Throughout time, athletes have sought out ways of becoming faster, stronger and more powerful. In order to reap the benefits of these physiological gains, it is important that athletes have optimum recovery post competition, often in short periods of time between competitions and training, depending on the sport. A technique that is gaining momentum is cold water immersion (CWI) (Bleakley and Davison, 2011). According to Wilcock et al. (2006), CWI is defined as being immersed in water temperature of less than 15 degrees Celsius. This is based on the athlete’s earliest signs of pain in past studies (Low and Reid, 1994). Though more recent research has used water temperatures as low as 8 degrees Celsius (Gill et al. 2006; Ingram et al. 2009) and 5 degrees Celsius (Sellwood et al. 2007).
However, the current body of literature remains open to debate and interpretation (Schniepp et al. 2002; Gill et al. 2006; Yamane et al. 2006; Crowe et al. 2007; Sellwood et al. 2007; Ingram et al. 2009 and Montgomery et al. 2008). And yet, there does not appear to be evidence for an optimal treatment procedure for CWI (Cochrane, 2004). Nonetheless, a number of different methods in the literature have been used. For example, combining cold and hot water intervals (Cochrane, 2004) and time intervals, 3 minutes in cold water followed by 3 minutes out of the water.
The rationale for using CWI according to the literature is that it has been found to reduce delayed onset of muscle soreness (DOMS) (Leeder et al. 2012). This includes physiological changes like intracellular-intravascular fluid shifts, a decline in muscle oedema and an increase in cardiovascular output which are evident without additional energy expenditure (Wilcock et al. 2006). This could be a potential benefit to the athlete as glycogen stores will already be depleted post activity or competition, thus being able to take part in recovery sessions that do not task the energy systems. This will help aid and speed up recovery, as well as the removal of lactic acid and the transportation of nutrients through the bloodstream (Wilcock et al. 2006). Conversely to this, a study by Tikuisis et al. (2000) found that being immersed neck deep in water temperature of 18 degrees Celsius increased the metabolic rate. This is due to shivering which occurs differently based on the square foot of body fat on the participants (Tikuisis et al. 2000). However, in Tikuisis et al.’s (2000) study participants were immersed for 90 minutes compared to Wilcock et al. (2006) who examined studies where participants were immersed for between 5 minutes and a maximum of 20 minutes.
In another study looking at the difference in recovery between CWI and thermoneutral water immersion in soccer players, Ascensao et al. (2011) found that CWI immediately after a one off soccer match reduces muscle damage and discomfort. The study also found CWI and thermoneutral groups had differential changes in creatine kinase in 30 minutes, 24 hours and 48 hours. There were similar changes in myoglobin after 30 minutes and in C-reactive protein after 30 minutes, 24 hours and 48 hours. There were also muscular adjustments in quadriceps strength after 24 hours, gastrocnemius after 24 hours and adductors after 30 minutes (Ascensao et al. 2011). This is further replicated by Leeder et al. (2012) who completed a meta-analysis (Fig. 1) across a number of studies comparing CWI and control of measures of DOMS. This is reporting a confidence interval of 95% for the speed of recovery of the lower limbs and upper limbs in favour of CWI.
In regards to injury recovery and reconditioning, CWI has been used after acute musculoskeletal injuries. This is to encourage vasoconstriction and decrease the initial inflammatory response generated by the body when an injury occurs (Goodall and Howatson. 2008). Finally, chronic effects of CWI on training adaptation in Yamane et al.’s (2006) study found that the control group showed substantial training effects compared to the cold group, deeming cooling as ineffective. Therefore, more research needs to be completed on training adaptations.
The aim of this study is to determine the effect CWI has on performance on repeated cycling sprints in the heat with university studies based at a university in the West Midlands, England. The hypothesis is that sprint peak power performance should improve after CWI. The second hypothesis using the recovery scale is that the participants should recover at a faster rate after being immersed in cold water.
The test protocol used 24 participants from the University respectively, testing involved 10 x 6 second cycling sprints with 24 seconds active recovery in an ambient temperature environment of 35-38 degrees Celsius. Following the sprints the participants were seated in cold water at 10 degrees Celsius for 10minutes before repeating the 10 x 6 seconds sprints again.
All specific procedures were followed in accordance with the University laboratory manual and all participants voluntarily gave consent to take part in the testing.
The data collected was, the participants peak power output, mean power output, fatigue index percentage and the rating of perceived exertion (RPE) (Rowsell et al., 2011) during the repeated sprints, as well as recovery rate scale. The results were compared between time points and conditions. All recorded data was entered into the Statistical practices of the social sciences (SPSS) where a repeated measures ANOVA and an independent samples T-test were completed. Size effect and Cohen’s d was calculated using www.uccs.edu/lbecker/.
