Novel Precooling Strategy Enhances Time Trial Cycling in the Heat : Medicine & Science in Sports & Exercise

Journal Logo


Novel Precooling Strategy Enhances Time Trial Cycling in the Heat


Author Information
Medicine & Science in Sports & Exercise 43(1):p 123-133, January 2011. | DOI: 10.1249/MSS.0b013e3181e93210
  • Free


Preparations for the 2008 Beijing Olympic Games were dominated by concerns of coaches and athletes as to how to achieve optimal performance of sustained high-intensity exercise in a hot and humid weather (7,36). Thorough reviews of the literature show that techniques that reduce core temperature immediately before a prolonged endurance exercise carried out under high thermal stress can enhance exercise capacity and performance (24,32). Effective techniques involving external cooling include immersion in cold water (5,21), direct application of cold materials to the skin (28,33), including the use of commercially available ice jackets (9,14,19), or combinations of these strategies (33).

Recently, the ingestion of large volumes of cold water has been investigated as a precooling strategy (22). This was based on calculations that ingestion of 1 L of water at 7°C by a 70-kg subject would reduce core temperature by ∼0.5°C if negative heat load was equally distributed through the body and the specific heat of the body was assumed to be 0.85 (20). When tested, the actual reduction in core temperature was observed to be 0.61°C ± 0.13°C at its maximum point, 20-25 min after ingestion, and remained 0.31°C ± 0.13°C lower than a control trial 55 min after drinking the cold water (20). Furthermore, Lee and Shirreffs (22) reported that the consumption of 1 L of cold (10°C) fluid during exercise in mild conditions attenuated the rise in rectal temperature (Trec) during steady-state cycling compared with ingestion of equal volumes of warm (37°C) and hot (50°C) fluids. Of course, the ingestion of large volumes of fluid before or during exercise is impractical in many sports, particularly those involving a high-intensity exercise because of the high risk of causing gastrointestinal upset or discomfort. A variation of this strategy, involving the ingestion of ice slurries, offers the potential for equal dissipation of heat from a smaller volume of "beverage." On the basis of the theory of enthalpy of fusion, ice requires substantially larger heat energy to cause a phase change from a solid to a liquid state (at 0°C) compared with the energy required to increase the temperature of liquid water (26).

Although there is a sound theoretical basis for using precooling strategies, "in-the-field" application during sporting competition requires identification of an ideal protocol for each unique event. Such a protocol not only needs to achieve the best outcomes for the event, which may include other physiological/nutritional benefits in addition to cooling, but also be practical to implement within the rules, logistics, and environment of the competition. The individual cycling time trial (TT) for men on the Beijing Olympic Games program involved two laps of a ∼23-km course with prolonged hill climbing. A temperature pill was ingested by an Australian professional cyclist before the Good Luck Beijing time trial test event in August 2007. The test event was held over one lap of the Olympic course, and the core temperature revealed a rise in Trec up to 39°C over the hill climb, which persisted throughout the descent (L. A. Garvican, exercise physiologist, Australian Road Cycling Team, unpublished observations). These data support the concept that cyclists participating in this event could benefit from precooling strategies that would increase their capacity for heat storage over two laps of the course.

Established precooling practices of elite Australian cyclists (hereafter known as standard cooling practices or STD COOL) are based on the combined use of a cooling jacket with cold water immersion (33). However, the location and logistics of the Beijing Olympic Games event were likely to prevent these techniques from being used. Therefore, the aim of our work was to identify whether a new precooling strategy, incorporating an internal cooling technique, would be both practical and effective in enhancing cycling performance in the heat, when used to achieve a reduction in Trec before a cycling race. In particular, we wanted to test the effectiveness of a new strategy against the standard precooling technique in enhancing the performance of a cycling protocol simulating the Beijing individual TT event. Our hypothesis was that, in comparison with a no-intervention trial, precooling techniques would reduce Trec before the start of exercise and enhance subsequent prolonged endurance performance. As well as addressing the immediate needs of Australian cyclists who competed in the Beijing Olympic Games, the results of this study can provide a model for approaching thermoregulatory challenges in other sports or competitions.


Before commencement of the pilot and main study, ethical clearance was obtained from the human research ethics committee of the Australian Institute of Sport. All subjects were informed of the nature and risks of each study before providing written informed consent.

Pilot Testing to Identify Useful Precooling Strategies

The cooling strategies used in this study were selected after an extensive pilot work, examining a variety of popular and novel approaches to precooling. Pilot work involved four subjects dressed only in cycling knicks and who completed eight different cooling strategies as well as a control condition [no cooling (CON)] in a counterbalanced experimental design. Trials were performed in a standardized hot environment [32°C-35°C and 50%-60% relative humidity (RH) as measured by a Kestrel 4000 Pocket Weather Tracker (Nielsen Kellerman, Boothwyn, PA)] and were conducted during a 90-min period that included 30 min of seated rest, 30 min of exposure to the experimental cooling strategy, and, finally, a 30-min structured warm-up exercise on a cycle ergometer (Lode Excalibur Sport, Groningen, The Netherlands). This warm-up protocol was similar to that used by elite cyclists before the individual TT and standardized for each individual (described below).

