +CON vs. HEAT, (*)COMB vs. COOL, (#)COMB vs. HEAT,
(+)COOL vs. HEAT, and number of symbols indicate probability (%)
of an increase or decrease (eg., * > 25–75%, possibly; ** > 75–95%,
likely; *** > 95–99%, very likely; **** > 99%, almost certainly)

European Journal of Applied Physiology
1 3
HEAT (0.7–1.2°C) and COMB (0.6°C) during the begin-
ning portions of the second-half (i.e. 0–8min), after which
between-group differences were unclear. Moreover, upon
commencement of the second-half, TSK in HEAT was higher
compared with COMB (0.6 ± 0.6°C). Time, condition and
interaction effects for changes in TB (Fig.6c) during the
second-half were p < 0.001, p = 0.092 and p < 0.001, respec-
tively. Changes in TB during the beginning portions of the
second-half were the lowest in COOL compared with CON
(0.3°C) and HEAT (0.3–0.5°C). In addition, TB in COMB
was lower compared with HEAT (0.3°C). Time, condition
and interaction for changes in Tes-Mus (Fig.6d) during the
second-half were all p < 0.001. Estimated muscle tempera-
tures in HEAT (37.9 ± 0.3°C) and COMB (37.8 ± 0.2°C)
at the start of the second-half were substantially higher
compared with CON (37.3 ± 0.4°C). In addition, Tes-Mus
in COOL (37.1 ± 0.4°C) was lower compared with CON
(− 0.3 to − 0.5°C), HEAT (− 0.3 to − 0.7°C) and COMB
(− 0.2 to − 0.7°C) at the beginning portions of the second-
half. Furthermore, Tes-Mus was substantially lower in COOL
(38.1 ± 0.3°C) compared with CON (38.3 ± 0.3°C) and
HEAT (38.3 ± 0.3°C) at the end of the second-half.
Changes in HR, RPE and TS are presented in Table1.
Time, condition and interaction effects for changes in HR
were p < 0.001, p = 0.573 and p = 0.076, respectively, with a
small difference in COOL compared with HEAT at the start
of second-half (111 ± 14 vs. 116 ± 14bpm), following which
all between-group differences in HR were unclear. Time,
condition and interaction effects for changes in RPE were
p < 0.001, p = 0.142 and p = 0.109, respectively. Changes in
RPE were similar at the start but were substantially greater
in HEAT and CON compared with COOL at 24min into
the second-half.
Time, condition and interaction effects for changes in
TS were all p < 0.001, respectively. Thermal sensation at
the start of second-half in COOL and COMB was lower
compared with HEAT and CON. Differences in TS were
substantially lower in COOL compared with HEAT and
CON, which persisted until 8min (Table1). In addition, TS
was lower in COMB at the onset of the second-half com-
pared with CON, and also substantially lower than HEAT
from the start of the second-half until the 8min. By 16min
into the second-half, all differences in TS between condi-
tions became unclear. Main effects for time, condition and
interaction for changes in USG were p = 0.044, p = 0.329
and p = 0.827, respectively. Differences in pre- (CON:
1.021 ± 0.007; COOL: 1.015 ± 0.010; HEAT: 1.018 ± 0.009;
COMB; 1.019 ± 0.006) and post-exercise USG (CON:
1.022 ± 0.006; COOL: 1.018 ± 0.01; HEAT: 1.019 ± 0.009;
COMB: 1.021 ± 0.008) were unclear between the groups.
