Athletes involved in team sports may be subject to varying degrees of sleep deprivation either before or after training and competition (4,26). The extent of this sleep disruption or loss can range from minor (2-4 h) to quite extensive (overnight), depending on the prevailing circumstances. As sleep is known to be critical in the restoration of metabolic processes (8,11,24) and regulation of hormone secretion (growth hormone, prolactin, and cortisol) (17), the effect of sleep loss before exercise may be detrimental to team-sport exercise. Moreover, sleep disruption has been associated with the increased perception of negative mood states (15,16), suppression of resting HR and core temperature (33), and reductions in aerobic oxidation capacity and decreased metabolic enzyme activity (34). Despite negative metabolic and physiological consequences at rest and during exercise after sleep loss, and the belief among athletes and coaches of the importance of adequate sleep for ensuing performance, the effect of sleep loss on team-sport exercise remains unclear.
The reality of many competition and training schedules results in athletes performing prolonged, high-intensity exercise bouts on consecutive days. Given the high physical, physiological, and metabolic demands associated with team sports (30), recovery from such exercise may be prolonged. Recovery may be further dampened if adequate sleep is not achieved because sleep deprivation/disruption has been associated with increased energy expenditure and metabolic demands (3). Furthermore, given that muscle glycogen depletion is evident after prolonged, repeated-sprint exercise (12), the increased energy expended during extended hours awake may affect the rate of resynthesis of the depleted muscle glycogen. In addition, sleep deprivation may also alter physiological functions including a decreased resting HR (7,33), body temperature (33), plasma catecholamine levels and increased minute ventilation and CO2 production (7). Further, during exercise, decreased evaporative heat loss (27), maximal HR (7), peak oxygen consumption (7,22), and peak ventilation (22) can be observed. Despite sleep deprivation affecting these metabolic and physiological parameters at rest and during exercise, the effect on team-sport athletes and subsequent self-paced, intermittent-sprint performance is not well understood.
The implications of altered physiological and metabolic states due to sleep deprivation and/or disruption may negatively affect subsequent exercise performance (9,14,20,25,29). Although no studies to date have specifically examined self-paced intermittent-sprint exercise (ISE), several studies have examined components of team-sport exercise including aerobic and anaerobic exercise performance and muscular strength, respectively. Prolonged, self-paced exercise protocols have shown that sleep deprivation reduces distance covered during 30 min of free-paced treadmill running in 11 healthy male participants (20); however, 30 h of sleep deprivation had minimal effect on exercise workload manipulation at a set RPE in 24 healthy students (16). Fixed-paced protocols also demonstrate contradictory findings, with 72-h sleep loss having minimal effect on steady-state exercise at 40%, 60%, and 80% V˙O2max in untrained subjects (9), whereas Martin (14) reported that sleep deprivation reduced time to exhaustion after 36-h sleep loss in healthy (19-27 yr) subjects. In addition, peak and mean power during a 30-s Wingate anaerobic test is reduced after 36 h of sleep loss (29). Finally, studies examining muscular strength have reported 60-h sleep loss to have minimal effect on 25 maximal isokinetic contractions of the upper or lower body in 11 healthy subjects (30). Conversely, Reilly and Piercy (25) reported that sleep restrictions of 3 h for three consecutive days reduced maximal strength during bench press, leg press, and dead lift and reduced submaximal lift capacity as well. Although previous studies provide insight into the possible effects of sleep deprivation on ISE, the specific effect on actual or simulated (intermittent-sprint) team-sport exercise performance remains unclear. Furthermore, the underlying mechanisms responsible for alterations in exercise intensity and thus pacing strategies are not well understood as the majority of studies have used constant-intensity protocols, which may not be indicative of team-sport exercise.
Despite the possibility of team-sport athletes experiencing sleep loss during a normal training or competition schedule, the relationship among sleep deprivation, recovery, and subsequent intermittent-sprint performance remains equivocal. Therefore, the aim of this study was to examine the effect of 30-h sleep deprivation on self-paced intermittent-sprint performance and pacing strategies. A secondary aim of the study was to examine sleep deprivation on muscle glycogen restoration, mood states, and neuromuscular function and implications on subsequent team-sport exercise. We hypothesized that ∼30-h sleep deprivation would negatively affect intermittent-sprint performance and recovery of muscle glycogen compared with a normal night's sleep.