The pilot precooling strategies tested were as follows: 1) cooling jacket (33)-participants wore a waist-length cooling jacket with long sleeves and a hood, constructed with a polyester blend outer shell containing phase change material (RMIT University, Melbourne, Australia) directly applied to the skin for 30 min; 2) arctic heat vest (36)-participants wore a rechargeable, waist-length cooling vest constructed with a Sportwool outer shell containing a viscose gel (Arctic Heat Products, Burleigh Heads, Australia) directly applied to the skin, recharged every ∼5 min, for 30 min; 3) iced towels (28)-participants wore bathroom towels dunked in icy water and wrung to extract liquid (three towels were constantly rotated to cover the skin of the torso and legs for 30 min); 4) large volume of cold fluid (22,23)-participants ingested 1 L of cold (4°C) sports drink (Gatorade; Pepsico Australia, NSW, Australia) in two boluses (at t = 30 and 45 min) and were given 15 min to consume each bolus; 5) small "slushie"-participants ingested 500 g of an ice slurry made from sports drink (Gatorade; Pepsico Australia) using a commercial machine (Essential Slush Company, Burleigh Heads, Australia) and were given 30 min to consume the bolus with the aid of a straw and spoon to maximize the ingestion of ice; 6) large slushie-participants ingested 1 kg of a sports drink slushie in two boluses (at t = 30 and 45 min) and were given 15 min to consume each bolus; 7) plunge (5,21)-participants were required to complete whole-body immersion in cold (10°C) water to the level of the mesosternale in a 1.6-m-long × 0.6-m-wide × 0.8-m-high inflatable pool (Portacovery, Canberra, Australia) for 10 min followed by 20 min of seated rest (temperature was maintained by the addition of ice and water); and 8) the combination of a large slushie and iced towels.

Figure 1 summarizes Trec changes during the observation protocol relative to Trec at t = 30 min (end of the stabilization phase), with Figure 1A summarizing the trials involving internal and combination precooling techniques and Figure 1B showing results for trials involving external techniques. Changes in TrecTrec) at the end of the precooling phase (t = 60 min) were used to reflect the effectiveness of the various cooling treatments. We were also interested to note the ΔTrec at the end of the warm-up phase (t = 90 min), which represented the potential differential for heat storage at the commencement of a subsequent performance effort. On the basis of the observations on cooling achieved by various precooling methods, which are presented in the literature (23,31), we categorized ΔTrec as either small (<0.3°C), moderate (0.3°C-0.6°C), large (0.6°C-0.8°C), or very large (>0.8°C). These results are summarized in Table 1 and demonstrate that the most successful cooling strategies were the plunge and the intake of a large slushie with and without the simultaneous application of iced towels.

Pilot study's relative ΔT rec. Internal and combination precooling (A) and external precooling (B) during 30 min of heat stabilization phase, 30 min of precooling, and 30 min of warm-up. Significant time effects from t = 30 min for are denoted by dark symbols. Significant treatment effects between large cold volume are denoted by alpha (α), large slushie are denoted by beta (β), large slushie + iced towels are denoted by chi (χ), icedtowels are denoted by delta (δ), arctic heat are denoted by epsilon (ε), and RMIT/AIS Jacket are denoted by phi (φ) when compared with CON.Commencement of the precooling phase is indicated with an arrow (↓). Statistical significance is set at the level of P < 0.05. Data are presented asmean ± SD.
Relative ΔT rec in response before cooling followed by a warm-up.

The results of this pilot study identified iced towels and ingestion of a large slushie as a novel, practical, and effective approach to precooling. The aim of the main study was then to investigate the effects of this new precooling strategy in comparison with the standard technique on thermoregulation and performance of an endurance cycling task in the heat.

Main Study

Study overview.

In a randomized crossover design, participants performed three experimental trials consisting of no intervention (CON), a standard cooling practice combining a cold water plunge followed by wearing a cooling jacket (STD COOL), and a new strategy combining the application of iced towels while ingesting a large slushie (NEW COOL). These experimental methods have been previously described (above) and were applied for a total for 30 min while dressed only in cycling knicks. All subjects were familiarized with the cycling protocol, and trials were separated by 3-7 d with a consistent recovery time between trials for each subject. The outcome variables were Trec, HR, thermal comfort (38), gastrointestinal comfort (five-point Likert scale), RPE (6), cycling performance (time and power outputs), and blood lactate concentration.


A total of 12 well-trained male A-grade cyclists aged 18-35 yr were recruited from the local cycling community. All cyclists had no previous history of heat intolerance and were without injury or illness. One subject withdrew from the study after the acclimation phase and thus was removed from the analysis. Characteristics of the subjects were as follows (mean ± SD): age = 33.0 ± 5.1 yr, mass = 72.1 ± 5.5 kg, maximum aerobic power (MAP) = 449 ± 26 W, peak oxygen uptake (V˙O2peak) = 71.6 ± 6.1 mL·kg−1·min−1.

Preparation for trials.

Before commencing the experimental phase, subjects visited the laboratory on at least nine occasions to acclimatize heat and familiarize with the ergometer (Velotron; Racermate, Inc., Seattle, WA) and the experimental exercise protocol [simulated Beijing Olympic TT course on the basis of road altitude [Polar 725S HR monitor with altimeter (barometric pressure)] and distance (speed pick-up)] data collected at the Good Luck Beijing time trial test event (L. A. Garvican, unpublished observations)]. Heat acclimation was completed during a 3-wk period and consisted of nine cycling sessions of at least 60 min in duration at a self-selected intensity. All acclimation sessions were conducted in a heat chamber set at the experimental climatic conditions (32°C-35°C at 50%-60% RH). All subjects completed at least one familiarization trial of the experimental cycling protocol in the heat chamber. Before the first experimental trial, subjects also performed a progressive maximal exercise test on a cycle ergometer (Lode Excalibur Sport) to determine peak oxygen consumption (V˙O2peak) and MAP output. After a 5-min warm-up at 150 W, the test protocol started at 175 W and increased 25 W every 60 s until volitional exhaustion. MAP was determined as the power output reached in the last completed stage. If the participant finished partway through a 60-s stage, MAP was calculated in a pro rata manner. Subjects' expired air was collected into a customized Douglas bag gas analysis system, which incorporated an automated piston that allowed the volume of air displaced to be quantified, with O2 and CO2 analyzers (AEI Technologies, Pittsburgh, PA). The operation and calibration details of this equipment have been described previously (34).