Fig. 6 Changes in core temperature (Tre, a), mean skin temperature
(TSK, b), mean body temperature (TB, c) and estimated vastus later-
alis muscle temperature (Tes-Mus, d) during the 2nd half of intermit-
tent cycling protocol. Error bars have been omitted to improve visual
clarity. Between group differences are denoted by symbols; *CON
vs. COMB, #CON vs. COOL, +CON vs. HEAT, (*)COMB vs. COOL,
(#)COMB vs. HEAT, (+)COOL vs. HEAT, and number of symbols
indicate probability (%) of an increase or decrease (eg., * > 25–75%,
possibly; ** > 75–95%, likely; *** > 95–99%, very likely; **** >
99%, almost certainly)

European Journal of Applied Physiology
1 3
Discussion
The present study investigated the effects of combined core
cooling and lower body heat maintenance during half-time
interval, on a second-half intermittent sprint performance in
the heat. It was hypothesised that such a combined strategy
could minimise the decline in muscle contractile function
following the half-time interval, yet increase the capacity
for heat storage, which could benefit prolonged exercise
performance. Our findings demonstrate that the use of a
blizzard blanket was effective in attenuating the decrease in
Tes-Mus during half-time as observed in HEAT and COMB,
with minimal influence on Tre (Fig.5a, d). However, the
expected improvement in sprinting performance during
the initial stages of the second-half was not observed when
compared to CON (Figs.2, 3, 4). In addition, the decline in
performance during the latter stages of the second-half was
most pronounced within HEAT and COMB (Figs.2, 3, 4).
Conversely, the use of cooling alone resulted in a substantial
decrease in body temperatures (i.e. Tre, TSK, TB; Figs.5 and
6), resulting in improved perceptual responses (i.e. RPE and
TS; Table1), in line with the better maintenance of sprint
performance, particularly at the latter stages of the exercise
protocol.
The use of a heat maintenance garment in HEAT and
COMB significantly attenuated the decline in Tes-Mus during
half-time (Fig.5d). This was evident following the end of
the 15min half-time where Tes-Mus in HEAT and COMB
was substantially higher when compared with CON and
COOL (HEAT: − 0.42 ± 0.24°C, COMB: − 0.57 ± 0.19°C,
COOL: − 1.20 ± 0.37°C and CON: − 1.00 ± 0.38 °C). Our
findings are supported by previous work demonstrating a
slower decline in post-warm-up Tm following the use of
insulated garments (Faulkner etal. 2012, 2013; Raccuglia
etal. 2016). However, comparing the decline in Tm with
these studies would be inappropriate given the differences
in experimental context (i.e. post-warm-up strategy vs. half-
time intervention). In this regard, to the authors’ knowledge,
this is the first study to report Tm (albeit estimated) during
two halves of intermittent sprint performances following the
use of a half-time lower body heat maintenance strategy.
Indeed, studies investigating the use of heated garments dur-
ing half-time have utilised TC as a surrogate measurement
of Tm with the assumption that the pattern of change will
be similar (Beaven etal. 2018; Kilduff etal. 2013; Russell
etal. 2018). Although caution is warranted as our Tm meas-
ures are estimates, it allows for determining the independent
changes in Tm and TC. Moreover, the estimation error can be
considered low, given that the difference between measured
and estimated Tm during exercise and post-exercise recovery
was reported to range between 0.06 and 0.22°C (Flouris
etal. 2015). As such, the findings of this study demonstrate
to a certain extent that the use of the blizzard blanket as a
half-time strategy was effective in the maintenance of Tm
Table 1 Changes in heart rate
(HR), ratings of perceived
exertion (RPE) and thermal
sensation (TS) during
intermittent cycling in the heat
All within- and between-condition differences were analysed in raw units, with the smallest worthwhile
change (0.2 × between-subject SD) determined from first-half. Probability of an increase or decrease
associated with within group differences are denoted by A (25–75%, possibly), B (> 75–95%, likely), C
(> 95–99%, very likely) and D (> 99%, almost certainly). All between-group differences are indicated in
parentheses by symbols specific to experimental condition (i.e. # vs. COOL; * vs. HEAT; + vs. COMB),
followed by quantitative chances (A: 25–75%, possibly, B: > 75–95%, likely, C: > 95–99%, very likely and
D: > 99%, almost certainly)
First half-end Time (min)–second-half
Start 8 16 24 End
HR (bpm)
CON 160 ± 17 116 ± 14 137 ± 15D163 ± 17D155 ± 15D160 ± 15D
COOL 163 ± 11 111 ± 14(*B) 137 ± 9D161 ± 15D154 ± 14D163 ± 11D
HEAT 159 ± 14 121 ± 16 142 ± 18D163 ± 17D157 ± 15D159 ± 16D
COMB 164 ± 13 113 ± 21 141 ± 13D162 ± 16D159 ± 15D163 ± 15D
RPE (A.U.)