Peak Power Watts
Peak power watts (W) was measured in a group of students while completing cycle sprints pre and post cooled and pre and post rested. The mean and Standard deviation (SD) for the pre cooled was 1016±286 peak power W and post cooled 1021±255 peak power W (Graph.1), with an effect size of 0.09 and Cohen’s d of 0.01. This shows that peak power watts increased after the cooled trial. The mean and SD for the pre rest was 1014±300 peak power W and post rest 930±344 peak power W (Graph. 1), with an effect size of 0.12 and Cohen’s d of 0.26. This shows that the peak power decreased after having a rest post sprint suggesting fatigue. The repeated measures ANOVA reported that there was no significant difference between the pre and post cooled results for peak power W, P=.606. Moreover, there was also no significant difference between the pre and post rest groups P=.558. (Fig. 3). The confidence interval for the peak power in watts for pre cooled was a mean of 1016.2W. The confidence interval for peak power watts after the participants sat in the cold water increased to 1021.5W (Fig. 11). This further suggests that the peak power watts generated by the participants increased. The confidence interval for peak power watts for pre rest was 1014.5W after rest the confidence interval decreased to 930.6W (Fig. 11). Furthermore, suggesting that the peak power watts decreased after rest in the cycle sprint trial.
Graph. 1. Peak power watts produced pre cooled/rest and post cooled and rest.
Mean Power Watts
Mean power W mean and SD for pre cooled was 710±200 and post cooled 776±194 with pre rest 709±210 and post rest 623±173 respectively (Graph. 2). The effect size was 0.16 and the Cohen’s d was 0.03 for pre and post cooled peak mean W. This suggests a significant difference according to the Cohen’s d for mean peak power watts between the pre and post cooled trial. For pre and post rest peak mean W the effect size was 0.21 and the Cohn’s d was 0.44. However, the results from the repeated measures ANOVA show that there was no significant difference between the pre and post cooled results, for mean peak W P=.820 and the rest pre and post results P=.115 (Fig. 5). The confidence interval for the mean pre power cooled was 710.9W, the post cooled mean power increased to 776W (Fig. 12). The confidence interval for the mean power pre rest was 709.6W, post rest this decreased to 623.4W (Fig. 12).
Graph. 2. Mean Power W pre cooled/rest and post cooled/rest.
The percent fatigue index (%IF) mean and SD for pre cooled was 57.8±14.2 and post cooled was 64.5±12.4. The pre rest %IF was 57.9±6.5 and post rest was 69.9±11.8 (Graph. 3). This suggests that even though participants fatigued in both trials, the cooled trial fatigued less than the rest trial, with fatigue increasing by 6.7% for cooled and 12% rest. The effect size for pre and post cooled %IF was 0.24 and the Cohen’s d was 0.50. The effect size for pre and post rest %IF was 0.53 and the Cohen’s d was 1.25. The results from the repeated measures ANOVA show that there is a significant difference between the pre and post cooled tests P=.017. However, there was no significant difference pre and post rest for %IF tests P=.476 (Fig. 7).
Graph. 3. %IF pre cooled/rest and post cooled/rest
Rate of Perceived Exertion
The rate of perceived exertion (RPE) mean and SD for pre cooled was 16.5±1.16 and post cooled was 16.4±.90. The pre rest RPE mean and SD was 16.5±1.38 and post RPE was 18.5±.90 (Graph. 4). This advises that the RPE that participants felt after the cooled trial decreased by 0.01 point and after the rested trial increased by 2 points. The effect size for pre and post RPE was 0.04 and the Cohen’s d was 0.09. Which suggests a significant difference between the cooled and rest trials. The confidence interval for RPE pre cooled was 16.5 the post was 16.4. The confidence interval for RPE pre rest was 16.5 and post 18.5 (Fig. 15). According to the repeated measures ANOVA there was a significant difference between the RPE pre and post cooled tests P=.008. This indicates that participants didn’t feel that they were working as hard in the post cooled trial. Moreover, the repeated measures ANOVA reported a significant difference between RPE pre and post rest P=.002 (Fig. 9). Highlighting that the participants felt that they had to work harder in the post rest trial.
