Many intermittent team sports, such as football and rugby, are played over 2 consecutive periods separated by a break at half-time. During the half-time, players are typically passive, and they engage in tactical debriefing and video analyses, receive medical attention, if needed, and ensure adequate nutritional replenishment (22). However, numerous studies have reported declines in the amounts of the high-intensity running and sprinting, which are important indicators of performance in players of intermittent team sports (17), after traditional passive half-time practices in thermoneutral environments (3).
In previous studies, a re–warm up (RW) of short duration at half-time involving approximately 7 minutes of moderate-intensity exercise has been shown to attenuate the losses in core and muscle temperatures (Tm) during half-time and avoid reductions in performance of sprint, vertical countermovement jump, and Yo-Yo intermittent recovery test immediately after half-time (12,13,19,21,31). For example, Russell et al. have suggested that RW attenuated the loss in core temperature by 0.5° C and increased repeated shuttle sprint ability and peak power output during countermovement jumps immediately after half-time (21). However, a recent review highlighted that the players may not be able to complete the RW of 7 minutes used in previous studies since they typically have approximately 3 minutes available for an RW in actual matches (22). Indeed, only 58% of the fitness coaches and sport scientists in the Premier League and Championship divisions of professional football instructed players to engage in RW either on the pitch or within the facilities of the venue, and 63% of them reported that one of the major situational limiting factors was a lack of time (28). Therefore, it is necessary to evaluate the effectiveness of very-short-duration RW (i.e., 3 minutes) during half-time in increasing the intermittent exercise performance after half-time.
A recent review regarding warm-up for precompetition preparation indicated that the efficacy of a warm-up is determined by its intensity and duration, as well as the time between the completion of the warm-up and start of the competition (15). Regarding intensity, a low- and moderate-intensity warm-up performed with no recovery time can be effective in improving the sprint and intermittent sprint performance as it results in little depletion in phosphocreatine and increases Tm (1,15). In fact, a previous study has demonstrated that a low-intensity warm-up (i.e., 39% of maximal oxygen uptake [V[Combining Dot Above]O2max]) before competitions increased the subsequent cycling sprint performance compared with a moderate-intensity (i.e., 56% of V[Combining Dot Above]O2max) or high-intensity (i.e., 74% and 80% of V[Combining Dot Above]O2max) warm-up (23). Wittekind et al. similarly reported that a low-intensity (i.e., 40% of peak aerobic power) warm-up was more effective in increasing the cycling sprint performance than was a high-intensity (i.e., 110% of peak aerobic power) warm-up (29). However, it is not known whether RW has similar, if any, benefits in increasing the subsequent exercise performance (23,29). In addition, although the effectiveness of moderate-intensity RW in preventing a decrement in exercise performance has been addressed in previous studies (12,13,19,21,31), to the best knowledge of the authors, there has been no study that investigated the effects of low-intensity RW on subsequent exercise performance.
Therefore, the purpose of this study was to investigate the effects of very-short-duration RW (i.e., 3 minutes) on subsequent intermittent sprint performance and compare the effects of low- and moderate-intensity RWs on subsequent intermittent sprint performance. We hypothesized that both low- and moderate-intensity RWs would increase the intermittent sprint performance, and that the low-intensity RW would have the same influence on intermittent sprint performance as moderate-intensity RW because previous studies have suggested that low- and moderate-intensity “warm-up” can increase intermittent sprint performance (1,15).
Experimental Approach to the Problem
Participants completed 3 trials in a randomized order. In the experimental trials, 2 consecutive intermittent exercises separated by a 15-minute half-time were performed on a cycle ergometer (Monark 894E; Monark, Varberg, Sweden). The interventions at half-time included 15 minutes of seated rest for the control trial and 3 minutes of cycling at 60 and 30% of V[Combining Dot Above]O2max in others. Exercise performance after half-time was evaluated using a Cycling Intermittent-Sprint Protocol (CISP) that has been described previously to assess intermittent sprint performance in athletes of intermittent team sports (10). A high reliability of work during sprints in the CISP has previously been reported (i.e., intraclass correlation = 0.9) (10).