CON 17.4 ± 2.2 6.0 ± 0 12.6 ± 1.9D, (*B) 16.1 ± 2.3D16.0 ± 1.6D (#B) 17.6 ± 1.7D
COOL 16.8 ± 2.0 6.0 ± 0 12.4 ± 3.0D14.9 ± 2.7D, (*B) 14.9 ± 2.0D, (*B) 17.2 ± 2.1D
HEAT 16.9 ± 2.5 6.0 ± 0 13.7 ± 1.9D16.2 ± 2.1D16.1 ± 2.0D17.2 ± 2.1D
COMB 17.0 ± 2.2 6.0 ± 0 12.8 ± 1.7D15.6 ± 2.4D15.2 ± 1.6D17.0 ± 2.2D
TS (A.U.)
CON 6.7 ± 1.1 5.2 ± 0.8(+C, #C) 5.7 ± 1.0B (#B) 6.2 ± 1.0D6.3 ± 1.0D6.4 ± 1.1C
COOL 6.2 ± 1.1 3.6 ± 1.4(*C) 4.9 ± 1.5(*B) 5.7 ± 1.3D5.9 ± 1.3D6.2 ± 1.4D
HEAT 6.2 ± 1.1 4.9 ± 0.9(+C) 5.9 ± 0.9D (+B) 6.2 ± 1.1D6.2 ± 1.1D6.2 ± 1.1D
COMB 6.3 ± 1.1 3.4 ± 1.3 5.1 ± 1.1D5.9 ± 1.3D6.2 ± 1.1D6.2 ± 1.1D

European Journal of Applied Physiology
1 3
independent of changes in Tre (Fig.5a, d). For instance,
it was evident that there was a substantial decrease in Tre
(COMB: − 0.54 ± 0.13°C vs. CON: − 0.34 ± 0.31°C), and
an attenuated decline in Tes-Mus (COMB: − 0.39 ± 0.18°C
vs. CON: − 0.85 ± 0.25°C) in COMB at the end of the
half-time interval. In addition, decline in Tes-Mus in HEAT
following half-time was substantially attenuated when
compared with CON (HEAT: − 0.42 ± 0.24°C vs. CON:
− 1.00 ± 0.38°C) while Tre responses were generally simi-
lar (HEAT: 38.3 ± 0.51°C vs. CON: 38.2 ± 0.43°C). Taken
together, it is likely that the use of a lower body heat main-
tenance garment in both conditions (HEAT and COMB) did
not influence Tre responses during the half-time interval.
The mixed-method cooling consisting of ice slurry
ingestion, hand cooling and the use of ice vest was effec-
tive in reducing Tre (Fig.5a). Due to the limited time
available during half-time, such combination of modali-
ties offers the appropriate stimulus to sufficiently reduce
TC, and yet at the same time practical to implement (Duf-
field etal. 2009). From the results, COMB and COOL
resulted in the largest decrease in Tre following half-time
(Fig.5a). A decrease of ~ 0.5°C in Tre was observed after
the half-time interval in both of these conditions (COMB
and COOL). Comparatively, Duffield etal. (2009) reported
a decrease of ~ 0.3°C following 20min of mixed-method
precooling. Furthermore, Minett etal. (2012) reported that
a longer precooling duration (20min vs. 10min) resulted
in a greater reduction in TC (d = 1.32). It is likely that the
combination of both internal and external cooling methods
applied here resulted in a greater cooling effect as com-
pared to the previous studies where only external cooling
methods were utilised (i.e. cooling vest, cold towels and
ice packs).
Although the decrease in Tes-Mus during half-time was
successfully minimised in HEAT, the expected improvement
in performance was not evident. Indeed, our results dem-
onstrate that the use of a lower body heat maintenance gar-
ment during half-time was detrimental to prolonged inter-
mittent sprint performance in the heat, particularly at the
latter portions of the second-half. The current findings are
in contrast to previous studies (Kilduff etal. 2013; Russell
etal. 2018), which have consistently demonstrated improved
sprinting performance following passive heating during half-
time, particularly during the initial portion of a 6 × 40m RS
protocol. These discrepant findings are likely explained by
differences in exercise duration and environmental condi-
tions. Indeed, in the aforementioned studies, sprints were
conducted in thermoneutral environments (~ 19–20°C),
and the sprinting protocol was substantially shorter. In the
current study, a 65–66 min intermittent exercise protocol
was utilised to better typify the physical demands associ-
ated with team-sport. Moreover, given the increasing preva-
lence of major tournaments being held in hot environmental
conditions (e.g. 2020 Tokyo Olympics and 2022 FIFA World
Cup in Qatar), the use of half-time heat maintenance strat-
egies needed further substantiation within such context.
