The Effects of Palm Cooling on Repeat Sprint Ability Following a Fatigue Inducing Protocol in Collegiate Female Athletes

Abstract

Cooling glabrous skin sites, such as the palms, can mitigate core body temperature rise and delay fatigue during physical activity, potentially enhancing performance. However, research on palm cooling (PC) in female athletes remains limited, and existing devices are often costly or impractical. This study evaluated the effects of PC using a novel, cost-effective portable device on repeat sprint ability in female collegiate athletes after a fatiguing protocol. Twenty female athletes (age: 20.1±1.4 years; height: 166.7±6.9 cm; mass: 66.6±9.8 kg; BMI: 24.0±3.9 kg/m²; body fat: 24.9±4.6%) participated in a randomized crossover study. Sessions included a modified Loughborough Intermittent Shuttle Test (LIST) followed by a 20-meter repeat sprint test. Athletes used either the PC device (7–15°C) during rest intervals or no cooling (NC). Physiological (heart rate, blood lactate), perceptual (RPE, thermal perception, affect), and performance outcomes (number sprints) were recorded. PC resulted in significantly more successful sprints (≥90% of maximum velocity) than NC (10.3±12.9 vs. 6.0±8.4; p=0.025). No significant differences were observed in perceptual or physiological measures (p>0.05). Palm cooling with a portable device improved repeat sprint ability in female athletes, supporting its use as a practical performance optimization strategy. Further research is warranted to investigate underlying mechanisms and applications across different sports and environmental conditions.

Introduction

During exercise, the cardiovascular system faces increased demands as working muscles require more blood supply. Fatigue occurs when the system can no longer maintain adequate arterial oxygen delivery to these muscles.1 As a result of this enhanced blood flow, core body temperature is elevated, contributing to fatigue and impairing performance.2 Aerobic capacity declines when skin and core temperatures become elevated. While hyperthermia-induced central fatigue does not appear to affect performance at brief maximal efforts, it has been found to affect performance during prolonged muscle activation periods.3 These prolonged periods of muscle activation occur primarily when continuous muscle activation is required throughout the duration of a practice or match. Potential solutions to prevent or delay the onset of fatigue are pre-cooling (cooling before physical activity), per-cooling (cooling during physical activity), or a combination of the two. Some of the most effective areas of the body for these cooling strategies are those composed of glabrous skin, such as the palm of the hand and bottom of the feet. Heat flows from the core to the environment within the blood through arteriovenous anastomoses (direct connections between small arteries and veins) located in the glabrous skin.4 During glabrous skin cooling strategies heat is transferred to the cooling device via these anastomoses resulting in core body temperature decreases.5

Past research has led to differing conclusions on the benefits of pre-cooling, per-cooling, and no cooling and has investigated the effects of various cooling strategies, including upper body, forearm, head, neck, palm, and lower body cooling as well as ingestion of ice or cold water. Several studies have shown that per-cooling on the hands and core is advantageous to the athlete because it reduces sensations of thermal strain and fatigue without compromising the release of heat in the hands due to vasoconstriction or the body’s natural acute inflammatory and metabolic responses to exercise.2,6 Other studies have reported that per-cooling alone, through ingestion and head cooling, was not as beneficial as using it in combination with pre-cooling. 7,8 Multiple studies suggested that continuous per-cooling, through ingestion, fans, shade, and cooling of the neck, thigh, upper body, and trunk had no effect or varying effects on performance, perceptual, and physiological measures, including rate of perceived exertion (RPE), speed of tennis serves, jump performance, total quality of recovery, heart rate, blood lactate, whole body sweat rate, thermal sensations, and body temperatures.6,9 However, another study investigating continuous palm cooling (PC) through cold-water immersion during cycling showed that per-cooling allowed participants to exercise longer.10 These results suggest that per-cooling through glabrous skin, specifically the palms, may be the most effective cooling strategy to enhance performance and decrease negative perceptual outcomes. Prior studies have investigated the ideal temperature in pre- and per-cooling that would not decrease blood circulation while still benefiting performance by decreasing thermal sensations. Studies found that temperatures between 7–15°C appear best for palm per-cooling, whether cooling continuously or during rest, when trying to achieve significant performance and recovery results.10–12 Only a few studies have been done on intermittent cooling techniques, but they suggest that techniques such as half-time and inter-set cooling could attenuate declines in performance from fatigue.11,13