Subjects followed a standardized diet and training protocol for up to 24 h before each experimental trial. Specifically, they were allowed to undertake a light exercise bout on the day before each trial (repeated for subsequent trials). In addition, they were required to consume a standardized diet, supplied in the form of prepackaged meals and snacks, providing 9 g of CHO per kilogram body mass (BM), 1.5 g of protein per kilogram BM, and 1.5 g of fat per kilogram BM, with a total energy goal of 230 kJ per kilogram BM. Subjects refrained from any intake of caffeine and alcohol during this period. Compliance to the diet and exercise protocol was determined from a checklist kept by each subject and presented on arrival to the laboratory before each trial.

Experimental trials.

All testing was carried out in the afternoon to mimic the schedule of the Beijing cycling TT. Approximately 2.5 h before a trial (−150 min before start of TT), subjects reported to the laboratory, having just eaten their prerace meal from the packaged diet (providing 2 g of CHO per kilogram BM). Subjects brought with them a "first waking" urine sample to determine specific gravity, which was used to ensure that cyclists attended the laboratory in a similar euhydrated state for each trial. At this time, their food and training diaries were checked for compliance to the standardization protocols.

Subjects then voided their bladders, inserted a single-use thermal probe (Mon-a-therm General Purpose Temperature Probe; Mallinckrodt Medical, Inc., St. Louis, MO) 12 cm beyond the anal sphincter, and fitted a chest strap for a Polar S810i HR monitor (Polar Electro Oy, Kempele, Finland). Rating of thermal comfort, BM, Trec, and HR were recorded before entering the heat chamber (−120 min before start of TT). Cyclists were able to drink chilled (4°C) water ad libitum throughout the period of stabilization to heat (−120 to −60 min before TT) and throughout the warm-up period (−30 to 0 min before TT), with the volume chosen on their first trial being recorded and repeated in subsequent trials. Subjects were also allowed to void their bladders when required during the 2-h period before the TT, with BM being recorded before and after each toilet break. During this stabilization to the hot environment, subjects were required to check the configuration of the bike ergometer at a set time; otherwise, subjects remained seated throughout this period.

The environmental conditions (temperature and RH) inside the chamber were measured every 10 min throughout the duration of the trial. During the heat stabilization period (−120 to −60 min before TT), the subject's HR, Trec, and thermal comfort were recorded every 5 min. At the commencement of the precooling phase (−60 min before TT), subjects were exposed to one of the three experimental treatments lasting 30 min, which were as follows: 1) CON-no treatment apart from the ad libitum consumption of cold water (4°C); 2) STD COOL-a combination protocol involving whole-body immersion in cold (10°C) water to the level of the mesosternale for 10 min followed by wearing a cooling jacket for 20 min; or 3) NEW COOL-a combination protocol where subjects consumed a total of 14 g of sports drink slushie per kilogram BM given in two 7-g boluses per kilogram BM and were given 15 min to consume each bolus while wearing iced towels as previously described. During this time and in addition to the measurements previously outlined, subjects were asked to provide ratings of the effectiveness of cooling and stomach fullness.

Subjects then completed a standardized 20-min warm-up on the Velotron ergometer fitted with an SRM cycling power meter (Scientific Version, eight-strain gauge; Schoberer Rad Meßtechnik, Jülich, Germany) sampling power output (W) at 1-s intervals, which was calibrated before the commencement of the first TT using a custom-built calibration rig previously described (13). This warm-up consisted of repetition of a protocol of 3 min at 25% MAP, 5 min at 60% MAP, 2 min at 80% MAP, and based on the actual protocols used by elite Australian TT cyclists in similar conditions. The final 10 min before the start of the TT allowed subjects to complete their own preparations, including toilet breaks and bicycle adjustments. During this time, subjects were provided with standard race instructions for each TT protocol, and the zero offset of the SRM crank was set according to the manufacturer's instructions. Drinks for subsequent intake during the TT were removed from ice storage and left in the heat chamber to simulate drink temperatures that would be experienced in a race situation with high ambient temperatures.

Subjects completed the cycling TT according to instructions, with HR and Trec being monitored continuously and manually recorded every 2 min, whereas self-reports of thermal sensation and stomach comfort were recorded at approximately 10-min intervals. A capillary blood sample was collected via a finger prick to measure blood lactate concentration (Lactate Pro; Arkray KDK Corp., Kyoto, Japan) at the TT start, at the "top of the climb" (12.5 and 35.7 km), half way, and at TT completion (23.2 and 46.4 km).

The ergometer was placed in front of a large television screen that displayed the course profile on the accompanying computer software (Velotron 3D Software; RacerMate, Inc., Seattle, WA). The feedback provided to the subject was limited to distance covered (km), cycling gear ratio (12-27/48-54), road gradient (%), and instantaneous velocity (km·h−1). There were two visual displays of the course: a topographical profile of the 46.4-km course, featuring an arrow to indicate the current course position of the cyclists, and a display of a road from the perspective of a rider. Subjects were provided with 350 mL of a 6% CHO-electrolyte drink (Gatorade; Pepsico Australia) at the "top of each climb" (12.5 and 37.5 km), which simulated the ideal time to consume fluid on the Beijing TT course, on the basis of the experience of professional cyclists during the Beijing test event. They were permitted to drink ad libitum for the next kilometer on the first trial. The volume that was consumed was measured and repeated for subsequent trials.