All trials were separated by at least 3 days. Participants performed 1 practice trial to familiarize themselves with the experiment at least 3 days before the first experimental trial. The mean temperature and humidity in the experimental trials were 20.6 ± 0.5° C and 50.8 ± 1.4% (mean ± SD), respectively.
Eleven healthy men who trained (i.e., more than an hour per session) for more than 2 days per week (mean ± SD: age: 22.7 ± 2.4 [range, 20–28] years; height: 1.73 ± 0.06 m; body mass: 65.3 ± 10.0 kg; and V[Combining Dot Above]O2max: 52.6 ± 6.2 ml·kg−1·min−1) participated in this study. This study was approved by the Ethics Review Committee on Research with Human Subjects of Waseda University (Approval number: 2017-048), and all participants provided written informed consent to participate in this study.
Participants were asked not to alter their regular lifestyle habits, exercise, and diet throughout the study. Participants recorded all the meals and drinks consumed for 24 hours before each experimental trial and replicated their dietary intake in subsequent trials, ensuring that they were standardized across trials. Participants refrained from intake of alcohol and caffeine for 24 hours before each experimental trial. Furthermore, participants fasted for 3 hours, except for consumption of water, before each experimental trial. All 3 experimental trials were performed at the same time of day to avoid any circadian rhythm-related variation in the obtained results.
Participants initially underwent a graded exercise test to determine their V[Combining Dot Above]O2max and maximum heart rate (HRmax) on a cycle ergometer (Monark 894E; Monark). The test started at 40 W with a target cadence of 80 rpm and increased by 40 W every 2 minutes until volitional exhaustion. Breath-by-breath gas analysis was performed using an automatic gas analyzer (Quark CPET; COSMED, Rome, Italy). Heart rate was measured using a wireless HR monitor at 5-second intervals (Polar RCX3; Polar Electro, Kempele, Finland).
In the experimental trials, the participants performed 2 consecutive intermittent exercises separated by a 15-minute half-time on a cycle ergometer (Monark 894E; Monark) (Figure 1). First, the participants rested on a chair for 5 minutes followed by a standardized warm-up (i.e., 5 minutes of cycling at 95 W and a 30-second bout at 120 W with a cadence of 80 rpm and 30-second rest between both bouts). Then, they performed the first 40-minute intermittent exercise, which consisted of 20, 2-minute periods. Each 2-minute period started with 15 seconds of passive rest followed by 25 seconds of unloaded cycling, 10 seconds of high-intensity cycling (i.e., 130% of V[Combining Dot Above]O2max), and ended with 70 seconds of moderate-intensity cycling (i.e., 60% of V[Combining Dot Above]O2max). The 40-minute intermittent exercise was based on a pilot work to determine that all participants could finish the required workload, and the proportions of rest and unloaded, high-intensity, and moderate-intensity cycling in this exercise reflect those of the time spent in a soccer match (17). During the half-time, participants either rested on the cycle ergometer for 15 minutes (control), rested on the cycle ergometer for 11 minutes followed by cycling at 60% of V[Combining Dot Above]O2max for 3 minutes (60% RW), or rested on the cycle ergometer for 11 minutes followed by cycling at 30% of V[Combining Dot Above]O2max for 3 minutes (30% RW). Each RW protocol was completed 1 minute before the start of the second intermittent exercise, using the CISP, performed for 20 minutes (10). The CISP consisted of 10, 2-minute periods. Each 2-minute period started with 10 seconds of passive rest followed by a 5-second maximal sprint against a resistance of 7.5% of body mass, followed by 105 seconds of active recovery (i.e., 50% of V[Combining Dot Above]O2max). Each sprint was initiated from a stationary start, with the right pedal crank at approximately 90° to the horizontal. The pedal cadence throughout the trial was 80 rpm, except during the 5-second maximal sprint.
Work was calculated as the mean power output multiplied by the duration of the sprint (5 seconds), using the Anaerobic Test Software (Monark ATS Software; Monark).