Ten male team-sport athletes, playing at club and representative levels with competitive matches and formal training sessions three or more times per week took part in this study. Mean ± SD characteristics were the following: age 21 ± 3 yr, mass 81.5 ± 9.5 kg, height 178.6 ± 9.2 cm, and maximal aerobic capacity (V˙O2max) 56.8 ± 5.3 mL·kg−1·min−1 (45.3-63.1 mL·kg−1·min−1). All participants were questioned about their sleeping patterns, and baseline sleep data were collected 2 d before initial testing, with participants excluded if substantial variation in sleeping patterns were evident. Participants were informed of the requirements and demands of the study, and written informed consent was obtained before the commencement of testing. This study was approved by the institutional human ethics committee.
Initially participants completed a familiarization session that also included a graded exercise test to determine V˙O2max and velocity of V˙O2max (vV˙O2max). After familiarization, participants completed a 1-d (baseline) trial (n = 7) and two-consecutive-day experimental trials (n = 10) in a counterbalanced crossover design. Experimental trials consisted of a normal night's sleep (CONT1 and CONT2) or no sleep (SDEP1 and SDEP2) between testing days. All testing procedures and exercise protocols were identical between the baseline and experimental sessions except for the inclusion of the muscle biopsy procedures. The exercise bouts performed on each day included a 30-min graded exercise run (GXR) followed by a 50-min ISE protocol. Muscle biopsies were obtained before GXR and 30-min after ISE during the baseline session, whereas a single biopsy was obtained before GXR on day 2 of both experimental trials. As all food, fluid, activity, and testing procedures were standardized, the baseline session served to provide information regarding representative muscle glycogen concentration of the vastus lateralis after exercise on day 1 for both conditions of the experimental trials. Accordingly, no biopsies were obtained on day 1 during actual experimental trials to avoid any delayed muscle soreness on day 2 due to the biopsy procedure. The experimental trials involved two-consecutive-day testing sessions of identical procedures separated by either one night of "normal" sleeping hours relative to the participant's sleep patterns (CONT) or a night without sleep (SDEP). Although extreme, the study was designed to simulate a night of minimal sleep due to extended travel commitments. After exercise on CONT1, participants remained supervised within the laboratory until after the meal provision, before returning home for a "normal" night sleep and returning to the laboratory by 8:30 a.m. the following morning. Conversely, during SDEP, participants were required to remain awake in the laboratory for the duration of the experimental trial and were supervised by the research team. Each trial was completed at the same time of day (3:00 p.m.) and was separated by 7 d to ensure adequate recovery and to allow participants to regain normal sleep patterns.
All food and fluid during the consecutive day trials were matched and standardized between the two conditions. The diet consumed by the participants was controlled by the research team ∼24 h before each session and between testing sessions during the experimental trials (total dietary control ∼60 h) with a CHO intake of ∼3 g·kg−1 body weight. Consumption and timing of food and fluid were recorded by participants in a diary and were inspected by the research team. Food and fluid were provided to participants and were matched for caloric intake between conditions; however, during SDEP, a smaller portion of the food allocated for dinner was consumed at the same time as CONT, with the remaining food spaced intermittently throughout the ensuing evening. Participants abstained from food, fluid, and caffeine 3 h before each testing session, and no strenuous activity was completed 24 h before each session, excluding the exercise protocol. Participants consumed 500 mL of water 1 h before testing to ensure the participants presented in a euhydrated state and consumed an additional 500 mL between the GXR and ISE protocols. Sleep patterns were monitored two nights before the initial testing session to determine "normal" sleeping patterns for each participant, with sleep diaries and sleep watches (Actiwatch; Philips Respironics, Murrysville, PA). The sleep watches were also worn, and diaries were completed throughout the experimental trials to ensure compliance with the experimental interventions. All testing sessions were completed within a controlled environment with an ambient temperature during the GXR and ISE of 19°C ± 1°C and 17°C ± 1°C, respectively.