Current evidence supports the use of glabrous skin cooling devices for their potential to delay fatigue and increase the amount of work that can be performed. However, cooling equipment (cooling mitts, Normatec Boots, immersion tubs, ice vests, and climatic chambers) for glabrous skin is generally expensive, bulky, and requires a power source which reduces its accessibility and potential use in team sports. Several commercially available small and portable PC devices, such as the Anti-Fatigue Charge Bar, have been developed to combat this issue, however, research using such devices is limited. Additionally, investigations of short-duration PC specifically on high-level athletes, young adults, and females, is lacking. Thus, the primary aim of this study is to investigate the effects of inter-set PC using a novel cost-effective portable device on repeated sprint ability following a fatiguing protocol in female collegiate athletes. Our findings will further inform coaches and athletes on the potential use of PC as an ergogenic aid during training or competition. Based on prior research we hypothesize that post-fatigue PC, in addition to inter-set PC, will increase the number of sprints completed by delaying early the onset of fatigue.

Methods

A randomized crossover design was used to investigate the effects of PC using a novel cost-effective portable device, on repeated sprint ability in female collegiate athletes following a fatiguing protocol. During the familiarization session, participants were oriented to testing instruments [heart rate (HR) monitors, blood lactate monitors, the OMNI Rate of Perceived Exertion (RPE) scale, the Thermal Perception (TP) Scale, the Feeling Scale (affect), and speed timing gates], the fatigue protocol (modified LIST), and the repeat sprint performance test (RST). Afterwards, participants performed 2–3 trials of a maximum effort 20-m sprint speed test. Following the familiarization session, participants experienced both the control (NC, no cooling) and intervention (PC) sessions in a randomly assigned order. During each trial, measurements of HR, RPE, TP, and affect were recorded before participants began the modified LIST and at each rest period. Blood lactate concentration was measured before the warmup, after the modified LIST, and after the RST. Additionally, the number of sprints performed at or above 90% of participants’ maximal speed during the RST were recorded. All familiarization and testing sessions were conducted at the facilities of Grove City College in the Human Performance Lab and intramural gymnasium in thermoneutral environments. All experimental procedures were approved by the Grove City College Institutional Review Board (protocol #121-2023) prior to implementation. This research was carried out fully in accordance with the ethical standards of the International Journal of Exercise Science.14

Participants

In a recent review by van de Kerkhof et. al,15 pre-cooling strategies yielded positive differences in performance outcomes with a large effect size of 0.74, while per-cooling yielded non-significant positive differences in outcomes with a moderate effect size of 0.45. These results were used to perform a priori power analysis (G*Power 3.1.9.7) using moderate to large effect sizes for a two-tailed t-test, with an error of probability of 0.05 and power of 0.80. This resulted in a suggested sample size of 17 to 41 participants. Following recruitment twenty-nine female participants aged 18–22 volunteered for this study. Two participants were excluded from participation prior to the familiarization session due to irregular menstrual cycles, two participants completed only one experimental session and dropped out prior to completion of the second experimental session, four participants did not complete either experimental session due to timing of their menstrual cycle, and one participant was excluded following both experimental sessions due to inconsistent eating patterns prior to testing, leaving 20 participants for final analysis. Participant demographics are listed in Table 1.

Table 1.