The electromyograms (EMG) from the muscle bellies of the right vastus lateralis were recorded during the maximal sprint using a surface electrode (SX230-1000; Biometrics, Newport, United Kingdom) at a sampling frequency of 1,000 Hz, with the ground electrode placed on the left wrist. To reduce impedance (<2 kΩ), the skin was abraded and washed before electrode placement. The root mean square (RMS) as the mean value between the onset and end of the burst were calculated per sprint, and then averaged at 55–65 and 65–75 minutes for simplicity and statistical analyses. Onsets of the bursts were defined by an electric threshold of ±0.2 mV (20). Three-second maximum voluntary isometric contraction against manual resistance was recorded to obtain the 100% EMG signal for normalization before each experimental trial. The 100% RMS value was obtained from 1-second window between 3-second maximum voluntary isometric contraction. The hip and knee joints were flexed at 90° during the maximum voluntary isometric contraction.
Skin temperature over the right vastus lateralis was measured using a thermistor thermometer (ITP082-24; Nikkiso-therm, Tokyo, Japan) during each experimental trial. The muscle temperature was estimated from the skin temperature as follows: Tm = 1.02 × skin temperature +0.89 (correlation between muscle and skin temperatures r 2 = 0.98), as per previously described methods (4).
Breath-by-breath gas analysis was continuously performed using an automatic gas analyzer (Quark CPET; COSMED) during the experimental trial. The mean V[Combining Dot Above]O2, carbon dioxide production (V[Combining Dot Above]CO2), and respiratory exchange ratio (RER) were calculated.
Muscle oxygenation was measured using a light-emitting diode sensor for near-infrared spatial resolved spectroscopy (NIRSRS: Hb14; ASTEM, Kanagawa, Japan) at 2 wave lengths (770 and 830 mm) and a sampling frequency of 5 Hz, with the diode located on the right vastus lateralis at 30% of the length between the patella and greater trochanter (26). The NIRSRS technique provides continuous, noninvasive monitoring of changes in oxygenated, deoxygenated, and total hemoglobin ([INCREMENT]oxy-Hb, [INCREMENT]deoxy-Hb, and [INCREMENT]total-Hb, respectively) concentrations in comparison with those at rest before the standardized warm-up and muscle oxygen saturation (SmO2). The probe consisted of 1 light source and 2 photodiode detectors, and the optode distances were 20 and 30 mm (26). In this study, [INCREMENT]oxy-Hb, [INCREMENT]deoxy-Hb, [INCREMENT]total-Hb, and SmO2 were calculated using fat-correction software (Hb11; ASTEM, Kanagawa, Japan) (26), with the thickness of the fat layer measured using ultrasonography (LogiQ3; GE Healthcare, Tokyo, Japan) before each experimental trial.
Heart rate was measured using a wireless HR monitor at 5-second intervals during the experimental trials (Polar RCX3; Polar Electro).
The rating of perceived exertion (RPE) (2) was assessed before and after the first intermittent exercise, after half-time, and at 10-minute intervals during the CISP (i.e., 0, 40, 55, 65, and 75 minutes).
The sample size was estimated using G*Power 3 (6), using the date from a previous study that investigated the effects of RW on repeated shuttle sprint ability (21). To detect improvements in exercise performance with a power of 80% and an alpha level of 5%, a sample size of ≥6 participants was required. All values are shown as mean ± SD. Statistics were computed using SPSS computer software (Version 24.0; SPSS Japan, Inc., Japan). The mean work, RMS, estimated Tm, V[Combining Dot Above]O2, V[Combining Dot Above]CO2, RER, [INCREMENT]oxy-Hb, [INCREMENT]deoxy-Hb, [INCREMENT]total-Hb, SmO2, HR, and RPE were compared using repeated-measures 2-factor (trial × time) analysis of variance. Where significant interaction and trial effect were found, the values were subsequently analyzed using a Bonferroni multiple comparison test. Partial η2 values are provided as an estimate of the effect size. The 95% confidence intervals (95% CI) for the mean absolute pairwise differences between the trials were calculated using the t-distribution and degrees of freedom (n−1). Root mean square, [INCREMENT]oxy-Hb, [INCREMENT]deoxy-Hb, [INCREMENT]total-Hb, and SmO2 data are represented for 10 participants because they were missing for 1 participant. Statistical significance was set at p ≤ 0.05.