For each respective day, participants commenced with a 30-min GXR on a motorized treadmill (True 825 SDFT System; ETL Testing Laboratories, Inc., Cortland, NY) at 60%, 70%, and 80% vV˙O2max for 10 min at each respective intensity. Exercise intensities were calculated based on the relative percentage intensity determined from vV˙O2max. The inclusion of the GXR was designed to simulate a warm-up and ensured the entire exercise completed and energy expended during a respective testing session was similar to a competition or training scenario. After a 10-min recovery, participants then completed a 50-min free-paced ISE protocol with 1-min breaks every 10 min. The self-paced exercise protocol was completed on a synthetic surface 20 m in length and involved a 15-m maximal sprint each minute, with a 5-m deceleration zone before impacting with a crash mat placed upright against a wall. Immediately after impact with the mat, participants completed a self-paced exercise at varying intensities for the remainder of the minute (∼50 s) (28). Exercise bouts were submaximal, free-paced activities of hard running, jogging, or walking. During the hard running bout, participants were instructed to "cover as much distance as possible," while selecting their own pace during the jogging and walking bouts, respectively. These bouts were completed in a shuttle-run format with only one mode completed each minute, rotating through each minute in the above order. The intraclass correlation (r) and coefficient of variation (CV) for the sprint times and total distance covered for this ISE protocol are r = 0.90 and 0.96 and CV = 1.9% and 1.5%, respectively. Participants returned at 50 s of each minute to complete the ensuing sprint. To invoke a greater eccentric load, every 7 min participants completed eight deep-squat, double-leg bounds, aiming to cover as much distance as possible. Participants were given verbal support and encouragement during sprints and hard running bouts and were not aware that distance was being counted to ensure no conscious manipulation of exercise intensity was implemented.
Performance measures recorded during the intermittent-sprint protocol included 15-m sprint time and distance covered during each submaximal exercise bout and double-leg bounds. Maximal 15-m sprint performances were recorded with an infrared timing gate system (Speed-Light, Swift, Australia), with mean and total sprint time calculated. Mean and total distance covered throughout each respective submaximal exercise bout (hard running, jogging, and walking) were measured by manually counting meters covered using 1-m floor markings. Finally, double-leg bounds were recorded by manually measuring the distance covered with a tape measure to determine mean and total distance covered. To assess pacing strategies implemented by the participants, mean sprint times and distance covered during self-paced exercise were reported as mean ± SD of each 10 min of the ISE protocol.
Maximal voluntary force and voluntary activation.
Maximal isometric voluntary contraction (MVC) and evoked twitch properties of the right knee extensors were assessed before GXR and ∼5 min after ISE. During pre-GXR assessment, participants completed a 3-min moderate-intensity (60-W) warm-up on a cycle ergometer (828E; Monark, Stockholm, Sweden). The neuromuscular test was completed on a modified dynamometer, in which participants were seated on a straight-back chair with the hip and knee angle at 90° (0° represents full extension). A Velcro strap was placed at the ankle, 1 cm above the lateral malleolus, which attached onto a suspended load cell (6000, ICI; Sensortronics, Covina, CA). The load cell was attached to the undercarriage frame of the chair and detected the force, which was amplified and recorded by a signal acquisition system (PowerLab, 8/30 and Chart v6.1.1; ADInstruments, Sydney, Australia). The load cell was zeroed and calibrated before each testing session. Participants were secured to the chair with a waist strap, and during testing, participants were instructed to place their arms across their chest to minimize additional forces contributing to the MVC.
Muscle stimulation was achieved using two 50 × 90-mm self-adhesive surface electrodes (Verity Medical, Ltd., Stockbridge, Hampshire, UK), placed on the anterior aspect of the right thigh, ∼2 cm below the inguinal fold and ∼3 cm above the superior border of patella. The current was delivered via a stimulator (Model DS 7A; Digitimer Ltd., Weleyn, Garden City, England) using a doublet square-wave pulse with a width of 200 μs. Initially, the current was applied in incremental steps until a plateau in twitch force was reached. The current was then increased by a further 25% to ensure that supramaximal stimulation was used for all tests. MVC testing consisted of a series of 15 maximal isometric contractions of right knee extensors. Participants were instructed to produce a maximal isometric effort for duration of ∼3 s for each contraction, with the start of each contraction 10 s apart. The first and final five contractions included a superimposed electrical stimulus manually delivered when peak torque was achieved (∼1 s after initiation of each contraction) and within 2 s of relaxation of MVC, a second stimulus was delivered, with the muscle at complete rest. All data were processed using a customized, formulated spreadsheet (Excel 2007; Microsoft Corp., Redmond, WA). Initially, the effect of gravity of the lower leg was corrected by calculating average load applied to the force transducer during the 50-ms period immediately before the force onset. Peak voluntary force (VF) was determined as the maximal force exerted before the delivery to electrical stimulation, and voluntary activation (VA) was calculated using the twitch interpolation technique (1). Mean ± SD VF and VA for all contractions were assessed within and between conditions.