As a further strategy to mimic riding on a hilly course, the cyclist was positioned in front of a large industrial fan (750 mm, 240 V, 50 Hz, and 380 W; model N11736; Trade Quip, Auckland, New Zealand). The speed of the fan was altered to simulate uphill or downhill wind speeds: specifically, the fan was fixed on low speed (1130 ± 5 rpm) for 0-12.5 and 23.2-35.7 km and was switched to high speed (1330 ± 5 rpm) for 12.5-23.2 and 35.7-46.4 km.

Split times, velocity, and power output data were collected for each trial, with the periods of interest being time to top of first climb (12.5 km), end of first lap (23.2 km), time to top of second climb (35.7 km), and finish (46.4 km). On the completion of each TT, subjects were asked a series of questions related to their effort ("How much of yourself did you give?"), using a modified Borg scale where effort ≤100% is reported, motivation ("How motivated were you to race today?"), sensation ("How did you feel during the time trial?"), and comfort ("How comfortable did you feel during the time trial?") presented as five-point Likert scales. This series of questions is a monitoring tool used routinely within the Australian road cycling team.

Statistical Analysis

Dependent variables including BM, percent dehydration, and postrace subjective ratings were analyzed for significant effects using ANOVA. A two-way (treatment × time) repeated-measures ANOVA was used to determine significant differences in dependent variables (rectal temperature, HR, blood lactate, thermal comfort, and stomach fullness) between treatment means at each time point. Pairwise comparisons were conducted to determine where the differences existed, using a Newman-Keuls post hoc test. These statistical tests were conducted using Statistica for Microsoft Windows (version 8; StatSoft, Tulsa, OK), and the data are presented as means and SD. For analysis, significance was accepted at P < 0.05.

The performance data from the three trials were analyzed using the magnitude-based inference approach recommended for studies in sports medicine and exercise science (3,17). A mixed modeling procedure in the Statistical Analysis System (version 9.1; SAS Institute, Cary, NC) was used to estimate means (fixed effects) and within-subject and between-subject variations (random effects, modeled as variances). The fixed effects were 1) treatment (CON, STD COOL, and NEW COOL) to adjust for a main effect and 2) trial number (nos. 1-3) to adjust for learning and habituation. The random effects were 1) identity (subjects 1-11) to control for the different abilities of the subjects and 2) extra variance on the first trial to allow for differences in familiarization between subjects. We also investigated the extra variance of the NEW COOL treatment to account for the individual differences to this treatment; however, the variance estimated was −0.5 (negative), and hence, this random effect was eliminated from the model. Performance data are represented by TT time and power output during the various segments of the course and are reported as means ± SD. The magnitude of the change in time was interpreted by using values of 0.3, 0.9, 1.6, 2.5, and 4.0 of the within-athlete variation (coefficient of variation) as thresholds for small, moderate, large, very large, and extremely large differences in the change score between the trials (17). The typical variation (coefficient of variation) for road cycling TT in top athletes has been previously established as 1.3% by Paton and Hopkins (30), with the smallest worthwhile change in performance time established at 0.4% (16). Finally, these data are presented with inference about the true value of a precooling treatment effect on simulated cycling TT performance. The practical interpretation of an effect is deemed "unclear" when the magnitude of change is substantial when the confidence interval (precision of estimation) could result in positive and negative outcomes (4,18). Data for the incomplete trial were calculated by comparing the performance decrement of the athlete's performance (time and power data) from trial 1 with the athlete's best performance (trial 3).


Monitoring of the subjects' compliance to pretrial standardization requirements showed that all subjects consumed the standard diet as requested before each of their trials and refrained from strenuous training in the 24 h before the commencement of the TT. Collection of "first waking" urine samples on the morning of each trial showed that subjects commenced each trial with similar within-subject and between-subject hydration status (data not shown). The mean changes in BM from entrance to the heat chamber to the completion of the TT were −2.43 ± 0.73, −2.30 ± 0.63, and −1.67 ± 0.51 kg for CON, STD COOL, and NEW COOL, representing mean losses of 3.3%, 3.1%, and 2.3% BM (CON, STD COOL > NEW COOL, P < 0.05). The volume of the sports drink consumed during the TT was 630 ± 70 g for all treatments, which provided a CHO intake of ∼38 g (∼0.5 g·kg−1 BM).

Monitoring of the heat chamber showed that all trials were carried out in similar conditions (mean conditions across the ∼3.5 h of heat exposure in each trial were 34.1°C ± 0.3°C and 52.8% ± 3.5% RH for CON, 33.8°C ± 0.5°C and 53.4% ± 4.1% RH for STD COOL, and 33.9°C ± 0.5°C and 53.4% ± 3.6% RH for NEW COOL, P > 0.05). Ten cyclists completed the three trials according to our protocol. One subject was instructed to cease cycling in one trial at the 38.8-km point of the 46.4-km course because of ethical obligations (Trec exceed 41°C and his self-reported thermal sensation was 7 = "very hot"). This situation occurred during his first trial (STD COOL). This subject was able to complete the two other trials without the same incident.

Trec at the end of the stabilization phase (t = −120 min before TT) was considered to be the baseline value for each trial. Figure 2 shows the relative changes in Trec during each trial. There was an observable treatment effect on Trec during the period before the TT, with Trec being lower at the completion of the cooling phase and throughout the warm-up after both precooling techniques (STD COOL < NEW COOL < CON, P < 0.05). The warm-up was associated with a rise in Trec in all trials so that it had moved above baseline values by the start of the TT. Trec continued to rise during the TT in all trials, such that there were no differences in Trec between treatments during this phase (P > 0.05).