Accordingly, the current findings do not advocate the use of
heat maintenance garments as a half-time intervention. The
observed decrease in performance is likely due to the attain-
ment of an elevated Tre (~ 38.6°C) and TB (~ 37.8°C), which
could have precipitated hyperthermia-induced fatigue, and
in turn negated any contractile benefits conferred by main-
taining a higher Tm (Drust etal. 2005). Specifically, while
elevations in Tm have been shown to improve maximal power
out during sprinting (Faulkner etal. 2012; Gray etal. 2006),
muscle contractile force and voluntary activation have been
shown to progressively decrease with increasing TC, inde-
pendent of changes in Tm, in passively heated participants
(Thomas etal. 2006).
In the current study, the authors investigated combining
cooling and heat maintenance strategies, targeted at mini-
mising thermoregulatory strain whilst retaining muscle con-
tractile benefits, respectively. In this respect, it was initially
suggested that the use of passive heating alone might exacer-
bate the increase in thermal load, leading to an earlier onset
of hyperthermia-induced fatigue. However, despite that
Tre and Tes-Mus were appropriately manipulated in COMB,
sprinting performances were not improved compared with
HEAT, and instead were substantially decreased compared
with CON and COOL (Figs.2, 3, 4). Our findings are in
contrast with recent findings by Beaven etal. (2018), who
demonstrated improved RS performance (5 × 40m) follow-
ing combined precooling and lower limb passive heating.
Specifically, sprint performances following the combination
treatment (precooling and passive heating) were improved
in sprints 1–3 compared with precooling and control, and
better maintained in sprints 4–5 compared with passive heat-
ing alone and control. While differences in experimental
design, exercise duration and environmental temperatures
between the studies may largely account for the disparity in
results, the current study demonstrates that relatively high
Tm may limit prolonged exercise performance in the heat,
despite some alleviations in Tre and TSK. The reasons under-
pinning our findings with COMB are unclear. It is possible
that the thermoregulatory, perceptive and CNS-mediated
benefits extended by cooling the skin and core may have
been negated by the maintenance of higher TB and Tes-Mus.
Indeed, high TSK and Tm have been suggested to provide an
inhibitory feedback to the central motor command, down-
regulating exercise intensity (Kayser 2003; Tucker etal.
2004; Ulmer 1996). Furthermore, high TSK may have medi-
ated for an anticipatory reduction in central recruitment and
power output (Tucker etal. 2004).
The COOL intervention seemed to be the most effective
in attenuating the decline in sprinting performance dur-
ing the second-half (Figs.2, 3, 4). Our finding are in line

European Journal of Applied Physiology
1 3
with the growing consensus (Castle etal. 2006; Duffield
etal. 2009; Skein etal. 2012) that prolonged intermittent
exercise performance may be maintained and/or improved
following prior cooling. However, the majority of studies
to date have investigated the effects of half-time cooling in
combination with pre-exercise cooling (Aldous etal. 2018;
Brade etal. 2014; Minett etal. 2011, 2012), making direct
comparisons difficult. We are only aware of two studies
that have specifically investigated the effects of half-time
cooling per se (Hornery etal. 2005; Maroni etal. 2018)
on second-half exercise performance. In both these stud-
ies, exercise performances were not enhanced following
half-time cooling procedures. As such, further research is
needed to elucidate the performance benefits associated
with half-time cooling interventions.
The improved exercise performance evident follow-
ing COOL is likely explained by the lower body tem-
peratures observed at various time-points throughout the
second-half (Fig.5a–c), as well as improved RPE and TS
(Table1). While enhanced heat storage capacity has been
traditionally regarded as the primary mechanism underpin-
ning improved exercise performance following precool-
ing (Marino 2002), there is emerging evidence implicat-
ing psychophysiological mechanisms as well (Choo etal.