On arrival, subjects provided a urine sample to measure hydration status (Refractometer 503; Nippon Optical, Works Co., Tokyo, Japan). Nude mass was recorded before GXR and 20 min after ISE protocol on a set of calibrated scales accurate to 10 g (Fitness Technology, Inc., O'Fallon, MO). Participants were instructed to "towel down" as much sweat as possible before stepping onto the scales, and the difference in mass was used to calculate sweat rate. HR was recorded before GXR, before ISE, and every 10 min during the ISE protocol with an HR monitor and wristwatch receiver (F1; Polar Electro-Oy, Kempele, Finland). Core temperature (Tcore) was measured with a telemetric capsule that was ingested 4-5 h before each testing session to ensure it had passed into the gastrointestinal tract and was unaffected by the ingested food and/or fluid. Core temperature was recorded from a handheld monitor receiving a telemetric measure from the ingested capsule before GXR and ISE and every 10 min during the ISE protocol (VitalSense; Mini Mitter, Bend, OR). After sterilization of the fingertip with an alcohol swab, the skin was perforated using a lancet and a 35-μL sample of capillary blood from the fingertip was obtained before GXR, before ISE, and at 30 min after ISE. Capillary blood samples were measured for lactate (La−, Lactate Pro; Arkray, Inc., Kyoto, Japan) and glucose (Glu, Accu-Chek Advantage; Roche, Mannheim, Germany) as per the manufacturer's instructions. CV for the Lactate Pro and Accu-Chek Advantage device were 3.1% and 2.6%, respectively.
Muscle biopsies were obtained using the Bergström needle biopsy technique before GXR and 30 min after ISE during the single (baseline) session and before GXR on day 2 of both experimental sessions. During biopsy extraction, participants laid supine with the biopsy site selected in the belly of the vastus lateralis muscle. Once the site was identified, the muscle was anaesthetized by injection of 1.5 mL of xylocaine under the skin. A small incision was made whereby a muscle biopsy needle was inserted, and with manual suction, a small piece (∼80-100 mg) of muscle was extracted from the vastus lateralis muscle of the left thigh (to avoid interference with the electrical stimulation site). Preexercise and postexercise samples were extracted from the same incision, while after exercise with the biopsy needle angled proximal to the previous pre-GXR sample site. Subsequent biopsy incision was made distal to the previous incisions. Biopsy samples were immediately frozen with liquid nitrogen and stored at −80°C until further analysis of muscle glycogen concentration. A small piece (4-6 mg) of freeze-dried muscle was removed and dissected of visible blood, fat, and connective tissue. One aliquot of freeze-dried muscle was homogenized for 1 min in perchloric acid. To determine muscle glycogen, homogenate was diluted with 350 μL of 2 M HCl and incubated in a heating block at 100°C for 2 h. After incubation, the homogenate was reweighed, vortexed, and neutralized to a pH 7-8 with 2 M K2CO3. The glycogen aliquots were centrifuged, and the supernatant was used for analysis. Muscle glycogen concentration was assessed using a spectrophotometer to assess absorbance with the addition of glucose-6-phosphate dehydrogenase followed by hexokinase. Glycogen concentration was expressed as millimoles of glycogen per kilogram of dry weight (dw).
Perceptual measures of RPE and a modified POMS were used throughout the protocol. RPE was recorded every 10 min during ISE protocol, whereas mood state was assessed using the modified POMS test assessing moods (lively, alert, energetic, and fatigued) that were deemed applicable to this particular study before GXR and 20 min after the ISE protocol.
Data are reported as mean ± SD. Initially, a Shapiro-Wilk test was used to confirm that data did not differ substantially from a Gaussian distribution. A repeated-measures ANOVA (time × day × condition) was used to determine main effects within and/or between each experimental condition. Significance was set at P = 0.05. A paired t-test was used to determine differences in muscle glycogen concentration between conditions. A Tukey post hoc was then applied where appropriate to determine statistical significance. In addition, Pearson product-moment correlation was completed to assess the relationship between preexercise muscle glycogen concentration and mean sprint time, total distance covered, and hard running distance covered.