Participant Characteristics (n = 20)

Variable	Mean ± SD
Age (years)	20.0 ± 1.4
Height (cm)	166.7 ± 6.9
Weight (kg)	66.6 ± 9.8
BMI (kg/m 2 )	24.0 ± 3.9
Body fat (%)	24.9 ± 4.6
Fat mass (kg)	16.9 ± 5.6
FFM (kg)	49.7 ± 5.0
Sprint time (s)	3.3 ± 0.2

BMI=body mass index, FFM=fat-free mass, cm=centimeters, kg=kilograms, s=seconds, Sprint time=best 20-meter sprint trial time in seconds

Participants were female varsity athletes with regular menstrual cycles participating in one of the following sports: basketball, volleyball, tennis, lacrosse, soccer, or track and field. Exclusion criteria included 1) the presence of any known cardiovascular, pulmonary, or metabolic disease; 2) current musculoskeletal injury; or 3) presence of any known diseases requiring medical clearance. Participants were instructed to maintain regular training and dietary habits throughout experimental sessions, and to refrain from eating 3 hours prior, consuming caffeine 24 hours prior, and performing any vigorous activity 24 hours prior to a session. Experimental sessions were completed during the first 7 days of the follicular phase of the menstrual cycle with at least 48 hours between sessions.

Protocol

Prior to the familiarization session, participants completed an electronic pre-screening form documenting their sport, current activity level, and menstrual cycle. In the initial familiarization visit, participants received an overview of the experimental study and all experimental protocols prior to providing their written informed consent. Participants then completed a Physical Activity Readiness Questionnaire (PAR-Q) and anthropometric measurements including height (cm), weight (kg), BMI (kg/m²), fat mass (% and kg), and fat free mass (kg). Height was measured using a Detect-Medic Balance Scale and the attached stadiometer (Detecto Scales, Inc., Brooklyn, NY) and was recorded to the nearest 0.5 cm. Body mass (kg) and body composition (fat and lean mass) was measured using a Tanita bioelectrical impedance analyzer (BIA) (TBF-310GS, Tanita Corporation of America, Arlington Heights, Illinois). Body fat was recorded to the nearest 0.1% and 0.1 kg using the athletic setting. Participants were then introduced to all equipment (palm cooling device, OMNI RPE scale, TP scale, affect scale, Lactate Pro Lactate Meter, Polar Heart Rate monitor, and Brower TCI Timing System) and exercise protocols (LIST protocol and sprint protocol) that were used in the experimental sessions. Participants then completed a 5 to 10-minute standardized dynamic warmup before completing one practice bout (~2 min) of the LIST protocol. Following this, participants performed 2 to 4 trials of a maximal velocity 20-meter sprint through the timing system (Brower TCI Timing System). The sprints had to be within 0.2 seconds to ensure participants were completing a maximal effort sprint trial. Participants contacted researchers to schedule their experimental sessions at the beginning of their menstruation phase, and sessions were scheduled within the first 7 days of the follicular phase with at least 48 hours between sessions.

All experimental sessions were completed in the intramural gym on Grove City College campus. The average temperature of the room during sessions was 22°C. At the start of both experimental sessions, capillary blood lactate levels were collected using a finger prick with a lancet following cleaning with alcohol. The first droplet was wiped away with a cotton swab and the subsequent droplet was used for analysis and measured in mmol/L (Lactate Plus, Nova Biomedical). A chest strap HR monitor (PolarH10, Polar Electro, Kempele, Finland) was then affixed just below the participant’s xiphoid process followed by completion of a 5 to 10-min warm up. The warmup consisted of jumping jacks in place for 1-minute, dynamic stretching covering a 10-meter distance, and three acceleration runs to prepare participants for sprinting at full speed. The dynamic stretches included calf scoops, figure fours, high kicks, backwards kicks, quad pulls, knee hugs, side lunges, forward lunges, high knees, and butt kicks. The 20-meter acceleration runs were performed at what participants felt was 50%, 75%, and 100% of their maximal sprint speed. Following the warmup, participants began the modified Loughborough Intermittent Shuttle Test (LIST) protocol which included 3 15-minute blocks of 3x20-meter walking, 1x20-meter sprinting, 3x20-meter running, and 3x20-meter jogging. Participants were given a 3-minute seated rest after each block of the LIST. During these rest periods HR, RPE, TP, and affect were recorded. TP ratings corresponded with the participants’ feelings of body temperature. Room temperature water was available ad libitum during the first experimental session to ensure proper hydration. During the second session participants were provided the same amount of water they consumed in the first session.

