Secondary Logo

Journal Logo

Original Research

Effect of Practical Precooling on Neuromuscular Function and 5-km Time-Trial Performance in Hot, Humid Conditions Among Well-Trained Male Runners

Randall, Carla A.1; Ross, Emma Z.1,2; Maxwell, Neil S.1

Author Information
Journal of Strength and Conditioning Research: July 2015 - Volume 29 - Issue 7 - p 1925-1936
doi: 10.1519/JSC.0000000000000840
  • Free



Athletic competition in the heat increases physiological strain and reduces performance during self-paced, endurance exercise (12,15). Exercising in hot, humid conditions disproportionately increases rate of heat storage, body temperature, and heat illness susceptibility, subsequently decreasing force production, muscle activation, and exercise intensity (23). An increase in core temperature to a critical level of 40° C and subsequent hyperthermia inhibits central nervous system function and central activation through reduced force production (15,28). Hyperthermia increases cardiovascular strain and relative metabolic strain and reduces oxygen supply (15). However, an increase in motor unit recruitment and work rate toward completion of prolonged exercise has been reported, regardless of heat storage rate and high core temperatures reached (13). Therefore, the application of precooling may counteract the increased physiological load and enhance subsequent performance.

Previous studies have established that precooling before exercise in hot conditions improves intermittent (7,35) and endurance performance (11,33,43). Precooling may minimize thermal strain and maintain muscular recruitment, essential for optimizing performance during long-distance training and competition. External precooling techniques, including ice vests or cold towels, reduce skin temperature, whereas ice packs, cold showers, cold-water immersion, or combined methods reduce skin, muscle, and core temperature (22,30,44). However, precooling is technique specific and duration dependent, consequently altering heat strain, pacing, and exercise performance (7,25).

Previous research has identified performance improvements during self-paced intermittent and continuous exercise in the heat without any alterations in end-exercise physiological perturbations (2,11,35). Therefore, the ergogenic benefits from precooling may include the prevention of hindered central motor drive because of thermal sensory feedback (30). Precooling may improve performance through greater heat storage capacity and blunting pre-exercise core temperature to enable increased muscle recruitment and to promote the selection and maintenance of higher intensity exercise (2,11). Furthermore, the selection of exercise intensity may be altered by afferent feedback from peripheral signals comprising cardiovascular and thermoregulatory strain (42). However, many studies exploring precooling on self-paced endurance performance have integrated warm-up protocols that are of insufficient quality and duration before athletic competition. At present, there is little evidence (2,36) establishing whether precooling during an active warm-up alters performance and neuromuscular function during a performance trial. Moreover, precooling techniques such as immersion and showers are not practical immediately before long-distance competition (2,30). Portable precooling techniques that could be used at the side of a track or road before a race, consisting of ice vests, may prevent augmented heat strain (systemic effect on the upper body) (8), whereas ice packs on the thighs may prevent increased muscle temperature (local effect involving direct muscle cooling) (7,10). Passive heating increases muscle temperature and may augment the rate of acidification, causing a greater decline in force production and subsequent performance (39). Nevertheless, the use of intermittent exercise does not accurately represent the typical physiological mechanisms occurring in endurance exercise. Therefore, it is unclear whether the changes in heat strain and muscle temperature established with torso or thigh precooling will similarly influence endurance exercise performance. Moreover, previous studies comparing a single precooling technique with a control condition of no cooling have denoted that precooling with either an ice vest (5.1%) or cooling packs (6.3%) does not elicit superior or inferior changes in performance (44). Thus, it remains uncertain whether torso precooling in contrast to thigh precooling is the superlative, practical technique. Furthermore, there is limited research examining the application of either an ice vest or cooling packs during a race-specific warm-up. With coaches and athletes needing better guidance on the usefulness of different cooling strategies before endurance running events, more evidence is required to evaluate simple, cost-effective cooling methods. Strength and conditioning professionals could also benefit in the knowledge that precooling strategies could help strength-based tasks that have a prolonged endurance element to them.

Accordingly, the aim of this investigation was to examine the effect of isolated torso and thigh precooling throughout a warm-up on neuromuscular function and 5-km time-trial performance in hot, humid conditions. It was hypothesized that torso and thigh precooling would aid running performance during a 5-km time-trial within a hot environment and minimize neuromuscular fatigue.


Experimental Approach to the Problem

Subjects were required to visit the laboratory on 4 occasions with each visit separated by 2–4 days of recovery. All subjects completed a familiarization session to certify understanding with the testing equipment and procedures and to establish optimal electrical stimulation intensities. They also performed an incremental exercise test to determine maximal oxygen uptake (V[Combining Dot Above]O2max). Then, in a randomized and counterbalanced order, 3 experimental trials were performed inside an environmental chamber in hot, humid conditions (32.2 ± 0.8° C; 48.6 ± 6.7% relative humidity) at similar times of day to minimize the influence of circadian variation (31). Before each testing session, electrical stimulation of the femoral nerve was performed during 3 maximal voluntary contractions (MVCs) at resting baseline to assess neuromuscular function and voluntary activation (VA). A prescribed 30-minute warm-up similar to that prescribed by Arngrïmsson et al. (2) and Stannard et al. (36) was then completed where the independent variable was either no cooling, where subjects wore a regular t-shirt with shorts containing neutral temperature packs over the thighs (Control), or precooling by wearing an ice vest (Vest) or frozen gel packs over the thighs (Packs). The vest or packs on the thighs were removed following the warm-up. The 5-km time-trial was then undertaken on a motorized treadmill to obtain the primary outcome–dependent variable of 5 km running time. During the warm-up and time-trial, core and skin temperatures (Tcore and Tsk, respectively), heart rate (HR), body heat content change, rating of perceived exertion, and thermal sensation were measured. Throughout the time-trial, 500-m cumulative and split times, velocity, and time to complete the 5 km were recorded. Finally, neuromuscular assessments were performed following 5-km time-trial completion.