2017; Stevens etal. 2018). For instance, it has been pur-
ported that ice ingestion can extend the voluntary drive
to exercise by influencing afferent feedback regarding
thermal state, independent of changes in body tempera-
tures (Siegel etal. 2011). In addition, changes in TSK have
been suggested to provide afferent feedback, which is con-
sidered an important regulator of exercise intensity and
thermal comfort during self-paced exercise (Flouris and
Schlader 2015; Sawka etal. 2012; Schlader etal. 2011;
Tatterson etal. 2000). Specifically, Schlader etal. (2011)
demonstrated that a lower pre-exercise TSK and TS resulted
in higher self-selected intensity (power output) at exercise
onset, which consequently led to an overall increase in
the amount of work completed. Despite the inherent dif-
ferences between continuous self-paced and intermittent
exercise protocols, it is likely that improved thermal per-
ception, in concert with the reduced thermal strain follow-
ing COLD modulated for an improved exercise intensity
(power output) during the second-half.
Conclusion
In summary, the results of the present study show that
lower body passive heating alone or in combination with
mixed modality cooling (i.e. ice slushy ingestion, hand
cooling and ice vest) were effective half-time strategies to
attenuate the decline in Tes-Mus. However, lower body heat
maintenance even when used concurrently with cooling
was detrimental to prolonged intermittent sprint perfor-
mance in the heat, particularly at the latter stages of the
exercise protocol. The combination of upper body cool-
ing modalities utilised in this study led to a substantial
decrease in Tre,TSK, and TS. Although the cooling effects
on Tre were slightly diminished following the onset of the
second-half, changes in TSK and by extension TB were
persistent, resulting in improved TS and RPE during the
earlier and latter segments of the second-half, respec-
tively. This likely contributed to better maintenance of
exercise performance observed during the latter stages of
the second-half.
Author contributions MI, GT, AR and SA conceived and designed the
research. MI, GT, JP and JS conducted the study. MI and JS analysed
the data and wrote the manuscript. All the authors read and approved
the manuscript.
Funding No sources of funding were acquired for this study.
Compliance with ethical standards:
Conflict of interests The authors have no conflict of interests.
References
Aldous JWF, Chrismas BCR, Akubat I, Stringer CA, Abt G, Taylor
L (2018) Mixed-methods pre-match cooling improves simulated
soccer performance in the heat. Eur J Sport Sci 19:1–10
Allen PB, Salyer SW, Dubick MA, Holcomb JB, Blackbourne LH
(2010) Preventing hypothermia: comparison of current devices
used by the US Army in an invitro warmed fluid model. J Trauma
Acute Care Surg 69:S154–S161
Beaven CM, Kilduff LP, Cook CJ (2018) Lower-limb passive heat
maintenance combined with pre-cooling improves repeated sprint
ability. Front Physiol 9:1064
Bishop D, Maxwell NS (2009) Effects of active warm up on thermoreg-
ulation and intermittent-sprint performance in hot conditions. J
Sci Med Sport 12:196–204
Borg GA (1982) Psychophysical bases of perceived exertion. Med sci
sports exerc 14:377–381
Brade C, Dawson B, Wallman K (2014) Effects of different precooling
techniques on repeat sprint ability in team sport athletes. Eur J
Sport Sci 14:S84–S91
Burton AC (1935) Human calorimetry: II The average temperature of
the tissues of the body: three figures. J Nutr 9:261–280
Castle PC, Macdonald AL, Philp A, Webborn A, Watt PW, Maxwell
NS (2006) Precooling leg muscle improves intermittent sprint
exercise performance in hot, humid conditions. J Appl Physiol
100:1377–1384
Choo HC, Nosaka K, Peiffer JJ, Ihsan M, Abbiss CR (2017) Ergogenic
effects of precooling with cold water immersion and ice ingestion:
a meta-analysis. Eur J Sport Sci 18:1–12
Drust B, Rasmussen P, Mohr M, Nielsen B, Nybo L (2005) Elevations
in core and muscle temperature impairs repeated sprint perfor-
mance. Acta Physiol 183:181–190

European Journal of Applied Physiology
1 3
Duffield R, Steinbacher G, Fairchild TJ (2009) The use of mixed-
method, part-body pre-cooling procedures for team-sport athletes
training in the heat. J Strength Condition Res 23:2524–2532
Edholm P, Krustrup P, Randers M (2015) Half-time re-warm up
increases performance capacity in male elite soccer players. Scand
J Med Sci Sports 25:e40–e49
Faulkner SH, Ferguson RA, Gerrett N, Hupperets M, Hodder SG,
Havenith G (2012) Reducing muscle temperature drop post warm-
up improves sprint cycling performance
Faulkner SH, Ferguson RA, Hodder SG, Havenith G (2013) External
muscle heating during warm-up does not provide added perfor-
mance benefit above external heating in the recovery period alone.