Sprint Performance and Neuromuscular Activity
There were a main effect of trial (p < 0.05, partial η2 = 0.384) and trial × time interaction (p < 0.05, partial η2 = 0.041) for the mean work. The mean work was higher in both RW trials than in the control trial (60% RW: p < 0.05, 95% CI: 26–428 J; 30% RW: p < 0.05, 95% CI: 1–331 J; Figure 2A). The mean work at 55–65 minutes was higher in both RW trials than in the control trial (60% RW: p < 0.05, 95% CI: 73–490 J; 30% RW: p < 0.05, 95% CI: 8–325 J; Figure 2A).
There were a main effect of trial (p < 0.05, partial η2 = 0.336) and time (p < 0.05, partial η2 = 0.384) for the mean RMS. The mean RMS was higher in both RW trials than in the control trial (60% RW: p < 0.05, 95% CI: 0.2–23.2%; 30% RW: p < 0.05, 95% CI: 0.5–35.0%; Figure 2B).
Estimated Muscle Temperature
There was a trial × time interaction (p < 0.05, partial η2 = 0.437) for the estimated Tm.
The estimated Tm at 60 minutes was higher in the 60% RW trial than in the control and 30% RW trials (control: p < 0.05, 95% CI: 0.4–2.0° C; 30% RW: p < 0.05, 95% CI: 0.2–1.5° C; Figure 3). The estimated Tm at 65 minutes tended to be higher in the 60% RW trial than in the control and 30% RW trials (control: p = 0.08, 95% CI: −0.1 to 1.5° C; 30% RW: p = 0.08, 95% CI: −0.1 to 1.1° C; Figure 3).
Gas Analysis and NIRS
The mean values of V[Combining Dot Above]O2, V[Combining Dot Above]CO2, and RER at 54–65 minutes are provided in Figure 4. There was a main effect of trial for the mean V[Combining Dot Above]CO2 and RER (V[Combining Dot Above]CO2: p < 0.05, partial η2 = 0.298, RER: p < 0.05, partial η2 = 0.368). The mean V[Combining Dot Above]CO2 and RER were higher in the 60% RW trial than in the control trial (V[Combining Dot Above]CO2: p < 0.05, 95% CI: 0.1–5.0 ml·kg−1·min−1, RER: p < 0.05, 95% CI: 0.01–0.08).
Changes in the mean values of NIRSRS variables at 54–65 minutes are provided in Figure 5. There was a main effect of trial for [INCREMENT]oxy-Hb and [INCREMENT]total-Hb ([INCREMENT]oxy-Hb: p < 0.05, partial η2 = 0.292, [INCREMENT]total-Hb: p < 0.05, partial η2 = 0.350). There was no significant difference among the 3 trials regarding [INCREMENT]oxy-Hb. [INCREMENT]total-Hb was higher in the 60% RW trial than in the control trial (p < 0.05, 95% CI: 0.4–9.7 µmol·L−1). There was a trial × time interaction for the mean [INCREMENT]oxy-Hb during the 105 seconds of active recovery (p < 0.05, partial η2 = 0.389). At 55–65 minutes, it tended to be higher in both RW trials than in the control trial (control: −2.0 ± 3.9 µmol·L−1; 60% RW: 1.8 ± 2.0 µmol·L−1, p = 0.06, 95% CI: −0.2 to 7.7 µmol·L−1; 30% RW: 1.0 ± 4.3 µmol·L−1, p = 0.06, 95% CI: −0.1 to 6.1 µmol·L−1). There was no significant difference in it at 65–75 minutes among the 3 trials.