No differences were present in mean sleep duration during the two nights preceding the experimental sessions, with 8.1 ± 1.6 h sleep during CONT and 8.4 ± 1.5 h sleep during SDEP (P > 0.05), indicating that regular and consistent sleeping patterns were evident for all participants. Sleep duration was significantly different between CONT (8.5 ± 1.7 h) and SDEP (0 ± 0 h) conditions during the evening of the experimental trials (P = 0.01).
Intermittent-sprint performance and pacing strategies.
Sleep deprivation resulted in significant differences between conditions for mean sprint times, with slower times recorded for 8 of the 10 participants during SDEP2 compared with CONT2 (P = 0.05; Table 1). Further, within the SDEP condition, mean and total sprint times were significantly slower during SDEP2 compared with SDEP1 (P = 0.04). Pacing of maximal sprint efforts seemed to be affected by sleep loss, with mean sprint times slower during the initial and final 10 min during SDEP2 compared with CONT2 (P = 0.01-0.03, respectively). Within the respective SDEP and CONT conditions, mean sprint times were slower during the initial 10 min on day 2 compared with day 1 (CONT1 = 2.63 ± 0.11 s, CONT2 = 2.68 ± 0.12 s, SDEP1 = 2.64 ± 0.18 s, SDEP2 = 2.74 ± 0.15 s, P < 0.02). Moreover, during the 11- to 20-min, 31- to 40-min, and 41- to 50-min phases, mean sprint times were slower during SDEP2 than SDEP1 (P = 0.03-0.05).
Total distance covered during the self-paced, submaximal exercise bouts was not significantly different between conditions, with 6 of 10 participants performing worse during SDEP2 compared with CONT2 (P > 0.05). However, distance covered during SDEP2 was significantly lower during the initial and final 10 min compared with SDEP1 (P = 0.01; Fig. 1A) and lower during the final 10 min compared with CONT2 (P = 0.02; Fig. 1A). No significant differences were evident within or between conditions for mean and total hard running, jogging, and walking distance covered with 6, 4, and 5 of 10 participants, respectively, performing worse after sleep deprivation (P > 0.05; Table 1). Finally, distance covered during double-leg bounds are presented in Table 1, and significant differences were noted within and between conditions, with less mean and total distance covered by 8 of 10 participants during SDEP2 compared with CONT2 (P = 0.01) and SDEP1 (P = 0.01), respectively.
Maximal VF and VA.
Peak VF was reduced after ISE during SDEP1 compared with CONT1 (P = 0.02; Fig. 2A) and after ISE in SDEP2 compared with CONT2 in 8 of 10 participants (P = 0.04). Peak VF was reduced before GXR on SDEP2 compared with SDEP1 (P = 0.01). Percentage change between CONT2 and SDEP2 preexercise VF was 11.7% ± 9.0%, whereas after exercise, it was reduced to 4.7% ± 9.7%. VA was significantly reduced in SDEP2 compared with CONT2 for both before GXR and after ISE in 8 of 10 participants (P = 0.03 and P = 0.05, respectively; Fig. 2B). Moreover, VA was reduced before GXR in SDEP2 compared with SDEP1 (P = 0.01). Preexercise VA percentage change between CONT2 and SDEP2 was 9.9% ± 10.2%, whereas after exercise, it was 6.3% ± 9.7%.
Muscle glycogen concentration was significantly reduced in all participants after exercise (122 ± 64 mmol·kg−1 dw) compared with before exercise (310 ± 67 mmol·kg−1 dw) during the baseline session (P = 0.001). Preexercise glycogen content was significantly lower in SDEP2 (209 ± 60 mmol·kg−1 dw) compared with baseline (P = 0.03) in all participants, although no differences were present between baseline and CONT2 (274 ± 54 mmol·kg−1 dw; P > 0.05). Further, muscle glycogen was significantly lower before exercise in SDEP2 compared with CONT2 (P = 0.05), and percentage change between conditions was 24.5% ± 10.7%. No significant correlations (P > 0.05) were evident between before GXR muscle glycogen concentration and mean sprint times (r = 0.13), total distance covered (r = 0.22-0.28), and hard running distance covered (r = 0.2-0.3).