Eight well-trained, male, club long-distance runners volunteered for the study (mean ± SD age: 34.8 ± 4.4 years; stature: 179.4 ± 4.6 cm; body mass: 72.0 ± 8.8 kg; training volume: 30.3 ± 13.7 km·wk−1; maximal aerobic velocity: 17.8 ± 1.2 km·h−1; maximal oxygen uptake: 65.5 ± 3.9 ml·kg−1·min−1). There were three 5,000 m, two 10,000 m, one half-marathon, one marathon, and one Ironman 70.3 specialists. All subjects completed more than 3 years of long-distance training, one 5-km time-trial per year, and performed 5 km in 19.5 ± 0.9 minutes (range: 18:22–21:07 minutes:seconds). Before testing, subjects had the experimental procedures explained to them. All subjects provided written informed consent to participate in this study, which was approved by the Ethics Committee at the University of Brighton and conformed to the Declaration of Helsinki 2013. Subjects completed trials during the British winter season (October to February) and were not heat acclimatized. Subjects were instructed to arrive at the laboratory in a fully hydrated state, maintain their normal diet, and replicate this before subsequent visits. They were asked to refrain from vigorous exercise for 48 hours, avoid alcohol for 24 hours, and avoid caffeine and food for 2 hours before testing.


Preliminary Session

On arrival, data were collected comprising age, height (Detecto Scales; Detecto, Webb City, MO, USA), and body mass (Seca 778; Seca, Hamburg, Germany). The subjects performed brief, isometric contractions of the knee extensors in hot, humid conditions until they were accustomed to the equipment. Subjects were acquainted with receiving femoral nerve stimulation during these MVCs. Subjects then completed the prescribed warm-up to become accustomed with the individualized warm-up intensities, stretching, and strides protocol. Afterwards, they performed an incremental exercise test to determine V[Combining Dot Above]O2max on a treadmill (Woodway ELG; Woodway, Weil am Rhein, Germany) in a temperate laboratory environment (18–20° C, ∼40% relative humidity). Following a 10-minute jog at 8 km·h−1, speed was increased to 10 km·h−1 with speed increments of 1 km·h−1 every 60 seconds until volitional exhaustion. Heart rate was recorded and respiratory indices were assessed breath-by-breath using the MetaLyzer Sport online gas analysis system (Cortex, Leipzig, Germany).

Precooling During Warm-up Protocol

The warm-up protocol completed before each 5-km time-trial was identical and intended to simulate stretches and distance runners covered preceding a race. Precooling occurred only during the warm-up to optimize neuromuscular performance, considered ideal for race preparation. Following neuromuscular function assessments, subjects mounted the treadmill with the precooling intervention. After collection of pre-warm-up measurements, subjects performed a 5-minute warm-up at their typical warm-up speed (determined during the preliminary session), followed by 10 minutes of prescribed static and dynamic stretching exercises. They completed a further 10-minute run on the treadmill at a faster (1.6 km·h−1) pace than the initial 5 minutes (2,36). Four 30-second strides at just below and just above race pace (±0.5 km·h−1) were performed with 45 seconds of standing recovery. Subjects removed the vest or packs on the thighs and subsequent pre-time-trial measures were recorded. The Vest (Arctic Heat Products, Westwood, New Jersey) and Packs (Hot-Cold Pack; Koolpak, Poole, United Kingdom) weighed 2,388 and 2,376 g, respectively. On removal from a 20° C freezer, the Vest and Packs with thermistors attached during cooling had a surface temperature of 10.7 ± 2.5° C and −16.0 ± 5.8° C, respectively (7). The Packs were secured within compartments of bespoke shorts to the anterior, lateral, and posterior aspects of the thighs to completely cover the quadriceps and hamstring muscle groups. The only difference between conditions was the addition of wearing unfrozen Packs (6 for Control, 3 for Vest) during the warm-up to ensure similar energy cost of the garment compared with the Packs condition.

Five-Kilometer Time-Trial

The 5-km time-trial was immediately commenced, with the treadmill set at 1% gradient to simulate ground running (20). Subjects increased or decreased the speed themselves and were instructed to perform maximally over the distance. Subjects were taken out of the environmental chamber following completion of the 5-km time-trial and neuromuscular assessments or on inability to continue or attainment of the ethics-approved Tcore limit (39.7° C). Subjects were informed on completion of every 500 m and aware that performance measures were being recorded. However, they were not provided with verbal encouragement, 500-m splits, total time elapsed during the time-trial, or 5-km performance times until study completion. Reliability of the 5-km time-trial protocol in our laboratory has found a typical error of measurement of 2.5% from a heterogeneous group of 14 physically active individuals. This is similar to the 2.0% typical error of measurement reported by Laursen et al. (21) among endurance-trained distance runners.

Force and Electromyographic Recordings

Knee extensor force throughout voluntary contractions was assessed using a calibrated load cell (Model 615; Tedea-Huntleigh, Chatsworth, California). The load cell was fixed to a bespoke chair and adjusted to a height that was in the direct line of applied force for each participant. The load cell was connected to a noncompliant cuff attached around the right leg just superior to the ankle malleoli. Subjects sat upright in the chair, secured with a shoulder and waist strap, with the hips and knees at 90° of flexion (16,34). Electromyographic (EMG) activity of the knee extensors and flexors was recorded from the vastus lateralis and biceps femoris. After the skin was shaved and swabbed with isopropyl 70% alcohol, surface electrodes (Kendall H59P; Covidien Ilc., Mansfield, MA, USA) were placed with an interelectrode distance of 2 cm over the muscle bellies. A reference electrode was positioned over the patella. The positions of all electrodes were marked with indelible ink to ensure reproducibility of placement throughout subsequent visits. Electrodes were replaced following the time-trial. Surface electrodes were used to monitor the compound muscle action potential (M-wave) obtained by electrical stimulation of the femoral nerve (17). Electromyographic electrodes were connected to data acquisition hardware (Bioamp Power Lab, 15T; ADInstruments, Bella Vista, Australia) and data were viewed by specific software (Lab Chart 7.3.5; ADInstruments). Signals were amplified (ADInstruments), band-pass filtered (EMG only: 20–2,000 Hz), digitized (4 kHz; ADInstruments), then acquired, and later analyzed (Lab Chart 7.3.5; ADInstruments).