Heart Rate and Rating of Perceived Exertion
There were a main effect of trial (p < 0.05, partial η2 = 0.575), time (p < 0.05, partial η2 = 0.979), and trial × time interaction (p < 0.05, partial η2 = 0.815) for the mean HR (Table 1). The mean HR at 54–55 minutes was higher in both RW trials than in the control trial (60% RW; p < 0.05, 95% CI: 12–22 b·min−1; 30% RW: p < 0.05, 95% CI: 2–10 b·min−1). It was also higher in the 60% RW trial than in the 30% RW trial (p < 0.05, 95% CI: 6–16 b·min−1). The mean HR at 55–65 minutes was higher in the 60% RW trial than in the control trial (p < 0.05, 95% CI: 2–8 b·min−1).
There were a main effect of time (p < 0.05, partial η2 = 0.787) and trial × time interaction (p < 0.05, partial η2 = 0.204) for RPE. Rating of perceived exertion at 55 minutes was higher in the 60% RW trial than in the 30% RW trial (p < 0.05, 95% CI: 0.3–2.9) (Table 1).
The major findings of this study were that a very-short-duration (i.e., 3 minutes) RW increased the intermittent cycling sprint performance over the subsequent 10 minutes compared with a traditional passive half-time practice, and that low-intensity (i.e., 30% of V[Combining Dot Above]O2max) RW was as effective in increasing cycling sprint performance as moderate-intensity (i.e., 60% of V[Combining Dot Above]O2max) RW when matched for the total RW duration. The magnitude of increase in cycling sprint performance after the 30% RW trial was similar to that in 10- and 30-m sprint performance reported in previous studies (present study: 4.2%, Lovell et al. (12): 5.0%, and Mohr et al. (19): 4.0%). These findings may have practical implications for players and coaches who are generally busy with other tasks during half-time.
A recent review has reported the effectiveness of a short-duration, moderate-intensity RW protocol (i.e., 7 minutes × 70% of HRmax) in increasing running-based sprint performance (22). Notably, in this study, we investigated the effectiveness of a very-short-duration, low-intensity RW protocol (i.e., 3 minutes × 30% of V[Combining Dot Above]O2max) on intermittent cycling sprint performance after the half-time. Our very-short-duration, low-intensity RW protocol has 2 strengths. First, the RW protocol is easier to implement in practice. Russell et al. discussed the importance of considering the duration of RW for practical applications (22). The short-duration (i.e., 7 minutes) of RW used in previous studies (12,13,19,21,31) may be too time-consuming to implement in practice. Indeed, Towlson et al. have addressed that many fitness coaches and sport scientists felt that approximately 2.6 minutes was available for RW activities during half-time in intermittent team sports (28). Therefore, a very-short-duration (i.e., 3 minutes) RW would be easily implemented in practice compared with the short-duration (i.e., 7 minutes) RW reported previously (12,13,19,21,31). Second, physiological load was lower in low-intensity RW than in moderate-intensity RW. The physiological load of any half-time RW is also important to consider. In this study, V[Combining Dot Above]CO2 and RER were higher in the 60% RW trial than in the control trial, with values being comparable for the 30% RW trial and control trial. These results have suggested that the increases in intermittent cycling sprint performance were comparable between the 30% and 60% RW trials with the former requiring less anaerobic energy. In intermittent team sports, the anaerobic energy stores, such as carbohydrates, decrease in the last part of the second half (18). Therefore, the restriction of anaerobic energy consumption is important for players. In addition, this study has demonstrated that RPE at 55 minutes (i.e., at the end of the half-time) was lower in the 30% RW trial than the 60% RW trial, indicating that participants perceived less exertion when performing the 30% RW trial compared with the 60% RW trial. These results are important in supporting the implementation of a very-short-duration, low-intensity RW protocol in practice for intermittent team sports.