No significant differences were evident within or between conditions for pre-GXR (CONT1 = 68 ± 5 bpm, CONT2 = 68 ± 10 bpm, SDEP1 = 66 ± 8 bpm, SDEP2 = 62 ± 9 bpm, P > 0.05) or pre-ISE HR (CONT1 = 98 ± 5 bpm, CONT2 = 97 ± 11 bpm, SDEP1 = 104 ± 8 bpm, SDEP2 = 100 ± 9 bpm, P > 0.05). Mean HR throughout the ISE protocol was significantly lower on day 2 compared with day 1 within both SDEP and CONT conditions, respectively (CONT1 = 176 ± 8 bpm, CONT2 = 173 ± 8 bpm, SDEP1 = 178 ± 9 bpm, SDEP 2 = 171 ± 12 bpm, P = 0.01; Fig. 3A). Moreover, mean HR was lower at 10, 20, 30, and 50 min of ISE during SDEP2 compared with SDEP1 (P = 0.01-0.03) and lower at 20 min during CONT2 (P = 0.04; Fig. 2A). Peak HR was not significantly different within or between conditions (CONT1 = 184 ± 8 bpm, CONT2 = 183 ± 8 bpm, SDEP1 = 185 ± 8 bpm, SDEP2 = 179 ± 9 bpm, P > 0.05). Core temperature (Fig. 3B) was not different (P > 0.05) within or between conditions at any stage of the exercise protocol on respective days. There were also no differences in urine specific gravity (USG) within or between conditions (CONT1 = 1.005 ± 0.002, CONT2 = 1.004 ± 0.002, SDEP1 = 1.011 ± 0.013, SDEP2 = 1.004 ± 0.002, P > 0.05). Differences in sweat rate indicated by change in nude mass were not different within the SDEP condition or between conditions but were evident within CONT with greater sweat rate on day 2 (1.4 ± 0.6 L) compared with day 1 (1.1 ± 0.4 L) (P = 0.008). Mean ISE blood lactate concentrations were lower on day 2 within CONT condition compared with day 1 (P = 0.01; Table 2). Finally, blood glucose concentrations were higher on day 2 compared with day 1 within the CONT and SDEP conditions (P = 0.01 and P = 0.04, respectively) but were lower in SDEP2 compared with SDEP1 at 30 min after ISE (P = 0.05; Table 2).
No significant differences were evident within and between conditions for RPE (P > 0.05; Fig. 3C). Differences were noted within and between conditions for the moods "lively, "alert," "energetic," and "fatigued" assessed in the modified POMS questionnaire. Between conditions, participants rated to be less "lively" before GXR on SDEP2 compared with CONT2 (P = 0.01; Table 3). Furthermore, pre-GXR participants were less "alert," "energetic," "lively," and more "fatigued" on SDEP2 compared with SDEP1 (P = 0.01-0.02). Finally, participants felt more "fatigued" before GXR on CONT2 compared with CONT1 (P = 0.04; Table 3).
The aim of the present study was to determine the effect of sleep deprivation on pacing strategies, intermittent-sprint performance, and physiological and perceptual recovery. Results indicated that ∼30 h of sleep deprivation between simulated team-sport exercise bouts resulted in slowed pacing strategies, reduced intermittent-sprint performance, reduced muscle glycogen content, reduced peak VF and VA, and negative perceptual strain. The reduction in performance during SDEP2 was most evident during sprint efforts, and distance covered during the bounds and self-paced efforts of the ISE. Despite the findings, it is difficult to determine whether sleep loss exclusively or associated factors were directly responsible for the observed performance decrements.
Some variation regarding experimental design and subsequent results are evident within the previous literature examining the effects of sleep disruption on exercise performance (14,16,18,29). The current study used a free-paced exercise protocol designed to allow the manipulation of the exercise intensities throughout the protocol by the participant, as observed in team-sport exercise. Similar to the decline in self-paced distance covered in the present study, Oliver et al. (20) reported reductions in distance covered during a 30-min self-paced treadmill run after a one-night sleep deprivation with no significant effects on Tcore, Tskin, and HR. Martin and Haney (16) also used a self-paced protocol; however, self-selection was achieved through the manipulation of the treadmill grade to exercise at a set RPE after either a 30-h sleep loss or a normal sleep. The differing protocols may explain the dissimilar findings, as Martin and Haney (16) reported that sleep loss had no effect on absolute treadmill grades. On the contrary, examinations of incremental tests have reported similar findings to the present study in that sleep loss negatively affects exercise performance. Although the present study highlighted sleep loss increased 15-m sprint times and reduced the volume of work completed, Martin (14) reported an 11% decrease in time to exhaustion during heavy treadmill walking, whereas Mougin et al. (17) reported partial sleep loss through late bedtimes and early rises reduced maximal work rate compared with baseline during an incremental test to exhaustion. That said, a previous study by Mougin et al. (18) reported no effect on time to exhaustion test when exposed to a disrupted night sleep. The inconsistencies within the existing literature may be due to the variations in protocol design, and therefore, the degree of influence on physiological and perceptual parameters and thus performance remains unclear. Regardless, this study is the first to use simulated team-sport activity, with sleep deprivation seemingly having a negative effect on intermittent-sprint activity.