Graph. 4. RPE pre cooled/rest and post cooled/rest
The mean and SD for recovery between cooled and rest was, cooled 7.25±1.13 and rest 4.91±.79 (Graph. 5). According to the perceived rate recovery status (RPS), this shows that the participants recovered faster after the cooled trial compared to the rest trial (Fig. 16). The effect size for recovery between the cooled and rest group was 0.76 and the Cohen’s d was 2.40. The confidence interval for recovery cooled and rest was 2.33. The results from the independent t-test shows a strong significant difference between recovery in the cooled and rest groups of p=.000 (Fig. 14). This advocates that the participants after being subjected to CWI recovered faster than when the participants just had a rest.
Graph. 5. Recovery scale after cooled and rest trials
The key results from this study suggests that the participants produced more peak power watts in the sprint tests after 10 minutes of CWI. However, this was only by a trivial amount. This suggests that the power applied by the participants after CWI increased. This would suggest that CWI acted specifically to recovery muscular power (Peiffer et al. 2010) which would suggest an improvement in sprint performance, or that CWI encourages type II muscle fibres to recover faster (Leeder et al. 2011). According to Leeder et al. (2011) an explanation for this is that type II muscle fibres are preferentially damaged after eccentric contraction exercises and are the prime fibre type during high velocity muscular contraction. Thus encompassing raised power production (Friden and Lieber, 2001). However, currently this is only speculated and requires further investigation (Leeder et al. 2011). Peiffer et al. (2010) suggests a different alternative that the increase in power could be due to a lowered perception of fatigue resulting in a faster pedal rate. This is due to the body temperature being cooler after being immersed in cold water. (Nybo et al. 2001).
Contrary to this, Delextrat et al. (2012) and Montgomery et al.’s (2008) studies found that there was no improvement in repeated sprint performance. However, participants were tested on running repeated sprints 24 hours after CWI. Compared to this study where the sprints were performed pre and post immersion on a cycle.
In the current study the participants also decreased in watts following just a rest break between sprints according to the confidence interval. The Repeated measures ANOVA however did not highlight this as a significant difference, though, the Cohen’s d did suggest a significance for the peak power watts between the sprint tests pre and post cooled. This was also the case for the mean peak power watts results.
The repeated measures ANOVA, Cohen’s d and effect size discovered a significant difference between the pre and post cooled %IF, pre and post cooled RPE and pre and post rest RPE. RPE decreased after CWI, but increased after the rest trial. This would suggest that the participants felt that they were working harder after the rest trial compared to after CWI. These results were also found in Peiffer et al.’s (2010) study.
The results from the independent t-test found a strong significant difference in recovery between the cooled and rest groups, with the rate of perceived recovery scale (RPS) (Fig. 16) highlighting that the participants recovered faster in the cooled trial compared to the rest trial. This agrees with the outcome of a number of studies in the literature (Delextrat et al. 2012; Montgomery et al. 2008; Peiffer et al. 2010).
There are a number of underlying mechanisms, for example, CWI is thought to be used to encourage a reduction in muscle blood flow and tissue temperature, which results in decreased inflammation from intense exercise (Gregson et al. 2011). This has shown to decrease femoral artery blood flow greater than 40%, and muscle temperature 2-4 degrees Celsius after being immersed in 8 degrees Celsius and 22 degrees Celsius for 10 minutes (Leeder et al. 2011). This proposed reduction in inflammation is thought to be linked with a decrease in the feeling of pain through reducing osmotic pressure of exudate (Swenson et al. 1996; Friden and Lieber, 2001 and Meeusen and Lievens, 1986). However, it is suggested that inflammatory response is critical for optimal repair of damaged tissue, so, CWI maybe unfavourable. Though, there is a lack of evidence to support this (Leeder et al. 2011).
Since CWI came onto the scene there has been a number of methods used to find a specific guideline for optimum recovery (Schniepp et al. 2002; Gill et al. 2006; Yamane et al. 2006; Crowe et al. 2007; Sellwood et al. 2007; Ingram et al. 2009 and Montgomery et al. 2008), yet to date these are inconclusive (Cochrane, 2004).
Numerous studies have suggested that athletes will recovery faster after CWI compared to other methods like warm water immersion, massage and compression garments (Delextrat et al. 2012; Montgomery et al. 2008). Although both studies found that CWI did not have an effect on repeated sprint ability, they both did suggest that CWI promotes better restoration and quicker recovery times compared to other recovery performance measures garments (Delextrat et al. 2012; Montgomery et al. 2008).
However, in respect of chronic training adaptation Yamane et al. (2006) found that CWI was ineffective in encouraging molecular and humoral changes linked with specific training effects like muscular hypertrophy. Jakeman et al. (2009) also found that CWI had no beneficial effects on recovery from plyometric exercise induced muscle damage after 24, 48, 72 and 96h. Nonetheless, more training studies are needed to explore the effects CWI has on training adaptation and performance (Versey et al. 2013).