Eur J Appl Physiol 113:2713–2721
Flouris A, Schlader Z (2015) Human behavioral thermoregulation dur-
ing exercise in the heat. Scand J Med Sci Sports 25:52–64
Flouris AD, Webb P, Kenny GP (2015) Noninvasive assessment of
muscle temperature during rest, exercise, and postexercise recov-
ery in different environments. J Appl Physiol 118:1310–1320
Girard O, Mendez-Villanueva A, Bishop D (2011) Repeated-sprint
ability—Part I. Sports Med 41:673–694
González-Alonso J, Teller C, Andersen SL, Jensen FB, Hyldig T,
Nielsen B (1999) Influence of body temperature on the devel-
opment of fatigue during prolonged exercise in the heat. J Appl
Physiol 86:1032–1039
Gray SR, De Vito G, Nimmo MA, Farina D, Ferguson RA (2006) Skel-
etal muscle ATP turnover and muscle fiber conduction velocity
are elevated at higher muscle temperatures during maximal power
output development in humans. Am J Physiol Regul Integr Comp
Physiol 290:376–382
Hopkins WG (2006) Spreadsheets for analysis of controlled trials, with
adjustment for a subject characteristic. Sport Sci 10:46–51
Hopkins W, Marshall S, Batterham A, Hanin J (2009) Progressive sta-
tistics for studies in sports medicine and exercise science. Med
Sci Sports Exerc 41:3
Hornery D, Papalia S, Mujika I, Hahn A (2005) Physiological and per-
formance benefits of halftime cooling. J Sci Med Sport 8:15–25
Ihsan M, Landers G, Brearley M, Peeling P (2010) Beneficial effects of
ice ingestion as a precooling strategy on 40-km cycling time-trial
performance. Int J Sports Physiol Perform 5:140–151
Kayser B (2003) Exercise starts and ends in the brain. Eur J Appl
Physiol 90:411–419
Kilduff LP, West DJ, Williams N, Cook CJ (2013) The influence of pas-
sive heat maintenance on lower body power output and repeated
sprint performance in professional rugby league players. J Sci
Med Sport 16:482–486
Marino F (2002) Methods, advantages, and limitations of body cooling
for exercise performance. Br J Sports Med 36:89–94
Maroni T, Dawson B, Dennis M, Naylor L, Brade C, Wallman K (2018)
Effects of half-time cooling using a cooling glove and jacket on
manual dexterity and repeated-sprint performance in heat. J Sports
Sci Med 17:485
Minett GM, Duffield R, Marino FE, Portus M (2011) Volume-depend-
ent response of precooling for intermittent-sprint exercise in the
heat. Med Sci Sports Exerc 43:1760–1769
Minett GM, Duffield R, Marino FE, Portus M (2012) Duration-depend-
ant response of mixed-method pre-cooling for intermittent-sprint
exercise in the heat. Eur J Appl Physiol 112:3655–3666
Mohr M, Krustrup P, Nybo L, Nielsen JJ, Bangsbo J (2004) Muscle
temperature and sprint performance during soccer matches–ben-
eficial effect of re-warm-up at half-time. Scand J Med Sci Sports
14:156–162
Nybo L (2008) Hyperthermia and fatigue. J Appl Physiol 104:871–878
Périard JD, Cramer MN, Chapman PG, Caillaud C, Thompson MW
(2011) Cardiovascular strain impairs prolonged self-paced exer-
cise in the heat. Exp Physiol 96:134–144
Raccuglia M, Lloyd A, Filingeri D, Faulkner SH, Hodder S, Havenith
G (2016) Post-warm-up muscle temperature maintenance: blood
flow contribution and external heating optimisation. Eur J Appl
Physiol 116:395–404
Racinais S, Wilson MG, Périard JD (2016) Passive heat acclimation
improves skeletal muscle contractility in humans. Am J Physiol
Regul Integr Comp Physiol 312:R101–R107
Ramanathan N (1964) A new weighting system for mean surface tem-
perature of the human body. J Appl Physiol 19:531–533
Russell M, West DJ, Briggs M, Bracken R, Cook C, Giroud T (2015)
A passive heat maintenance strategy implemented during a simu-
lated half-time improves lower body power output and repeated
sprint ability in professional rugby union players. PLoS ONE
10:e0119374
Russell M, West DJ, Harper LD, Cook CJ, Kilduff LP (2015) Half-
time strategies to enhance second-half performance in team-sports
players: a review and recommendations. Sports Med 45:353–364
Russell M, Tucker R, Cook CJ, Giroud T, Kilduff LP (2018) A compar-
ison of different heat maintenance methods implemented during a
simulated half-time period in professional Rugby Union players.