One of the mechanisms for the acute increase in intermittent exercise performance after the 60% RW may be an increase in Tm, which is a major factor for increased exercise performance after RW (19). Mohr et al. have suggested that the mechanism of increased 30-m sprint performance after RW includes increased Tm because there is a correlation (r = 0.6) between the change in Tm and sprint performance during half-time, and the decrement of Tm during half-time was not observed after performing a moderate-intensity RW of 7 minutes (19). In this study, the estimated Tm, which was not directly measured from the muscle, increased after the 60% RW trial, but not 30% RW trial. A previous study reported the relationship between warm-up intensity and temperature (30). Therefore, 60% RW may be enough to increase Tm compared with 30% RW. Interestingly, despite a lack of significant increases in Tm in the 30% RW trial, a similar intermittent cycling sprint performance was observed between both RW trials. Thus, there might be another potential mechanism for the acute increase in intermittent cycling sprint performance after the 30% RW trial.
This study has demonstrated that the mean RMS was higher in the 30% RW trial than in the control trial. The result has suggested that enhanced muscle activation as evidenced by increased EMG activity during sprints may be related to the acute increase in intermittent cycling sprint performance after the 30% RW because there is a linear relationship between RMS and power output during cycling (11). Furthermore, a previous study has suggested that a low-intensity warm-up, such as half-squat exercise performed at 25 and 35% of 1 repetition maximum, enhanced muscle activation as evidenced by increased EMG activity of vastus lateralis and enhanced countermovement jump performance (24). Therefore, 30% RW may result in increased intermittent cycling sprint performance through enhanced muscle activation as evidenced by increased EMG activity. However, the mechanisms for enhanced muscle activation in this study are not clear. Increase in EMG activity is caused by increased motor unit recruitment and firing frequency (11). Future research should consider the effects of low-intensity RW on motor unit recruitment and firing frequency.
The mean [INCREMENT]oxy-Hb at 105 seconds of active recovery at 55–65 minutes tended to be higher in both RW trials than in the control trial. Increased oxygen availability after maximal exercise accelerates the resynthesis of phosphocreatine (8,9,14). Thus, the present results have suggested that both RWs may increase the oxygen availability in the muscle and contribute to the resynthesis of phosphocreatine during active recovery at 55–65 minutes. In this study, the mean work and RMS during maximum cycling sprint were also higher in both RW trials than in the control trial. It is possible to assume that adenosine triphosphate utilization was higher in both RW trials than in the control trial, and an increase in adenosine triphosphate resynthesis may be required in both RW trials. To restore phosphocreatine, the mean [INCREMENT]oxy-Hb at 105 seconds of active recovery at 55–65 minutes may have increased in both RW trials. We have, however, no direct data to support these speculations and thus such speculations require further study.
The mechanism of increased mean [INCREMENT]oxy-Hb, which is an indicator of the balance between oxygen supply and utilization (25), may be related to an increase in the oxygen supply to the muscles (1). A previous study has reported that oxygen utilization in muscles was strongly correlated with the activity of the muscle (16). In this study, the mean RMS was higher in both RW trials than in the control trial, and therefore, oxygen utilization may also be higher in both RW trials than in the control trial, thus, suggesting an increase in oxygen supply, which is greater than oxygen utilization, in both RW trials. Indeed, a previous study has suggested that oxy-Hb relatively increased by warm-up, and this may result from an increase in the oxygen supply to the muscles due to the increased blood flow to the muscle and shifted the oxygen dissociation curve to the right (27). Although oxygen supply to the muscles was not measured in this study, HR was higher after both RW trials than after the control trial. Heart rate has been used as an indicator of presumable oxygen supply to the muscles with an inherent limitation (5). Therefore, the present results have suggested that oxygen supply may be elevated after both RWs. By contrast, there was no significant difference in the mean [INCREMENT]oxy-Hb during the 105 seconds of active recovery at 65–75 minutes among the 3 trials. This finding may relate to the finding that there was no significant difference in the mean HR at 65–75 minutes among the 3 trials.