The present study has highlighted that CWI has potential benefits on recovery speed and RPE between cycle sprints compared to rest, as well as a reduced %IF. This would suggest that CWI can be used for high intensity intermittent exercise. Though more studies need to be investigated with athletic populations. In respect of practicality, CWI allows for full body immersion, however, many sports teams or athletes may not have the facility for this or the funding to pay the cost of using a specialised pool. Therefore, a potential cheaper and more environmentally productive alternative could be the use of cooling vests that can be worn individually by the athletes at any time (Hausswirth and Mujika, 2013). This would be very useful during training in hotter climates like the Brazilian Olympics in 2016 and the Qatar FIFA World Cup in 2022.
The participants demonstrated a faster recovery in the cooled group compared to the rest group, they also had significantly less fatigue following the CWI in the second sprint cycle trial compared to the rest trial. This would agree with hypothesis 1 that recovery rate improved following CWI. The participants also reported less RPE after cooling compared to rest for the second trial. Although the Repeated measures ANOVA reported no significant difference in peak power watts, the Cohen’s d reported a strong significant difference in peak power W output for the sprint cycle trial following CWI. This would agree with hypothesis 2 that peak power W sprint performance increased. These results agree with the literature that CWI improves recovery of muscle power (Peiffer et al. 2010; Leeder et al. 2011; Friden and Lieber, 2001 and Nybo et al. 2001). Though, there is no obvious explanation for this and requires further exploration, potentially using Electromyography (EMG) may give some light to this.
It is also concluded that more research needs to be completed on CWI effects on chronic training adaptations as Peiffer et al. (2010), speculated that CWI could be linked with heightened type 2 muscle fibre recovery, following strenuous exercise. Yet, Jakeman et al. (2009) Delextrat et al. (2012) and Montgomery et al. (2008) found no such benefit following plyometric training and repeated sprint training, which would recruit type 2 muscle fibres.
Finally, CWI has been found to reduce the onset of DOMS post exercise and inflammation post injury (Swenson et al. 1996; Friden and Lieber, 2001 and Meeusen and Lievens, 1986). However, one has to air on the side of caution here because inflammation after an injury is the body’s natural inflammatory response to protect the area. Thus, is it actually a benefit that CWI reduces inflammation (Leeder et al. 2011)? Further clarification into this is required in order to set out guidelines of best practice for practitioners.
Ascensao, A. Leite, M. Rebelo, A. N. Magalhaes, S and Magalhaes, J, (2011). Effects of cold water immersion on the recovery of physical performance and muscle damage following a one-off soccer match. Journal of Sports Science. 29(3): 217-225.
Bleakley, C M and Davison, G. W, (2011). What is the biochemical and physiological rationale for using cold-water immersion in sports recovery? A systematic review. British Journal of Sports Medicine, 44; 179-187.
Cochrane, D. J (2004). Alternating hot and cold water immersion for athlete recovery: a review. Physio Therapy of Sport. 5; 26-32.
Crowe, M. J. O’Connor, D. Rudd, D, (2007). Cold water recovery reduces anaerobic performance. International Journal of Sports Medicine. 28; 994-8.
Delextrat, A. Calleja-Gonzalez, J. Hippocrate, A and Clarke, N, (2012). Effects of sports massage and intermittent cold-water immersion on recovery from matches by basketball players. Journal of Sports Sciences. 1-9.
Friden, J and Lieber, R. L, (2001). Eccentic exercise-induced injuries to contractile and cytoskeletal muscle fibre components. Acta Physiologica Scandinavica. 171; 321-6.
Gill, N. D. Beaven, C. M and Cook, C, (2006). Effectiveness of post-match recovery strategies in rugby players. British Journal of Sports Medicine. 40; 260-3.
Goodall, S and Howatson, G, (2008). The effects of multiple cold water immersions on indices of muscle damage. Journal of Sports Science and Medicine. 7; 235-241.
Gregson, W. Black, M. A. Jones, H et al. (2011). Influence of cold water immersion on limb and cutaneous blood flow at rest. American Journal of Sports Medicine. 33; 483-503
Hausswirth, C and Mujika, I, (2013). Recovery for performance in sport. Champaign IL, Human Kinetics
Ingram, J. Dawson, B. Goodman, C et al. (2009). Effect of water immersion methods on post-exercise recovery from simulated team sports exercise. Journal of Science Medicine Sport. 12; 417-21.
Jakeman, J. R. Macrae, R and Eston. R. (2009). A single 10-min bout of cold-water immersion therapy after strenuous plyometric exercise has no beneficial effect on recovery from the symptoms of exercise-induced muscle damage. Journal of the chartered institute for Ergonomics and Human Factors, 52; 4.