Neuromuscular Function

Force and EMG variables were assessed inside the environmental chamber before the warm-up and immediately following the 5-km time-trial. During each MVC, subjects positioned their arms across their chest. Maximal voluntary contraction force was established from 3 maximal contractions (duration 5 seconds, 30 seconds recovery). Thereafter, 3 MVCs were performed with femoral nerve stimulation delivered during each MVC, and an additional stimulus was delivered at rest, approximately 2 seconds following the superimposed stimulus, to determine potentiated quadriceps twitch force (Qtw,pot) and peripheral VA (see Data Analysis) (17,34). Subjects were provided with strong verbal encouragement during voluntary efforts. Measurements during the post-time-trial neuromuscular assessments were completed within 3 minutes after exercise termination.

Femoral Nerve Stimulation

Single electrical stimuli of 1,000 μs pulse width were delivered to the right femoral nerve by 50-mm diameter surface electrodes (Starburst 627SB; Tyco Healthcare Uni-Patch, Wabasha, MN, USA) using a constant current stimulator (DS71; Digitimer Ltd., Welwyn Garden City, United Kingdom). The cathode was positioned over the nerve high in the femoral triangle, and the anode was placed midway between the greater trochanter and iliac crest (17). The site of stimulation that produced the largest resting twitch amplitude was located. Stimulation intensity began at 10 mA and was progressively increased by 10 mA until a plateau in peak twitch force was attained. The final intensity was further increased by 30% (e.g., supramaximal; mean current: 147.9 ± 20.8 mA) and maintained for subsequent sessions. Muscle contractility was assessed by measuring the amplitude of the potentiated muscle twitch evoked by motor nerve stimulation (Qtw,pot). Membrane excitability was established by the measurement of the peak-to-peak amplitude and area of the electrically evoked M-wave (Mmax).

Data Analysis

For MVCs, data from the largest generated peak force and largest Qtw,pot were taken for subsequent analysis. Peripheral VA was assessed using twitch interpolation (24). In brief, the force generated during a superimposed single twitch (SIT) delivered within 0.5 seconds of reaching peak force during the MVC was compared with the force generated by the single twitch delivered during relaxation ∼2 seconds after the MVC: VA (%) = [1 − (SIT/Qtw,pot)] × 100. The reliability of the femoral nerve stimulation protocol for the assessment of peripheral VA for the knee extensors was determined in our laboratory (within- and between-day coefficients of variation: 3.9 and 5.3%, respectively).

Before experimental sessions, subjects provided a urine sample on arrival to the laboratory. Urine osmolality (Osmocheck; Vitech Scientific Ltd., Horsham, UK) and urine specific gravity (Hand Refractometer; Atago, Tokyo, Japan) were assessed to establish hydration status and ensure all subjects commenced exercise euhydrated (urine specific gravity: <1.020; urine osmolality: <700 mOsm·L−1) (6). Towel-dried nude body mass was assessed before and after exercise on a set of scales to determine nonurine fluid loss. Fluid intake was not permitted in any condition.

Tcore (4600 Thermometer; Henleys Medical Supplies, Welwyn Garden City, United Kingdom) was monitored from a depth of 10 cm past the anal sphincter. Tsk was recorded using surface thermistors (Squirrel 1002; Grant Instruments, Cambridge, United Kingdom) attached under running apparel to the skin on the right side of the body on the chest, upper arm, thigh, and calf as described by Ramanathan (32). Tcore and Tsk were measured pre warm-up, every 10 minutes during the warm-up, post warm-up, and at every 1 km during the time-trial. Body heat content (Hb) change was calculated based on the equation of Jay and Kenny (19).

Heart rate (Polar sports tester; Polar Electro, Kempele, Finland), thermal sensation (TS; 8-point Likert scale) (40), and rating of perceived exertion (RPE) according to Borg's 6- to 20-point scale were recorded were measured pre warm-up, every 10 minutes during the warm-up, post warm-up, and at every 1 km during the time-trial.

Statistical Analyses

Data were checked for normality, and sphericity was adjusted using the Huynh-Feldt method. For the dependent variables that display change across time (performance data: velocity, cumulative and split times; physiological measures: Tcore, Tsk, Hb, and HR; and perceptual ratings: RPE, TS), a 2-way repeated-measures ANOVA (condition × time) was used to determine the influence of the interventions within and between conditions, with Bonferroni corrected pairwise comparisons performed. Where significance was not obtained, effect size data were calculated (partial eta squared: η2) to determine the magnitude of the interventions on performance and certain neuromuscular measures. An effect size of 0.01 was classified as a “small,” 0.06 as a “moderate,” and >0.14 as a “large” effect. Student's paired t-test was used to assess baseline to postexercise differences for body mass and neuromuscular function. Pearson's product moment correlation coefficient was calculated to establish the relationship between changes in performance time and MVC differences. Data were analyzed using a standard statistical package (SPSS version 20.0) and reported as mean ± SD. Statistical significance was accepted at p ≤ 0.05.



Four subjects completed all 3 conditions. Four subjects completed the entire time-trial in Control, and 6 subjects completed the time-trial in Vest and Packs. All remaining subjects terminated exercise early because of attainment of the Tcore limit. At a distance of 3,500 m, 2 subjects terminated time-trial completion in Control. At 4,500 m, 1 additional subject in Control and Vest and 2 subjects in Packs terminated exercise. By 5,000 m, 1 further subject in Control and Vest terminated time-trial completion.