Following the final 3-minute rest of the LIST protocol, participants performed a 20-meter maximal RST through the timing gates. Participants were given a 5-second countdown before the start of each sprint. They repeated the sprint test until they were unable to maintain 90% of their maximum speed. Each sprint was followed by 30 seconds of rest.

Experimental condition order was randomly assigned using a counterbalancing technique. During the PC session, participants held the cooling device (Anti-Fatigue Charge Bar, AVA Cooling Technology, Sun Valley ID) with both hands during the final 3-minute rest of the LIST protocol and during the 30-second rest between the 20-meter maximal sprints. The PC device was filled with cold water and placed in a refrigerator in the Grove City College Human Performance Lab to charge prior to experimental sessions. The device was kept at a temperature of 7–15°C during experimental trials by storing it in a portable cooler between use and refilling the device with cold water to re-charge it. The device cap was fitted with a thermometer to provide real time temperature readings. No palm cooling was administered during the NC control session.

Statistical Analysis

Statistical analyses were performed using SPSS version 28. (SPSS Inc., Chicago, IL), with significance set at a priori at p<0.05. Descriptive statistics were calculated for all performance and perceptual variables. Data was tested for normality using the Shapiro-Wilk test. A dependent samples t-test was used to evaluate differences between conditions in the number of repeat sprints. A 2 (condition: PC & NC) x 3 (time: baseline, post LIST, post RST) repeated measures of ANOVA assessed changes in blood lactate. A 2 (condition: PC & NC) x 5 (time: baseline, post LIST block 1, post LIST block 2, post LIST and post RST) repeated measures ANOVA assessed differences in heart rate between conditions, and a 2 (condition: PC & NC) x 4 (time: post LIST block 1, post LIST block 2, post LIST and post RST) repeated measures ANOVA assessed changes in RPE, TP, and affect between conditions. Bonferroni post hoc assessments were used to examine significant main and interaction effects. The assumption of sphericity was confirmed using Mauchly’s test. Greenhouse-Geisser epsilon corrections were used when the assumption of sphericity was violated. Effect sizes were calculated using Cohen’s d (d: small=0.2, medium=0.5, & large=0.8), partial eta squared (h_p²: small=0.01, medium=0.06, & large=0.14).

Results

RST

Figure 1. shows the number of successful repeat sprints completed by each participant and the average completed per experimental condition. There was a statistically significant difference in the mean number of successful repeat sprints performed after the fatiguing protocol between PC and NC (10.3±12.9 vs. 6.0±8.4, p=0.025, d=0.546).

Figure 1.

Number of repeat sprints performed during each condition (n=20). Values are mean ± SD. PC=palm cooling; NC=no cooling; *=statistically significant difference between conditions (p=0.025).

Blood Lactate

There was no significant interaction between condition and time observed for blood lactate levels (F=0.425; p=0.657, h_p²=0.023) nor was there a significant main effect for condition (F=0.669; p=0.424, h_p²=0.036). There was a significant main effect for time (F=37.915; p<0.001, h_p² =0.678). Bonferroni post hoc analysis showed that the blood lactate levels significantly increased from pre-LIST (1.2±0.1) to post-LIST (3.9±0.4) (p<0.001) and from pre-LIST to post-RST (4.8±0.5) (p< 0.001). There was no significant difference in blood lactate levels between post-LIST and post- RST (p=0.140).

Figure 2.

Blood lactate (mmol/L) measured pre-LIST, post-LIST, and post-RST (n=20). Values are mean ± SD. LIST=Loughborough Intermittent Shuttle Test; RST=repeat sprint test; PC=palm cooling; NC=no cooling; *=Pre- LIST blood lactate significantly less than Post-LIST and Post-RST (p<0.001).