Relative ΔT rec during precooling (A), a warm-up (B), and, a 46.4-km simulated TT (C). Significant time effects from t = −120 min before TT are denoted by dark symbols. Significant treatment effects between STD COOL and NEW COOL when compared with CON are denoted by an asterisk (*). Significant interaction effects between STD COOL and NEW COOL are denoted by a hash (#). Commencement of the precooling phase is indicated with an arrow (↓). Statistical significance is set at the level of P < 0.05. Data are presented as mean ± SD.

Figure 3 shows the changes in HR as a percentage of maximal HR (%HRmax) during each trial. Figure 3A shows no change in relative HR during the cooling phase. However, with the onset of the warm-up (Fig. 3B), there was a concomitant increase in relative HR with an observable treatment effect: NEW COOL, STD COOL < CON at −25, −20, and −10 min before TT (P < 0.05). There was a significant difference in HR between trials at −5 min before TT; however, HR was not significantly different at the TT start. HR rose quickly above baseline values at the start of the TT and continued to rise throughout. There were no differences in relative HR profiles between trials.

%HRmax during precooling (A), a warm-up (B), and a 46.4-km simulated TT (C). Significant time effects from t = −120 min before TT are denoted by dark symbols. Significant treatment effects between STD COOL and NEW COOL are denoted by an asterisk (*1and *2, respectively). Significant interaction effects between STD COOL and NEW COOL are denoted by a hash (#). Commencement of the precooling phase is indicated with an arrow (↓). Statistical significance is set at the level of P < 0.05. Data are presented as mean ± SD.

Performance information from each of the TT is presented in Table 2 and includes time (h:min:s) and power output (W) for the entire TT, for each of the laps, and for each of the four segments (climbs 1 and 2 and descents 1 and 2). The error of measurement across all trials in the current study was established as 1.7%. Overall, NEW COOL was associated with a 3.0% increase in power output (∼8 W, P = 0.04) and a 1.3% improvement in performance time (∼1:06 min, P = 0.08) compared with the CON trial, with the true likely effects ranging from a trivial to a large benefit. The effect of the STD COOL trial compared with that of the CON trial was "unclear" for power output (1.1%; ±2.4%, P = 0.43) and performance time (−0.5%; ±1.2%, P = 0.53). There was an interaction between the two experimental treatments on descent 2, with NEW COOL achieving a small to a very large benefit in power output (4.9%; likely range = 1.3%-8.5%; ∼14 W, P = 0.03) and a small to a large benefit in performance time (−1.6%; likely range = −0.9% to −2.3%; ∼12.2 s, P = 0.009) than STD COOL. There was no evidence of any extra variance (an individual response) on the NEW COOL treatment under the conditions of this study.

Summary of cycling time trial performance data (performance time and power output).

Blood lactate results were similar across all experimental conditions. There was a trend for lactate concentrations to mirror the course profile, being slightly higher at the top of each climb and lower at the end of climb 1. There was an effect of time, whereby blood lactate concentration was greater than baseline values (3.2 ± 1.3, 3.1 ± 1.4, and 3.4 ± 1.4 mmol·L−1 for CON, STD COOL, and NEW COOL, respectively, P > 0.05) at the top of climb 1 after STD COOL (5.1 ± 1.8, P < 0.05) and at the completion of the TT (5.4 ± 1.5, 5.5 ± 1.6, and 6.7 ± 1.4 for CON, STD COOL, and NEW COOL, respectively, P < 0.05).

Figure 4 shows changes in the subjects' thermal comfort (Fig. 4A) and stomach fullness (Fig. 4B) during each trial. There was no significant change in the rating of thermal comfort after athletes entered the chamber to stabilize to the hot and humid conditions for 60 min. However, once precooling commenced at t = −60 min before TT, the rating of thermal comfort was significantly reduced (from a rating of ∼4.6 to 0), such that subjects reported feeling cooler when treated with STD COOL (−55 to −27 min before TT, P < 0.05) and NEW COOL (−60 to −30 min before TT, P < 0.05) treatments compared with CON. The warm-up caused an increase in ratings of thermal comfort so that, in all treatments, subjects rated their thermal comfort as being warmer than baseline levels (P < 0.05). Once the TT had commenced, thermal comfort deteriorated such that subjects progressively felt warmer, and there were no differences detected between trials. There was no significant change in the ratings of perceived stomach fullness across the three trials.

Subjective ratings of comfort. Thermal comfort (A) and stomach fullness (B). Significant time effects from t = −120 min before TT are denoted by dark symbols. Significant treatment effects between STD COOL and NEW COOL are denoted by an asterisk (*1 and *2, respectively). Commencement of the precooling phase is indicated with an arrow (↓). Statistical significance is set at the level of P < 0.05. Data are presented as mean ± SD.

Subjective information provided by each subject at the completion of each trial suggested that there was no effect of any treatment on their ratings of effort given (98.0% ± 6.0%, 98.2% ± 3.4%, and 96.8% ± 7.5% for CON, STD COOL, and NEW COOL, respectively, P > 0.05), where 100% equals "I gave everything I had," their sensation (3.1 ± 0.8, 3.3 ± 0.6, and 3.1 ± 1.1 for CON, STD COOL, and NEW COOL, respectively, P < 0.05), where a score of 3 indicates "felt okay," motivation (4.0 ± 0.09, 3.8 ± 0.8, and 4.2 ± 0.08 for CON, STD COOL, and NEW COOL, respectively, P > 0.05), where a score of 4 indicates "motivated," and comfort (2.3 ± 0.9, 3.0 ± 0.9, and 2.6 ± 1.0 for CON, STD COOL, and NEW COOL, respectively, P > 0.05), where a score of 2 indicates "uncomfortable."