No significant main effects were present for cumulative or split time. There were no significant differences in cumulative time until the last kilometer, when a faster cumulative time occurred at 4,000 m (Control n = 6, Packs n = 8) and 5,000 m (Control n = 4, Packs n = 6) in Packs compared with Control (4,000 m: 1,137 ± 75 seconds vs. 1,219 ± 65 seconds; 6.7%; p < 0.01; 5,000 m: 1,407 ± 80 seconds vs. 1,492 ± 88 seconds; 5.7%; p ≤ 0.05 for Packs and Control, respectively; Table 1). However, a large effect was present for quicker cumulative time in Packs compared with Control (η2 = 0.37; p = 0.22). Precooling by Vest (4,500 m: n = 7; 5,000 m: n = 6) compared with Control did not result in a significantly quicker cumulative time during the time-trial, especially at 5,000 m (3.2%; 1,444 ± 71 seconds vs. 1,492 ± 88 seconds; η2 = 0.18; p = 0.17). No significant differences were observed in Packs compared with Vest (2.6%; η2 = 0.08; p = 0.53).

Table 1:
Cumulative time and split time during the 5-km time-trial following the 30-minute warm-up in the heat with Control, Vest, and Packs precooling.

There were no significant differences in split time in Packs compared with Control; however, a large effect was present for quicker split time with precooling by Packs (η2 = 0.30; p = 0.14). There were no differences during the time-trial in Packs until at 1,500 m, when a faster split time than the initial 500 m occurred (p < 0.01; Table 1). No significant differences were denoted until 3,000 m, when precooling by Vest compared with Control caused a quicker split time at 3,500 m (p ≤ 0.05). A large effect was present for split time in Vest compared with Control (η2 = 0.14; p = 0.17). No significant differences and a moderate effect size were identified in Packs compared with Vest (η2 = 0.08; p = 0.24).

No significant main effects were present for mean velocity. There was a significant increment in mean velocity in Packs compared with Control at 1,000, 1,500, and 3,500 m (p ≤ 0.05; Figure 1). A large effect size was present for increased velocity in Packs compared with Control (η2 = 0.28; p = 0.06). No significant differences were identified in Vest compared with Control (η2 = 0.13; p = 0.21) or Packs (η2 = 0.09; p = 0.14). During the last kilometer of each time-trial, change in performance time was greater in the final 500 m than the penultimate 500 m (Control: 8.3 vs. 2.0% and Packs: 4.3 vs. 0%; p ≤ 0.05; Vest: 3.8 vs. 0.5%; p < 0.01) but was not significant between conditions.

Figure 1:
Velocity during the 5-km time-trial following the 30-minute warm-up in the heat with control (Control), ice vest (Vest), and frozen gel packs (Packs) precooling. †Significant difference between control and packs, p < 0.01. aControl (n = 6); bControl (n = 5), Vest (n = 7), Packs (n = 6); cControl (n = 4), Vest and Packs (n = 6).

Muscle Function

Baseline neuromuscular function measures did not vary across conditions. The difference in MVC force after the time-trial in Control, Packs, and Vest is shown for each individual in Figure 2A. Compared with Control, an equal or greater reduction in ΔMVC force was observed in Packs in 5 of 8 subjects. Therefore, post-time-trial, ΔMVC force was not significantly reduced below baseline or between conditions (Control: −7%; Packs: −9%; and Vest: −13%; p > 0.05). The difference in performance time was positively correlated to MVC differences obtained before warm-up and after time-trial in Packs compared with Control (r = 0.73; p ≤ 0.05) but remained nonsignificant in Vest compared with Control (r = 0.59; p = 0.12), and Packs compared with Vest (r = −0.21; p > 0.05).

Figure 2:
Individual differences in (A) maximum voluntary contraction (MVC) and (B) voluntary activation between Control, Vest, and Packs precooling conditions immediately after the 5-km time-trial. Mean data and SD for the difference in the mean of 8 well-trained male runners are also presented.

After the time-trial, VA and Qtw,pot were not significantly different from baseline or between conditions (VA: 1, −7, and −5%; Qtw,pot: 0, −13, and −2%; p > 0.05, for Control, Packs, and Vest, respectively; Table 2). A greater reduction from baseline in VA was identified in Packs and Vest in 4 and 2 subjects, respectively (Figure 2B). In only 2 subjects, VA increased above baseline in either Control or Vest. Mmax amplitude and area were not significantly different post-time-trial or between conditions (Table 2). However, a small-to-moderate effect was evident for an increased Mmax amplitude in Packs compared with Vest (η2 = 0.02; p = 0.31).

Table 2:
Neuromuscular function before the 30-minute warm-up in the heat with Control, Vest, and Packs precooling and immediately after the 5-km time-trial.*

Urine, Nude Body Mass, and Heart Rate

Pre-exercise hydration status did not differ between conditions for urine specific gravity (1.009 ± 0.007; 1.007 ± 0.006; 1.010 ± 0.007; p > 0.05) or urine osmolality (325 ± 248 mOsm·L−1; 305 ± 182 mOsm·L−1; 369 ± 256 mOsm·L−1; p > 0.05 for Control, Packs and Vest, respectively). The change in nude body mass following the time-trial, denoting nonurine fluid loss, was not significantly different post-time-trial or between conditions (Control: −1.6 ± 0.4 kg; Packs: −1.5 ± 0.2 kg; and Vest: −1.4 ± 0.6 kg; p > 0.05).

A significant main effect was present for an increase in HR during the time-trial (p ≤ 0.05) but not the warm-up. Heart rate was not significantly different between conditions during the warm-up or time-trial (p > 0.05; Table 3).

Table 3:
Cardiorespiratory, thermoregulatory, and perceptual responses during the warm-up and 5-km time-trial in the heat with Control, Vest, and Packs precooling.