Heart Rate

There was no significant interaction (F=0.378; p=0.823, h_p²=0.020) or main effect for condition (F=0.029; p=0.867, h_p²=0.002) for HR. There was a significant main effect for time (F=350.557; p<0.001, h_p²=0.949). Bonferroni post hoc analysis showed that HR pre-LIST (79.2±1.6) was significantly lower than all other timepoints (p<0.001), with no significant differences elsewhere (p>0.05).

RPE

There was no significant interaction (F=3.142; p=0.061, h_p²=0.142) or main effect for condition (F=0.231; p=0.636, h_p²=0.012) for RPE. There was a significant main effect for time (F=53.116; p<0.001, h_p²=0.737). Bonferroni post hoc analysis showed that RPE post-LIST block 1 (6.1±0.2) was significantly lower than all other timepoints (p<0.001). RPE post-LIST block 2 (7.6±0.2) was significantly lower than post-LIST (8.3±0.1, p<0.001) and post-RST (8.7±0.2, p=0.002). There were no significant differences in RPE between post-LIST and post-RST (p>0.550).

Thermal Perception

There was no significant interaction (F=1.51; p=0.872, h_p²=0.008) or main effect for condition (F=1.717; p=0.206, h_p²=0.083) for TP. There was a significant main effect for time (F=9.122; p<0.001, h_p²=0.324). Bonferroni post hoc analysis showed that TP post-LIST block 1 (1.9±0.1) was significantly lower than post-LIST block 2 (2.3±0.1; p<0.001) and post-LIST (2.5±0.1, p=0.003) with no difference between post-RST (2.25±0.1, p=0.084). There were no significant differences in TP between post-LIST block 2, post-LIST, and post-RST (p≥0.342) (Figure 3).

Figure 3.

Thermal Perception (−3 to +3) measured post-LIST block 1, block 2, block 3, and RST (n=20). Values are mean ± SD. LIST=Loughborough Intermittent Shuttle Test; RST=repeat sprint test; PC =palm cooling; NC=no cooling; *=Pre-LIST block 1 significantly less than post-LIST block 2 (p<0.001) and block 3 (p=0.003).

Affect

There was no significant interaction (F=1.794; p=0.159, h_p²=0.086) or main effect for condition (F=0.100; p=0.755, h_p²=0.005) for affect. There was a significant main effect for time (F=19.936; p<0.001, h_p²=0.512). Bonferroni post hoc analysis showed that affect post-LIST block 1 (1.5±0.3) was significantly higher than post-LIST block 2 (0.6±0.3; p<0.001), post-LIST (0.2±0.3, p<0.001), and post-RST (−0.2±0.4, p<0.001). Post-LIST block 2 was significantly higher than post-LIST (p=0.035) and post-RST (p=0.035) with no difference between post-LIST and post-RST (p=0.385).

Discussion

The purpose of our study was to investigate the effects of inter-set PC using a novel cost-effective portable device on repeated sprint ability following a fatiguing protocol in female athletes. Based on past research regarding short term recovery on sprint performance, it was hypothesized palm cooling would increase the number of sprints completed by delaying early onset fatigue. To our knowledge, this is the first study investigating the effects of palm cooling on repeat sprint performance in females. Cooling was conducted both post fatigue and between sprint bouts, creating a unique blend of per-cooling techniques. Similar to previous work, the physiology of local working muscles was not impacted by cooling just the palms.16 Our results showed that this blend of per-cooling palm cooling strategies significantly improved repeat sprint performance, allowing athletes to complete roughly 4 more successful sprints at or above 90% of their top speed than the control condition. No differences in perceptual (RPE, TP, or affect) or physiological variables (HR and blood lactate) were observed between conditions. The portable and practical nature of the cooling device used in this study makes it highly applicable to athletic settings, particularly for team sports requiring repeated sprints. Its cost-effectiveness and ease of use during rest periods could enhance accessibility for athletes and teams lacking resources for traditional cooling equipment. These results and the aforementioned findings reinforce the benefits of cooling regions of the body distal to the working muscles used in exercise.16