The purposes of the current study were to identify a practical cooling strategy that was effective in achieving a reducing in Trec before a simulated cycling TT in hot and humid conditions and to compare the effectiveness of this new strategy against standard precooling practices in enhancing cycling endurance performance. Our study showed that a new precooling strategy, combining the external application of iced towels with the internal consumption of ice slurry (slushie) made from sports drink, enhanced the performance of a laboratory cycling protocol simulating the Beijing Olympic Games TT in hot and humid conditions. In contrast, a precooling protocol on the basis of the previously established strategy of a cold water plunge followed by the use of a cooling jacket failed to provide clear benefits. The benefits to cycling performance achieved by the new precooling strategy were most evident in the second half of the TT, in both the "climb" and "descent" portions of the laboratory protocol. This new precooling strategy represents a practical and effective technique that could be used by cyclists in preparation for races of similar characteristics to the Beijing Olympic TT course that are undertaken in hot and humid conditions.

In the present study, both NEW COOL and STD COOL achieved a noticeable ("moderate&" and "very large," respectively) cooling effect on subjects who had been exposed to a hot and humid environment for 60 min. There were differences in the characteristics of these cooling methodologies, however, which might explain why only the NEW COOL protocol achieved a clear performance benefit over the CON trial. The timing of the peak cooling effectiveness (time of the lowest Trec) was similar in both treatments, occurring 5 min after the completion of the cooling protocol (i.e., after 5 min of active warming via the TT warm-up). However, the magnitude of cooling achieved before the TT was ∼0.6°C greater with the STD COOL trial, where we were able to replicate a similar magnitude to standard precooling practices on the basis of the work of Quod et al. (33). Unlike previous studies (22,23,25,27,35), we cannot directly compare the effects of each treatment on changes in Trec during the TT because the workload was self-paced. Indeed, the HR and Trec recorded during the TT were similar between trials, but the important finding was that, overall, a higher mean power output and faster performance were achieved with the NEW COOL treatment. This is a common finding among the literature during self-paced trials, whereby a successful intervention is associated with faster times or higher power outputs for the similar perturbations in physiological responses such as HR, RPE, and core temperature (5,10,21,33). Such findings give support to the model of teleoanticipation proposed by Ulmer (37), whereby efferent feedback during an intensive close-looped exercise task allows an individual to choose an exercise pace that will allow him/her to finish the task at a common metabolic end point.

It is of interest to speculate why the NEW COOL treatment achieved a beneficial effect on TT performance, whereas the effects of the STD COOL treatment, which produced a larger cooling response, were "unclear." The statistical treatment and our observations of subject behavior suggest that the larger cooling effect may introduce a poorer pacing strategy on the basis of "feeling better" at the start of the TT. Performance benefits were seen in climb 1 after STD COOL, with an increase in concentrations of blood lactate above resting values. In fact, by the end of the first climb, the pre-TT differences in Trec were abolished. This pacing strategy could not be sustained during descent 1, with cyclists experiencing a trivial to a large reduction in performance in STD COOL compared with that in CON. It is possible that with greater practice or education, subjects might be able to learn how to better pace efforts after whole-body cooling using a plunge so that there is better coupling between thermal sensations and potential for heat storage.

Another recently conducted study has also demonstrated the utility of the ingestion of ice on running endurance in a comparably hot environment (35). In that study, subjects consumed 7.5 g·kg−1 of an energy-containing drink presented as an icy slurry (−1°C) or as a cold beverage (4°C) immediately before undertaking a sustained high-intensity treadmill run. The slushie drink was associated with a reduced Trec and an increased running time to exhaustion (50.2 ± 8.5 vs 40.7 ± 7.2 min, P = 0.001). However, there was an increased Trec at the point of fatigue with the slushie treatment, suggesting a surprising uncoupling of Trec and the reaching of fatigue. It was speculated that the oral ingestion of the icy drink may have cooled the brain or produced greater cooling sensations that allowed additional work to be undertaken before volitional fatigue. There is the potential for negative effects from such a strategy if exercise continues beyond the "healthy" limits of the body's thermal load. Clearly, further investigations of the use of internal cooling strategies are merited.

Although the focus of these interventions was the optimization of thermoregulation via precooling, we acknowledge that the slushie consumption associated with the NEW COOL strategy involved intake of additional amounts of fluid and CHO; nutritional factors that might enhance performance per se. In fact, there were modest but significant differences in pretrial fluid intake between treatments so that the mean "fluid deficit"(change in BM) incurred from entry to the heat chamber to the completion of the TT was 2.3% in NEW COOL versus ∼3.2% in the other trials. Thus, it is possible that a small enhancement in hydration status will have contributed to better cycling performance with the NEW COOL strategy. Unfortunately, the literature on the effects of small gradations in fluid levels on cycling performance is limited to two studies. In one investigation, Dugas et al. (11) measured performance of six male cyclists during repeat 80-km TT in hot conditions in which subjects consumed different volumes of fluid to incur fluid deficits equivalent to ∼0.5%, 2%, 3%, and 4% BM. Their study failed to detect differences in performance between trials using traditional probability statistics; however, data pool ing to counter the risk of a type II error associated with low sample size showed that the higher fluid trials (0.5%-2% BM loss) showed greater power outputs and a strong trend to faster performance times particularly in the second half of the TT than the lower fluid intake trials (3%-4% BM loss). The other study conducted in temperate conditions failed to detect any differences in the performance of a 1-h cycling TT when cyclists consumed low, medium, and high volumes of fluid to incur 1275, 1025, and 538 g of BM loss, respectively, during the protocol (2). However, other studies that have investigated subtle differences in fluid deficits on physiological variables during steady cycling (25) and on basketball skills (3) have shown a progressive decline in variables, with the progressive increases in the degree of dehydration between 1% and 4% BM.