Core and Skin Temperature and Hb

No significant main effects were present for Tcore and Tsk. Tcore was not significantly different between conditions during the warm-up or time-trial (p > 0.05; Figure 3A). In Packs and Vest, Tsk was significantly reduced after the 10th minute during the warm-up compared with Control (p < 0.01) but not in Packs compared with Vest (Figure 3B). In Packs, ΔTsk was significantly increased post-warm-up (0.65 ± 1.5° C) but remained significantly lower than Control (33.3 ± 0.8° C vs. 34.8 ± 0.7° C; p ≤ 0.05) but not in Vest compared with Control (33.6 ± 1.1° C vs. 34.8 ± 0.7° C) or Packs compared with Vest. Tsk was not significantly different between conditions during the time-trial.

Figure 3:
Core temperature (A) and skin temperature (B) during the 5-km time-trial following the 30-minute warm-up in the heat with Control, Vest, and Packs precooling. *Significant difference between Control and Vest, p < 0.01. †Significant difference between Control and Packs, p < 0.01. ‡Significant difference between Control and Packs, p ≤ 0.05. §Significant difference from the start of the warm-up in packs, p ≤ 0.05. aControl (n = 6); bControl (n = 4), Vest, and Packs (n = 6).

A significant main effect was present for Hb change during the warm-up (p ≤ 0.05) but not the time-trial. In Packs, Hb change was significantly lower during and post-warm-up compared with Control (p ≤ 0.05; Table 3). Hb change was not significantly different in Vest compared with Control, or Packs compared with Vest during the warm-up. Hb change was significantly increased post-warm-up in all conditions (p < 0.01); however, it was not significantly different between conditions during the time-trial (p > 0.05).

Perceptual Responses

No significant main effects were present for RPE during the warm-up or time-trial. Rating of perceived exertion was not significantly different between conditions during the warm-up or time-trial (p > 0.05; Table 3). A significant main effect was present for an increase in TS during the time-trial (p ≤ 0.05) but not the warm-up. Thermal sensation was not significantly different in Packs compared with Control, or Packs compared with Vest during the warm-up. However, TS was significantly reduced in Vest compared with Control (p < 0.01–0.05) until the 30th minute of the warm-up when no significant reductions were observed (p = 0.054). Thermal sensation was not significantly different between conditions during the time-trial (Table 3).


The application of thigh precooling during a 30-minute warm-up improved 5-km time-trial performance compared with the control condition but was not improved with torso precooling. In particular, velocity was increased at 1,000, 1,500, and 3,500 m in Packs and subsequent cumulative time improved during the last kilometer of the time-trial. Both precooling interventions did not affect voluntary force production, muscle contractility, or membrane excitability. However, after 10 minutes of the warm-up, precooling by Vest and Packs reduced skin temperature. Furthermore, body heat content change reduced in Packs, whereas thermal sensation reduced in Vest during the warm-up. At the start of the time-trial, Packs reduced skin temperature.

Coinciding with previous research, precooling aids performance during endurance exercise (5,11). Several studies have identified equivalent findings during prolonged self-paced exercise, with a 13-second improvement for a 5-km running time-trial (2), a 304-m improvement in performance for a 30-minute running time-trial (5), and a 20-W increase in mean power during a 40-minute cycling time-trial (11). These studies similarly identified reductions in skin temperature and perceived thermal stress. In contrast, the present study did not observe the reductions of core temperature or sweat loss, possibly because of the lack of precooling before exercise. Although precooling reduces core temperature and remains lower during exercise (5), cold-water immersion is likely to cause a greater depth of cooling compared with the current methods. Therefore, the limited sample size is another explanation. The present data, alongside research using time-to-exhaustion exercise (43), emphasize the ergogenic benefits of external precooling for endurance exercise in the heat. Furthermore, low variability of 5-km time-trials established in our laboratory and among endurance-trained distance runners (typical error of measurement as a coefficient of variation: 2.5 and 2.0%, respectively) accentuates that enhanced endurance performance likely results from the precooling interventions (21). Nevertheless, few studies have explored the influence of precooling on neuromuscular function and performance during a trial among track runners. Moreover, only some studies have used precooling during an active warm-up to identify whether similar ergogenic benefits occur compared with impractical methods, such as cold-water immersion.

The performance improvement apparent at the last kilometer of the time-trial in Packs may have resulted from increased exercise intensity depicted at 1,000, 1,500, and 3,500 m. Furthermore, the improved performance time of 48 seconds observed in Vest converts into a 168-m advantage at the average velocity run in Vest, which is considerable in 5,000-m competition. Together, the large effects present for quicker cumulative and split time, and increased velocity following Packs, and moderate-to-large effects after Vest indicates the ergogenic benefits of precooling, although a larger sample size could confirm this implication (Control, n = 4; Packs, n = 6; Vest n = 6). Nevertheless, the effect of precooling induced by Packs seemed greater compared with Vest, with an improvement in 5-km performance time by 37 seconds. Tucker et al. (42) and Marino (22) denoted that self-paced, prolonged exercise in the heat causes a premature reduction in performance compared with thermoneutral conditions. Therefore, coinciding with previous studies (2,10), precooling promotes higher exercise intensities to be sustained during endurance exercise in the heat, despite the benefits remaining unclear.

The earlier reductions in exercise intensity observed in hot conditions may be because of reduced muscle recruitment by afferent feedback derived from the environmental condition and endogenous load (27). This may affect function of the central nervous system, where this decline in skeletal muscle activation subsequently reduces exercise intensity (15). Further physiological responses may reduce muscle recruitment or ability to maintain a higher exercise intensity as apparent in the Control condition, including increased brain temperature (28) or decreased neural transmission (27). Increased core temperature to a critical level of 40° C may also reduce muscle recruitment and cause an early termination of exercise (15,42). Nevertheless, highly trained athletes can maintain exercise intensity and produce an endspurt toward completion of an 8-km running time-trial, despite considerable heat storage and core temperature surpassing 40° C (13,41). In the present study, an endspurt occurred approximately 1,000 m before time-trial completion, even among participants terminating exercise early because of reaching the safety limit of 39.7° C. Although some subjects may be heat intolerant (29), this indicates either the involvement of centrally mediated factors or reserve from anaerobic energy pathways, or both.