Previous research investigating cooling techniques excluded female participants due to differing cold sensitivity than males, as well as the effects of the menstrual cycle on core temperature and temperature sensation. Females have been found to have a significantly higher cold sensitivity than males on some sites, including the middle part of the top of the palm and the medial area at the base of the palm.17 Other research found that females are more susceptible to changes in core temperature and benefits from cooling techniques.18 In particular, one study found that the time to exhaustion of participants was significantly longer in the follicular phase of the menstrual cycle than the luteal phase when exercising in hot, humid conditions; however, the phase of the menstrual cycle did not affect the performance of the participants when in temperate conditions.19 In our study, the phase of the cycle was controlled for to limit the impact that differing thermoregulation could have during different phases of the cycle.

The temperature of cooling devices and the duration of cooling appear to be critical to these strategies. Previous research has investigated cooling techniques at a range of temperatures between 5–24°C. Collectively, current evidence suggests that temperatures ranging from 8°C to 15°C is optimal for improving performance with 8°C appearing to be the lowest effective temperature without restricting blood flow.10,11,20 In our study, the temperature of the PC devices was successfully maintained between 7.2 and 15.6°C during trials. Regarding the duration of cooling, several approaches have been investigated including short (70 seconds), moderate (150–180 seconds), and long (5 minutes or more) interventions.9,21,22 Existing evidence appears to suggest an optimal duration of between 90 and 180 seconds .9,21 Consistent with these recommendations we were able to apply a 3-minute cooling period following the LIST protocol. The 30 second cooling periods between repeat sprint bouts was employed to help maintain the cooling effect attained prior to the RST.

A primary limitation of this study is the lack of core body or skin temperature assessments. Yanoaka et al observed meaningful increases (Δ ~ 1.6°C) in core body temperature measured by ingestible telemetric pill during the LIST protocol completed in a thermoneutral environment.23 Additionally, research has demonstrated significant positive correlations between rising blood lactate levels and body temperature during exercise.24,25 Thus, it is reasonable to assume that similar increases in body temperature could have occurred with our participants. Based on previous research we speculate that PC could have mitigated the rise in core body temperature following the LIST protocol, potentially delaying fatigue, and allowing for significantly more successful high-speed sprints to be performed.5,23 Although there were no statistically significant differences in TP, the PC condition resulted in lower TP values at each time point measured, supporting this suggested mechanism of improvement. The TP scale used here, although valid and reliable, only uses a range of −3 to +3. This may hinder its sensitivity, especially in thermoneutral environments, therefore limiting our potential to observe significant differences between conditions. Finally, we cannot rule out the potential influence of a placebo/belief effect. An athlete’s belief in a substance or piece of equipment might be a significant contributing factor influencing the perception of their ability to perform a given task26 which may explain why placebo/belief effects elicit benefits in up to 35% of cases in research.27 As an example, prior research displayed greater muscular strength and improvements in psychological measures in a cold-water immersion group than a thermoneutral-water immersion group but observed similar performance when compared to a thermoneutral-water placebo group who were manipulated to believe that thermoneutral-water was also ergogenic.28 We recommend future research involving palm cooling attempts to directly or indirectly assess core body temperature changes as well as belief effects to better understand the potential mechanisms for any observed performance improvements.

To our knowledge, this is the first study investigating the effects of PC on repeat sprint ability in female athletes using a unique blend of per-cooling throughout the experimental sessions. PC for 3-minutes following a fatiguing protocol and during an RST with an affordable, portable device resulted in significantly more repeat sprints than NC in a thermoneutral environment. These findings align with previous research and suggest that palm cooling may be an effective strategy for optimizing performance during training or competition. The portable and practical nature of the cooling device used in this study makes it highly applicable to athletic settings, particularly for team sports requiring repeated sprints. Its cost-effectiveness and ease of use during rest periods could enhance accessibility for athletes and teams lacking resources for traditional cooling equipment. These findings highlight the need for further research on accessible cooling technologies and their application across genders, sports, and environments.