It is possible that the structured fluid regimen associated with the slushie ingestion will be better for cycling performance than the "voluntary dehydration" typically associated with ad libitum drinking patterns in real-life sporting situations and also seen in our study. This is a contentious issue (29), and there is contradictory evidence from the sparse literature comparing performance outcomes associated with ad libitum versus structured fluid intake during exercise in a hot environment. One study (8) found that a structured drinking plan involving a larger volume of fluid intake did not enhance time to fatigue after 90 min of running in the heat compared with ad libitum; in fact, several subjects incurred severe gastrointestinal discomfort associated with the higher fluid intake. Meanwhile, Dugas et al. (11) concluded from the previously discussed investigation that there were no detectable differences in performance between an ad libitum fluid intake (associated with a 2% BM loss) and trials that incurred either higher or lower fluid deficits.

The NEW COOL protocol provided subjects with an additional CHO intake (0.8 g·kg−1) compared with the other trials. Although failure to match fuel availability represents another potential explanation of performance differences between trials, we think that this is unlikely. The CHO content of the slushie is small in comparison with the other strategies on the basis of current sports nutrition guidelines (1) used to promote CHO availability in the other trials (9 g·kg−1 in the 24 h before trial, including 2 g·kg−1 in the pre-event meals, and an additional 0.5 g·kg−1 during the TT). Furthermore, the TT protocol used in our study is unlikely to be limited by CHO availability (12,15,31).

On this basis, we feel confident that the main effect of the NEW COOL treatment on cycling performance was achieved via its cooling effect. There is a small literature concerning the benefits of ingesting cold water before and/or during exercise in hot conditions to provide a heat sink. Lee et al. (23) found that the ingestion of 1500 mL of cold fluids (4°C) before and during steady-state cycling in hot and humid conditions reduced heat accumulation and increased time to exhaustion by 23% (∼64 vs ∼52 min, P < 0.001) compared with a trial in which an equal volume of warm fluids (37°C) was consumed. Similarly, the consumption of cold fluids (4°C) was more palatable and was associated with an attenuated rise in rectal temperature and HR and an 11% increase in endurance during steady-state cycling in the heat (∼62 vs ∼55 min, P < 0.05) compared with the intake of a more neutral temperature drink (19°C) (27).

Practical implications.

The combination of applying iced towels and ingesting a large slushie provides a practical solution for sport scientists to precool athletes in the field. The logistical ease of preparing iced towels in a portable icebox allows for minimal demand on equipment, storage, transport, cost, and personnel. A sports drink slushie can be prepared using a portable commercial machine and can be simplified further by freezing and part-thawing ready-to-drink commercially available beverages. For instance, when precooling with this strategy at an event with similar logistical limitations such as the Beijing Olympic Games, athletes are able to use standard hotel towels prepared in an icy slurry of water and shaved ice while ingesting a slushie. In contrast, logistical difficulties associated with using other popular cooling strategies (i.e., cold water immersion, ice jackets, and ice vests) can be limited by equipment, access to sufficient water and/or electricity, freezer space, and transport demands. Therefore, the NEW COOL strategy provides a practical and logistically simpler strategy to precool athletes in the field.


In summary, we developed a novel precooling strategy combining internal and external cooling techniques, which may also offer benefits of better fuel and fluid status over the currently self-chosen patterns of cyclists and which can be practically applied in field settings. Moreover, this study demonstrated that this technique was more effective than the precooling technique, previously established and commonly used in the sport of cycling, in enhancing the performance of a TT simulating the event included in the Beijing Olympic Games. The benefits of the new technique were most evident in the second half of the TT. Further studies should be undertaken to elucidate the mechanisms associated with the performance benefits seen with our strategy and to determine the range of sports in which this protocol might be useful.

This study was supported by a Nestle Australia funding to the Australian Institute of Sport Sports Nutrition research activities and by a grant from the Performance Research Centre, Australian Institute of Sport. The significant technical assistance of Mr. Nathan Versey and Mr. Jamie Plowman is gratefully acknowledged.

No competing financial interests, agreements, or professional relationships exist whereby a third party may benefit from the results of the present study.

The results of the present study do not constitute endorsement by the American College of Sports Medicine.