Previously, Duffield et al. (11) reported that 20-minute lower-body precooling did not alter contractile function following 40 minutes of self-paced cycling. In agreement, the present study identified no alterations to voluntary force production, VA, or Qtw,pot between conditions. However, individual differences may exist in how precooling influences muscle function because force considerably reduced or remained relatively unaltered in 5 subjects after the time-trial with Packs. Furthermore, a superior decline in VA was observed with Packs and Vest among several subjects. Considering that fatigue resulting from the time-trial will probably supersede any effect of the precooling, the current study sought to establish the influence of precooling methods on muscle function and subsequent performance. A reduction in force production reported in the plantar flexors following 5-km running in thermoneutral conditions (14) implies that precooling minimally affects overall force generating capacity of the knee extensors. Maintained VA suggests that neural input reaching the neuromuscular junction was not impaired therefore enabling full ability to drive the motor neurons (14), although using techniques such as transcranial magnetic stimulation may further identify the location of central fatigue. The assessment of muscle contractility and membrane excitability, of which were not altered by Packs or Vest precooling, denotes that no changes in excitation-contraction coupling occurred (1,14). Furthermore, no alterations of Qtw,pot with Packs or Vest highlight that there was an adequate quantity of formed cross-bridges between actin and myosin and their rate of attachment (1). No changes in maximum M-wave amplitude would signify that muscle excitability (ionic disturbances) was maintained (3). Because performance time improved in Packs and without alterations in neuromuscular function, it is expected that the higher sustained velocity, particularly in Packs, was because of the maintenance of muscle recruitment during exercise (26).

The findings from the present study indicate that a downregulation of intensity during the performance time-trial may be related to greater thermoregulatory strain because of exercise within hot conditions. Packs seemed to ameliorate body heat content change, while similar to Arngrïmsson et al. (2), subjects perceived their thermal stress lower in Vest during the warm-up. The precooling interventions may have altered thermoregulatory strain differently, whereby the hybrid gel Vest containing ice crystals superficially cooled the skin, whereas the gel within Packs caused a greater depth of cooling. Furthermore, the thigh is likely to have lesser subcutaneous fat thickness than the torso, therefore Packs may have caused more vasoconstriction than Vest. Previously used before prolonged intermittent exercise in the heat (7), Packs may generate larger thermal gradients for conductive cooling subsequently reducing the temperature of the muscle. Although muscle temperature was not measured, previous work in our laboratory has denoted that 25 minutes of Packs during rest reduces mean muscle temperature of a predetermined depth by 15° C (7). Therefore, Packs may have produced a greater heat sink than Vest, explaining the lower-body heat content change identified. Conflicting with Arngrïmsson et al. (2) and Cotter et al. (9) using a Vest combined with or without thigh cooling, core temperature was not significantly different in Vest or Packs during the warm-up or time-trial between conditions. However, previous studies have reported that precooling aids endurance performance despite core temperature remaining constant or even increasing (11,43). Therefore, precooling by Packs and Vest during a warm-up may be used to reduce skin temperature and increase the difference of starting temperatures before commencing a time-trial in the heat. Interestingly, Arngrïmsson et al. (2) reported that the differences in skin temperature had disappeared by 3.2 km of the 5-km time-trial when differences in pacing were evident. However, in the present study, all physiological and perceptual differences had dissipated by the first kilometer. Together, precooling may have enabled the selection of higher exercise intensities derived from the expectation of subsequent constraints based on the reduced physiological responses following thigh and torso precooling.

Previous research highlights that numerous models explore exercise regulation in the heat by feedback (15,28) or feed-forward control (38). The central governor model states that the central nervous system regulates exercise intensity and skeletal muscle output through both feedback and feed-forward control to inhibit catastrophic disruptions in homeostasis (26,37). Consistent with Castle et al. (7), these responses vary with the precooling site and dose, altering overall performance time differently as Packs was improved by 2.6 and 5.7% compared with Vest and Control, respectively. In particular, the reduced skin temperature but no changes in core temperature potentially allowed Vest to create a small heat sink (45), therefore facilitating the management of fatigue associated with heat stress. In Packs, the central governor may have processed a sufficient quantity of sensory information to alter the pacing strategy (7,22) and increase motor unit recruitment and exercise intensity following Packs (11). Despite no significant performance differences between Packs and Vest, the selection of higher exercise intensity indicates that Packs precooling during a warm-up may be an effective technique. Moreover, the similar RPE and thermal and cardiovascular strain during the self-paced time-trial indicates that despite working at higher intensities following Packs, subjects perceived the exercise demands and environmental condition to be similar to Vest and Control. Therefore, exercising at a higher intensity for a certain RPE may have arisen from the reduced thermoregulatory and cardiovascular strain following the thigh cooling, opposed to the identified reductions following torso cooling.

The altered physiological and perceptual responses generated from Packs and Vest may also influence force production differently. In particular, a greater reduction in force production was associated with faster performance time in Packs. Blunting the increase in thermoregulatory strain may have allowed the maintenance of higher exercise intensities and subsequently induced greater reductions in force generation because of the greater endogenous load. Furthermore, the small-to-moderate effect size present for increased maximum M-wave amplitude in Packs compared with Vest indicates that thigh precooling may preserve muscle function. More specifically, completing the warm-up in the heat possibly increased muscle temperature and subsequent enzyme activity and muscle contractile properties (4), whereas Packs may have increased action potential propagation and adequately countered the expected postexercise fatigue because of the cooler muscle (35).