J Sci Med sport 21:327–332
Sargeant AJ (1987) Effect of muscle temperature on leg extension
force and short-term power output in humans. Eur J Appl Physiol
56:693–698
Sawka MN, Cheuvront SN, Kenefick RW (2012) High skin tempera-
ture and hypohydration impair aerobic performance. Exp Physiol
97:327–332
Schlader ZJ, Simmons SE, Stannard SR, Mündel T (2011) Skin tem-
perature as a thermal controller of exercise intensity. Eur J Appl
Physiol 111:1631–1639
Schneiker KT, Bishop D, Dawson B, Hackett LP (2006) Effects of
caffeine on prolonged intermittent-sprint ability in team-sport
athletes. Med Sci Sports Exerc 38:578–585
Shaffrath JD, Adams WC (1984) Effects of airflow and work load
on cardiovascular drift and skin blood flow. J Appl Physiol
56:1411–1417
Siegel R, Mate J, Brearley MB, Watson G, Nosaka K, Laursen PB
(2010) Ice slurry ingestion increases core temperature capacity
and running time in the heat. Med Sci Sports Exerc 42:717–725
Siegel R, Maté J, Watson G, Nosaka K, Laursen PB (2011) The influ-
ence of ice slurry ingestion on maximal voluntary contraction
following exercise-induced hyperthermia. Eur J Appl Physiol
111:2517–2524
Skein M, Duffield R, Cannon J, Marino FE (2012) Self-paced intermit-
tent-sprint performance and pacing strategies following respective
pre-cooling and heating. Eur J Appl Physiol 112:253–266
Stevens CJ, Mauger AR, Hassmen P, Taylor L (2018) Endurance
performance is influenced by perceptions of pain and tempera-
ture: theory, applications and safety considerations. Sports Med
48:1–13
Tatterson AJ, Hahn AG, Martini DT, Febbraio MA (2000) Effects of
heat stress on physiological responses and exercise performance
in elite cyclists. J Sci Med Sport 3:186–193
Thomas MM, Cheung SS, Elder GC, Sleivert GG (2006) Voluntary
muscle activation is impaired by core temperature rather than
local muscle temperature. J Appl Physiol 100:1361–1369
Toner MM, Drolet LL, Pandolf KB (1986) Perceptual and physiologi-
cal responses during exercise in cool and cold water. Percept Mot
Skills 62:211–220

European Journal of Applied Physiology
1 3
Tucker R, Rauch L, Harley YX, Noakes TD (2004) Impaired exercise
performance in the heat is associated with an anticipatory reduc-
tion in skeletal muscle recruitment. Pflügers Archiv 448:422–430
Ulmer H-V (1996) Concept of an extracellular regulation of muscular
metabolic rate during heavy exercise in humans by psychophysi-
ological feedback. Experientia 52:416–420
Zimmermann M, Landers G, Wallman K, Kent G (2018) Precooling
with crushed ice: as effective as heat acclimation at improving
cycling time-trial performance in the heat. Int J Sports Physiol
Perform 13:228–234
Publisher’s Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.