Leeder, J. Gissane, C. Someren, K. V. Gregson, W and Howatson, G. (2012). Cold water immersion and recovery from strenuous exercise: a meta-analysis. British Journal of Sports Medicine. 2-8.
Low, J and Reid, A, (1994). Electrotherapy explained: Principles and Practice, 2nd Edition, Oxford. Butterworth and Heinemann.
Meeusen, R and Lievens, P. (1986). The use of cryotherapy in sports injuries. Journal of Sports Medicine. 36; 398-414.
Montgomery, P. G. Pyne, D. B. Hopkins, W. G et al. (2008). The effect of recovery strategies on physical performance and cumulative fatigue in competitive basketball. Journal of Sports Science. 26; 1135-45.
Nybo, L and Nielsen, B, (2001). Perceived exertion is associated with an altered brain activity during exercise with progressive hyperthermia. Journal of Applied Physiology. 91; 2017-23.
Peiffer, J. J. Abbiss, C. R. Watson, G. Nosaka, K and Laursen, P. B. (2010). Effect of a 5-min cold-waer immersion recovery on exercise performance in the heat. British Journal of Sports Medicine. 44; 461-465.
Rowsell, G.J. Coutts, A. Reaburn, P and Hill-Haas, S. (2011). Effect of post-match cold-water immersion on subsequent match running performance in junior soccer players during tournament play. Journal of Sports Science. 29(1); 1-6.
Schniepp, J. Campbell, T. S, Powell, K. L et al. (2002). The effects of cold-water immersion on power output and heart rate in elite cyclists. Journal of Strength and Conditioning Research, 16; 561-6.
Sellwood, K. L. Brukner, P, Williams, D et al. (2007). Ice-water immersion and delayed onset muscle soreness: a randomised controlled trial. British Journal of Sports Medicine. 12; 392-7.
Swenson, C. Sward, L and Karlsson, J. (1996). Cryotherapy in sports medicine. Scandinavian Journal of Medicine and Science Sports. 6; 193-200.
Tikuisis, P. Jacobs, I. Moroz, D. Vallerand, A. L and Martineau, L. (2000). Comparison of thermoregulatory responses between men and women immersed in cold water, Journal of Applied Physiology, 89; 4, 1403-1411.
Versey, N. G. Halson, S. L and Dawson, B. T. (2013). Water immersion recovery for athletes: Effect on exercise performance and practical recommendations, Journal of Sports Medicine, 43; 1101-1130.
Wilcock, I. M. Cronin, J. B and Hing, W. A (2006). Physiological response to water immersion: a method for sports recovery? Journal of Sports Medicine, 36; 747-65.
Yamane, M. Teruya, H. Nakano, M et al. (2006). Post-exercise leg and forearm flexor muscle cooling in humans attenuates endurance and resistance training effects on muscle performance and on circulatory adaptation. European Journal of Applied Physiology. 96; 572-80.
Fig. 1. Meta-analysis for cold water immersion recovery times throughout studies over 24, 48, 72 and 96 hours post exercise.
Fig 2. Mean and Standard deviation for pre and post cooled and rested trials.
Fig. 3. ANOVA of peak power pre post cold water immersion and rest
Fig. 4. Mean peak watts mean and standard deviation for pre and post cooled and rest.
Fig. 5. Repeated measures ANOVA for mean peak watts for pre and post cooled and rest.
Fig. 6. Mean and SD for %IF pre and post cooled and rest.
Fig. 7. Repeated measures ANOVA for %IF pre and post cooled and rest.
Fig. 8. Mean and SD for Rate of perceived exertion Cooled and rest.
Fig. 9. Repeated measures ANOVA results for RPE cooled and rest.
Fig. 10. Confidence interval for recovery
Fig. 11. The confidence interval for Peak power W pre and post cooled and rest.
Fig. 12. The confidence interval for Mean power W pre and post cooled and rest.
Fig. 13. Recovery between cooled and rest groups mean and SD.
Fig. 14. Recovery between cooled and rest tests independent t-test.
Fig. 15. Confidence interval for RPE pre and post cooled and rest
Fig. 16. Rate of perceived recovery chart; Laurent, C. M; Green, J. M; Bishop, Phillip A; Sjökvist, J; Schumacker, R. E; Richardson, M. T; Curtner-Smith, M. (2011). A Practical Approach to Monitoring Recovery: Development of a Perceived Recovery Status Scale. Journal of Strength & Conditioning Research. 25 (3); 620-628.