During the warm-up in Control, wearing bespoke shorts containing neutral temperature packs ensured that changes in endurance performance resulted from precooling opposed to differing weight between conditions. Whereas in Vest, the heat had melted the ice vest and diminished any feeling of cool sensations. Therefore, exactly half-way through the warm-up, participants wore a different ice vest to ensure that the physiological benefits associated with precooling could fully function because of a colder melting temperature (18). Interestingly, 5 of 6 subjects providing quantitative feedback reported that the ice vest was the most comfortable garment and believed to optimally aid performance and perception of the heat during the time-trial.

Similar to other studies assessing fatigue of the knee extensors (16,17), the fatigue measurements were completed within 3 minutes after the termination of exercise. Peripheral VA and maximal force production have been reported to remain unchanged within 2.5 minutes following exercise in normoxia (16). Therefore, the present experimental design may have inadequately portrayed all elements of peripheral and central fatigue. However, the duration of the fatigue assessment following exercise remained constant for all 3 trials. Furthermore, force and VA have been shown to recover during cooling from hyperthermia potentially because of a reduction in core temperature (34). Thus, the fatigue measurements were undertaken inside the environmental chamber to ensure that neuromuscular function, alongside the physiological and perceptual responses to the heat were not alleviated.

In conclusion, torso and thigh precooling during a 30-minute active warm-up reduces thermoregulatory strain. Torso precooling with Vest reduces perception of thermal stress, whereas thigh precooling with Packs reduces change in body heat content. However, precooling with Packs was the most effective technique to reduce thermoregulatory strain, therefore enabling an improved selection of exercise intensities and improved performance during the last kilometer of the 5-km time-trial. Although precooling method neither improves nor inhibits neuromuscular function, it is expected that thigh precooling prevents the downregulation of exercise intensity evident in hot, humid conditions. Future investigation should explore the use of precooling during a warm-up in the heat and the prevention of reduced exercise intensity in performance trials potentially resulting from altered sensory feedback and muscle recruitment.

Practical Applications

For athletes competing in long-distance events undertaken within hot, humid environments, warming up with ice packs covering the thighs compared with an ice vest does not seem to provide a superior advantage in performance. Despite a limited sample size in the present study, both practical precooling techniques offer some benefits for exercise in the heat. Furthermore, both methods are simple to set up, inexpensive, and easily transportable, therefore should be considered for use by athletes and coaches. Trialing both techniques before various training sessions and competitions may be advantageous to establish individual preferences, especially with regards to comfort. However, it seems that frozen gel packs covering the upper legs could be the most effective to blunt the rise in thermoregulatory strain. In turn, this may enable athletes to select and maintain a higher exercise intensity, crucial for optimal performance. If deciding to complete a warm-up with a precooling garment, particularly with packs, caution should be taken to ensure athletes do not select an initial pace that is too fast.


The authors thank the support of the University of Brighton and acknowledge the participants for their contribution. The results of the present study do not constitute endorsement of the product by the authors or the National Strength and Conditioning Association.