1. American College of Sports Medicine. Position Stand: nutrition and athletic performance. Med Sci Sports Exerc. 2009;41(3):709-31.
2. Backx K, van Someren KA, Palmer GS. One hour cycling performance is not affected by ingested fluid volume. Int J Sport Nutr Exerc Metab. 2003;13(3):333-42.
3. Baker LB, Dougherty KA, Chow M, Kenney WL. Progressive dehydration causes a progressive decline in basketball skill performance. Med Sci Sports Exerc. 2007;39(7):1114-23.
4. Batterham AM, Hopkins WG. Making meaningful inferences about magnitudes. Int J Sports Physiol Perform. 2006;1(1):50-7.
5. Booth J, Marino F, Ward JJ. Improved running performance in hot humid conditions following whole body precooling. Med Sci Sports Exerc. 1997;29(7):943-9.
6. Borg G. Perceived exertion as an indicator of somatic stress. Scand J Rehabil Med. 1970;2(2):92-8.
7. Castle PC, Macdonald AL, Philp A, Webborn A, Watt PW, Maxwell NS. Precooling leg muscle improves intermittent sprint exercise performance in hot, humid conditions. J Appl Physiol. 2006;100(4):1377-84.
8. Daries HN, Noakes TD, Dennis SC. Effect of fluid intake volume on 2-h running performances in a 25 degrees C environment. Med Sci Sports Exerc. 2000;32(10):1783-9.
9. Duffield R, Dawson B, Bishop D, Fitzsimons M, Lawrence S. Effect of wearing an ice cooling jacket on repeat sprint performance in warm/humid conditions. Br J Sports Med. 2003;37(2):164-9.
10. Duffield R, Green R, Castle P, Maxwell N. Precooling can prevent the reduction of self-paced exercise intensity in the heat. Med Sci Sports Exerc. 2010;42(3):577-84.
11. Dugas JP, Oosthuizen U, Tucker R, Noakes TD. Rates of fluid ingestion alter pacing but not thermoregulatory responses during prolonged exercise in hot and humid conditions with appropriate convective cooling. Eur J Appl Physiol. 2009;105(1):69-80.
12. Galloway SD, Shirreffs SM, Leiper JB, Maughan RJ. Exercise in the heat: factors limiting exercise capacity and methods for improving heat tolerance. Sports Exerc Inj. 1997;3(1):19-24.
13. Gardner AS, Stephens S, Martin DT, Lawton E, Lee H, Jenkins D. Accuracy of SRM and power tap power monitoring systems for bicycling. Med Sci Sports Exerc. 2004;36(7):1252-8.
14. Hasegawa H, Takatori T, Komura T, Yamasaki M. Wearing a cooling jacket during exercise reduces thermal strain and improves endurance exercise performance in a warm environment. J Strength Cond Res. 2005;19(1):122-8.
15. Hawley JA, Palmer GS, Noakes TD. Effects of 3 days of carbohydrate supplementation on muscle glycogen content and utilisation during a 1-h cycling performance. Eur J Appl Physiol Occup Physiol. 1997;75(5):407-12.
16. Hopkins WG. Magnitude matters: effect size in research and clinical practice. Sportscience. 2006;10:58.
17. Hopkins WG, Batterham AM, Marshall SW, Hanin J. Progressive statistics. Sportscience. 2009;13:55-70.
18. Hopkins WG, Marshall SW, Batterham AM, Hanin J. Progressive statistics for studies in sports medicine and exercise science. Med Sci Sports Exerc. 2009;41(1):3-13.
19. Hornery DJ, Papalia S, Mujika I, Hahn A. Physiological and performance benefits of halftime cooling. J Sci Med Sport. 2005;8(1):15-25.
20. Imms FJ, Lighten AD. The cooling effects of a cold drink. In: Mercer JB, editor. Thermal Physiology. New York: Elsevier Science; 1989. p. 135-9.
21. Kay D, Taaffe DR, Marino FE. Whole-body pre-cooling and heat storage during self-paced cycling performance in warm humid conditions. J Sports Sci. 1999;17(12):937-44.
22. Lee JK, Shirreffs SM. The influence of drink temperature on thermoregulatory responses during prolonged exercise in a moderate environment. J Sports Sci. 2007;25(9):975-85.
23. Lee JK, Shirreffs SM, Maughan RJ. Cold drink ingestion improves exercise endurance capacity in the heat. Med Sci Sports Exerc. 2008;40(9):1637-44.
24. Marino FE. Methods, advantages, and limitations of body cooling for exercise performance. Br J Sports Med. 2002;36(2):89-94.
25. McConell GK, Burge CM, Skinner SL, Hargreaves M. Influence of ingested fluid volume on physiological responses during prolonged exercise. Acta Physiol Scand. 1997;160(2):149-56.
26. Merrick MA, Jutte LS, Smith ME. Cold modalities with different thermodynamic properties produce different surface and intramuscular temperatures. J Athl Train. 2003;38(1):28-33.
27. Mundel T, King J, Collacott E, Jones DA. Drink temperature influences fluid intake and endurance capacity in men during exercise in a hot, dry environment. Exp Physiol. 2006;91(5):925-33.
28. Myler GR, Hahn AG, Tumilty D. The effect of preliminary skin cooling on performance of rowers in hot conditions. Excel. 1989;6(1):17-21.
29. Noakes TD. Drinking guidelines for exercise: what evidence is there that athletes should drink "as much as tolerable," "to replace the weight lost during exercise" or "ad libitum"? J Sports Sci. 2007;25(7):781-96.
30. Paton CD, Hopkins WG. Variation in performance of elite cyclists from race to race. Eur J Sport Sci. 2006;6(1):25-31.
31. Pitsiladis YP, Maughan RJ. The effects of exercise and diet manipulation on the capacity to perform prolonged exercise in the heat and in the cold in trained humans. J Physiol. 1999;517(Pt 3):919-30.
32. Quod MJ, Martin DT, Laursen PB. Cooling athletes before competition in the heat: comparison of techniques and practical considerations. Sports Med. 2006;36(8):671-82.
33. Quod MJ, Martin DT, Laursen PB, et al. Practical precooling: effect on cycling time trial performance in warm conditions. J Sports Sci. 2008;26(14):1477-87.
34. Russell G, Gore CJ, Ashenden MJ, Parisotto R, Hahn AG. Effects of prolonged low doses of recombinant human erythropoietin during submaximal and maximal exercise. Eur J Appl Physiol. 2002;86(5):442-9.
35. Siegel R, Mate J, Brearley MB, Watson G, Nosaka K, Laursen PB. Ice slurry ingestion increases core temperature capacity and running time in the heat. Med Sci Sports Exerc. 2010;42(4):717-25.
36. Uckert S, Joch W. Effects of warm-up and precooling on endurance performance in the heat. Br J Sports Med. 2007;41(6):380-4.
37. Ulmer HV. Concept of an extracellular regulation of muscular metabolic rate during heavy exercise in humans by psychophysiological feedback. Experientia. 1996;52(5):416-20.
38. 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(3):1218-23.


©2011The American College of Sports Medicine