The limitations of this study need to be acknowledged in the interpretation of results. The mode and intensity of 2 intermittent exercises (i.e., simulated each half) we used in this study has limited practical applications in many intermittent team sports. This study evaluated the effectiveness of very-short-duration, low-intensity RW protocol using a cycling exercise, whereas most intermittent team sports use overground running and activities such as jumping and multidirectional running. It is not possible to conclude whether the present results can be applied in these activities. Moreover, the type of the participants we employed in this study (i.e., healthy men who trained for more than 2 days per week) does not allow to make a comparison with professional athletes. Nonetheless, it would be interesting to determine whether the RW used in this study would have similar effects in professional athletes when other modes of repeated exercises are performed for an extended period after half-time because the work during cycling sprint is proportional to sprint performance in over-ground running (7).
In conclusion, both very-short-duration RW at low and moderate intensities were equally effective in improving intermittent cycling sprint performance over 10 minutes after half-time, in comparison with a traditional passive half-time practice. These findings challenge the half-time practice and indicate that a very-short-duration low-intensity RW is beneficial in improving exercise performance after half-time.
This study revealed that players of intermittent team sports may be able to increase their intermittent sprint performance using very-short-duration RW at low and moderate intensities. These findings would support its use in the field of athletics because very-short-duration RW may allow to accommodate with other specific ergogenic strategies during half-time. Moreover, this study have suggested that a low-intensity RW has same increase rate for exercise performance and less anaerobic energy compared with moderate-intensity RW. Therefore, very-short-duration, low-intensity RW may be recommended for players of intermittent team sports.
The authors thank the participants for their cooperation. The present study was supported by a Grant-in-Aid for JSPS Research Fellow (no: 17J02878) and internal funding from the Waseda University. The authors have no conflicts of interest to disclose. The results of this study do not constitute endorsement of the equipment by the authors or the National Strength and Conditioning Association.
1. Bishop D. Warm up II: Performance changes following active warm up and how to structure the warm up. Sports Med 33: 483–498, 2003.
2. Borg GA. Psychophysical bases of perceived exertion. Med Sci Sports Exerc 14: 377–381, 1982.
3. Bradley PS, Sheldon W, Wooster B, Olsen P, Boanas P, Krustrup P. High-intensity running in English FA Premier League soccer matches. J Sports Sci 27: 159–168, 2009.
4. De Ruiter CJ, Jones DA, Sargeant AJ, De Haan A. Temperature effect on the rates of isometric force development and relaxation in the fresh and fatigued human adductor pollicis muscle. Exp Physiol 84: 1137–1150, 1999.
5. DeLorey DS, Kowalchuk JM, Paterson DH. Effects of prior heavy-intensity exercise on pulmonary O2
uptake and muscle deoxygenation kinetics in young and older adult humans. J Appl Physiol (1985) 97: 998–1005, 2004.
6. Faul F, Erdfelder E, Lang AG, Buchner A. G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods 39: 175–191, 2007.
7. Fitzsimons M, Dawson B, Ward D, Wilkinson A. Cycling and running tests of repeated sprint ability. Aust J Sci Med Sport 25: 82–87, 1993.
8. Girard O, Mendez-Villanueva A, Bishop D. Repeated-sprint ability - part I: Factors contributing to fatigue. Sports Med 41: 673–694, 2011.
9. Haseler LJ, Hogan MC, Richardson RS. Skeletal muscle phosphocreatine recovery in exercise-trained humans is dependent on O2
availability. J Appl Physiol (1985) 86: 2013–2018, 1999.
10. Hayes M, Smith D, Castle PC, Watt PW, Ross EZ, Maxwell NS. Peak power output provides the most reliable measure of performance in prolonged intermittent-sprint cycling. J Sports Sci 31: 565–572, 2013.
11. Hug F, Dorel S. Electromyographic analysis of pedaling: A review. J Electromyogr Kinesiol 19: 182–198, 2009.
12. Lovell R, Midgley A, Barrett S, Carter D, Small K. Effects of different half-time strategies on second half soccer-specific speed, power and dynamic strength. Scand J Med Sci Sports 23: 105–113, 2013.