1. Allen DG, Lamb GD, Westerblad H. Skeletal muscle fatigue: Cellular mechanisms. Physiol Rev 88: 287–332, 2008.
2. Arngrïmsson SA, Petitt DS, Stueck MG, Jorgensen DK, Cureton KJ. Cooling vest worn during active warm-up improves 5-km run performance in the heat. J Appl Physiol (1985) 96: 1867–1874, 2004.
3. Bigland-Ritchie B, Jones DA, Woods JJ. Excitation frequency and muscle fatigue: Electrical responses during human voluntary and stimulated contractions. Exp Neurol 64: 414–427, 1979.
4. Bishop D. Warm up I: Potential mechanisms and the effects of passive warm up on exercise performance. Sports Med 33: 439–454, 2003.
5. Booth J, Marino F, Ward JJ. Improved running performance in hot humid conditions following whole body precooling. Med Sci Sports Exerc 29: 943–949, 1997.
6. Casa DJ, Clarkson PM, Roberts WO. American College of Sports Medicine roundtable on hydration and physical activity: Consensus statements. Curr Sports Med Rep 4: 115–127, 2005.
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 (1985) 100: 1377–1384, 2006.
8. Castle PC, Spencer M, McNaught-Davis P, Watt PW, Maxwell NS. Quantification of heat strain during intermittent sprint exercise in male games players while under heat stress. J Physiol 565P: PC31, 2004.
9. Cotter JD, Sleivert GG, Roberts WS, Febbraio MA. Effect of pre-cooling, with and without thigh cooling, on strain and endurance exercise performance in the heat. Comp Biochem Physiol 128: 667–677, 2001.
10. Drust B, Rasmussen P, Mohr M, Nielsen B, Nybo L. Elevations in core and muscle temperature impairs repeated sprint performance. Acta Physiol Scand 183: 181–190, 2005.
11. 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 42: 577–584, 2010.
12. Ely BR, Cheuvront SN, Kenefick RW, Sawka MN. Aerobic performance is degraded, despite modest hyperthermia, in hot environments. Med Sci Sports Exerc 42: 135–141, 2010.
13. Ely BR, Ely MR, Cheuvront SM, Kenefick RW, DeGroot DW, Montain SJ. Evidence against a 40°C core temperature threshold for fatigue in humans. J Appl Physiol (1985) 107: 1519–1525, 2009.
14. Girard O, Millet GP, Micallef JP, Racinais S. Alteration in neuromuscular function after a 5 km running time trial. Eur J Appl Physiol 112: 2323–2330, 2012.
15. González-Alonso J, Teller C, Andersen SL, Jensen FB, Hyldig T, Nielsen B. Influence of body temperature on the development of fatigue during prolonged exercise in the heat. J Appl Physiol (1985) 86: 1032–1039, 1999.
16. Goodall S, González-Alonso J, Ali L, Ross EZ, Romer LM. Supraspinal fatigue after normoxic and hypoxic exercise in humans. J Physiol 590: 2767–2782, 2012.
17. Goodall S, Ross EZ, Romer LM. Effect of graded hypoxia on supraspinal contributions to fatigue with unilateral knee-extensor contractions. J Appl Physiol (1985) 109: 1842–1851, 2010.
18. House JR, Lunt HC, Taylor R, Milligan G, Lyons JA, House CM. The impact of a phase-change cooling vest on heat strain and the effect of different cooling pack melting temperatures. Eur J Appl Physiol 113: 1223–1231, 2013.
19. Jay O, Kenny GP. The determination of changes in body heat content during exercise using calorimetry and thermometry. J Hum Environ Syst 10: 19–29, 2007.
20. Jones AM, Doust JH. A 1% treadmill grade most accurately reflects the energy cost of outdoor running. J Sports Sci 14: 321–327, 1996.
21. Laursen PB, Francis GT, Abbiss CR, Newton MJ, Nosaka K. Reliability of time-to-exhaustion versus time-trial running tests in runners. Med Sci Sports Exerc 39: 1374–1379, 2007.
22. Marino FE. Methods, advantages and limitations of body cooling for exercise performance. Br J Sports Med 36: 89–94, 2002.
23. Marino FE, Lambert MI, Noakes TD. Superior performance of African runners in warm humid but not in cool environments. J Appl Physiol (1985) 96: 124–130, 2004.
24. Merton PA. Voluntary strength and fatigue. J Physiol 123: 553–564, 1954.
25. Minett GM, Duffield R, Marino FE, Portus M. Volume-dependent response of pre-cooling for intermittent-sprint exercise in the heat. Med Sci Sports Exerc 43: 1760–1769, 2011.
26. Nummela AT, Heath KA, Paavolainen LM, Lambert MI, St Clair Gibson A, Rusko HK, Noakes TD. Fatigue during a 5 km running time trial. Int J Sports Med 29: 738–745, 2008.
27. Nybo L. Hyperthermia and fatigue. J Appl Physiol (1985) 104: 871–878, 2008.
28. Nybo L, Nielsen B. Hyperthermia and central fatigue during prolonged exercise in humans. J Appl Physiol (1985) 91: 1055–1060, 2001.
29. Périard JD, Caillaud C, Thompson MW. The role of aerobic fitness and exercise intensity on endurance performance in uncompensable heat stress conditions. Eur J Appl Physiol 112: 1989–1999, 2012.
30. Quod MJ, Martin DT, Laursen PB. Cooling athletes before competition in the heat: Comparison of techniques and practical considerations. Sports Med 36: 671–682, 2006.
31. Racinais S. Different effects of heat exposure upon exercise performance in the morning and afternoon. Scand J Med Sci Sports 20: 80–89, 2010.
32. Ramanathan NLA. A new weighting system for mean surface temperature of the human body. J Appl Physiol (1985) 19: 531–533, 1964.
33. Ranalli GF, Demartini JK, Casa DJ, McDermott BP, Armstrong LE, Maresh CM. Effect of body cooling on subsequent aerobic and anaerobic exercise performance: A systematic review. J Strength Cond Res 24: 3488–3496, 2010.
34. Ross EZ, Cotter JD, Wilson L, Fan JL, Lucas SJ, Ainslie PN. Cerebrovascular and corticomotor function during progressive passive hyperthermia in humans. J Appl Physiol (1985) 112: 748–758, 2012.
35. Skein M, Duffield R, Cannon J, Marino FE. Self-paced intermittent-sprint performance and pacing strategies following respective pre-cooling and heating. Eur J Appl Physiol 112: 253–266, 2012.
36. Stannard AB, Brandenburg JP, Pitney WA, Lukaszuk JM. Effects of wearing a cooling vest during the warm-up on 10-km run performance. J Strength Cond Res 25: 2018–2024, 2011.
37. St Clair Gibson A, Lambert EV, Rauch LHG, Tucker R, Baden DA, Foster C, Noakes TD. The role of information processing between the brain and peripheral physiological systems in pacing and perception of effort. Sports Med 36: 705–722, 2006.
38. St Clair Gibson A, Noakes TD. Evidence for complex system integration and dynamic neural regulation of skeletal muscle recruitment during exercise in humans. Br J Sports Med 38: 797–806, 2004.
39. Thornley LJ, Maxwell NS, Cheung SS. Local tissue temperature effects on peak torque and muscular endurance during isometric knee extension. Eur J Appl Physiol 90: 588–594, 2003.
40. Toner MM, Sawka MN, Foley ME, Pandolf KB. Effects of body mass and morphology on thermal responses in water. J Appl Physiol (1985) 60: 521–525, 1986.
41. Tucker R. The anticipatory regulation of performance: The physiological basis for pacing strategies and the development of a perception-based model for exercise performance. Br J Sports Med 43: 392–400, 2009.
42. Tucker R, Marle T, Lambert EV, Noakes TD. The rate of heat storage mediates an anticipatory reduction in exercise intensity during cycling at a fixed rating of perceived exertion. J Physiol 574: 905–915, 2006.
43. Ückert S, Joch W. Effects of warm-up and precooling on endurance performance in the heat. Br J Sports Med 41: 380–384, 2007.
44. Wegmann M, Faude O, Poppendieck W, Hecksteden A, Fröhlich M, Meyer T. Pre-cooling and sports performance: A meta-analytical review. Sports Med 42: 545–564, 2012.
45. White AT, Davis SL, Wilson TE. Metabolic, thermoregulatory, and perceptual responses during exercise after lower vs. whole body precooling. J Appl Physiol (1985) 94: 1039–1044, 2002.

ice vest; thigh precooling; endurance running; voluntary activation; thermoregulation

Copyright © 2015 by the National Strength & Conditioning Association.