13. Lovell RJ, Kirke I, Siegler J, McNaughton LR, Greig MP. Soccer half-time strategy influences thermoregulation and endurance performance. J Sports Med Phys Fitness 47: 263–269, 2007.
14. McCully KK, Lotti S, Kendrick K, Wang Z, Posner JD, Leigh J Jr, et al. Simultaneous in vivo measurements of HbO2
saturation and PCr kinetics after exercise in normal humans. J Appl Physiol (1985) 77: 5–10, 1994.
15. McGowan CJ, Pyne DB, Thompson KG, Rattray B. Warm-up strategies for sport and exercise: Mechanisms and applications. Sports Med 45: 1523–1546, 2015.
16. Miura H, Araki H, Matoba H. Relationship between oxygenation and myoelectric activity in vastus lateralis muscle during exercise with constant work rate. Med Sci Sports Exerc 29: S180, 1997.
17. Mohr M, Krustrup P, Bangsbo J. Match performance of high-standard soccer players with special reference to development of fatigue. J Sports Sci 21: 519–528, 2003.
18. Mohr M, Krustrup P, Bangsbo J. Fatigue in soccer: A brief review. J Sports Sci 23: 593–599, 2005.
19. Mohr M, Krustrup P, Nybo L, Nielsen JJ, Bangsbo J. Muscle temperature and sprint performance during soccer matches–beneficial effect of re-warm up at half-time. Scand J Med Sci Sports 14: 156–162, 2004.
20. Racinais S, Bishop D, Denis R, Lattier G, Mendez-Villaneuva A, Perrey S. Muscle deoxygenation and neural drive to the muscle during repeated sprint cycling. Med Sci Sports Exerc 39: 268–274, 2007.
21. Russell M, Tucker R, Cook CJ, Giroud T, Kilduff LP. A comparison of different heat maintenance methods implemented during a simulated half-time period in professional Rugby Union players. J Sci Med Sport 21: 327–332, 2018.
22. Russell M, West DJ, Harper LD, Cook CJ, Kilduff LP. Half-time strategies to enhance second-half performance in team-sports players: A review and recommendations. Sports Med 45: 353–364, 2015.
23. Sargeant AJ, Dolan P. Effect of prior exercise on maximal short-term power output in humans. J Appl Physiol 63: 1475–1480, 1985, 1987.
24. Sotiropoulos K, Smilios I, Christou M, Barzouka K, Spaias A, Douda H, et al. Effects of warm-up on vertical jump performance and muscle electrical activity using half-squats at low and moderate intensity. J Sports Sci Med 9: 326–331, 2010.
25. Takagi S. Skeletal muscle oxygen dynamics and peak aerobic capacity. J Phys Fitness Sports Med 5: 379–383, 2016.
26. Takagi S, Murase N, Kime R, Niwayama M, Osada T, Katsumura T. Aerobic training enhances muscle deoxygenation in early post-myocardial infarction. Eur J Appl Physiol 116: 673–685, 2016.
27. Takizawa K, Ishii K. Relationship between muscle oxygenation
, and high intensity aerobic exercise performance improving effect of warm-up. Adv Exer Sport Physiol 12: 127–133, 2006.
28. Towlson C, Midgley AW, Lovell R. Warm-up strategies of professional soccer players: Practitioners' perspectives. J Sports Sci 31: 1393–1401, 2013.
29. Wittekind A, Cooper CE, Elwell CE, Leung TS, Beneke R. Warm-up effects on muscle oxygenation
, metabolism and sprint cycling performance. Eur J Appl Physiol 112: 3129–3139, 2012.
30. Yaicharoen P, Wallman K, Morton A, Bishop D. The effect of warm-up on intermittent sprint performance and selected thermoregulatory parameters. J Sci Med Sport 15: 451–456, 2012.
31. Yanaoka T, Yamagami J, Kidokoro T, Kashiwabara K, Miyashita M. Halftime re-warm up with intermittent exercise improves the subsequent exercise performance of soccer referees. J Strength Cond Res 32: 211–216, 2018.