Although differences in muscle glycogen content were small (65 mmol·kg−1 dw) between conditions, sleep loss significantly reduced muscle glycogen concentrations before exercise on day 2 compared with a normal night sleep. Because the exercise protocol and the food consumption for the baseline session and experimental trials were identical, it could be assumed that muscle glycogen depletion would be similar after day 1 during the experimental trials. Interestingly, despite the consumption of an identical diet between conditions, participants presented with reduced muscle glycogen concentrations in SDEP2. Because of the standardization of the diet and supervision by the research team, either a difference in the timing of food consumed and/or the additional energy expended while in a sleep deprived state may be responsible for the reduced muscle glycogen concentration. It has been reported that sleep aids in the reduction of energy expenditure below rest alone by 10%-15% (21), reduces metabolic requirements, and is involved in energy conservation (8). Moreover, it has also been reported that, when the same volume of food and CHO is ingested, the frequency of food ingestion may not be a major factor on muscle glycogen concentration (6). Accordingly, it seems evident that the additional energy expenditure while awake may be responsible for the reduced glycogen content evident on day 2 during SDEP. Accordingly, although functionally the differences in glycogen content may not have been large, it is worth noting that, after sleep deprivation, athletes may commence exercise bouts with a lower muscle glycogen content, which may be harmful to prolonged endurance bouts (1).
Previous research reports evidence indicating a state of reduced muscle glycogen may negatively affect exercise performance (2); thus, the reduction in muscle glycogen during SDEP2 may have had a contributing effect on the subsequent performance decline. Balsom et al. (2) reported that a low-CHO diet significantly reduced preexercise muscle glycogen compared with a high-CHO diet, resulting in a 33% reduction in high-intensity exercise during small-sided football games. In addition, Winnick et al. (35) reported the ingestion of a CHO solution to improve shuttle running and vertical jumps in team-sport athletes and maintain CNS function in the latter stages of exercise compared with a placebo. Although the present study involved a standardized diet, similar findings are evident in that reduced muscle glycogen contributes toward a reduction in high-intensity efforts during self-paced exercise. Similar to the present study, Rauch et al. (23) also reported a down-regulation of pacing strategies during a 2-h cycling time trial during a non-CHO-loaded compared with a CHO-loaded diet. Rauch et al. (23) suggest that the down-regulation in pacing may be related to integrated feedback from the periphery regarding muscle glycogen content. However, owing to the smaller difference in muscle glycogen concentration between conditions in the present study, and the supporting analysis that highlighted no correlation between glycogen content and the main performance parameters, it is likely that muscle glycogen content was not the sole contributing factor to performance declines. It may be suggested that performance and pacing declines were attributed to a combination of sleep deprivation itself and reduced glycogen content and perceptual/mood states after sleep loss.
The reductions in preexercise peak VF on SDEP2 may suggest that sleep loss and/or muscle glycogen content had a direct effect on recruitment of the exercising muscles (19,23). It is, however, unclear as to the level of contribution sleep itself and/or reduced muscle glycogen had on exercise performance, either of which may relate to the reductions observed in VF. In relation to the effect of sleep loss on peak force, Bulbulian et al. (5) previously reported a combined 30-h sleep loss and exercise to reduce flexion and extension peak torque but no effect on a calculated fatigue index, whereas Symons et al. (32) reported no effect of 60-h sleep loss on maximal isometric or isokinetic strength of the upper and lower body. Alternatively, St Clair Gibson et al. (31) reported reduced muscle glycogen concentration to have no effect on maximal isometric contraction force, whereas Nybo (19) reported a reduction in mean force and after placebo compared with glucose supplementation. Nybo (19) also reported that exercise-induced hypoglycemia reduced CNS activation during muscular contractions, which may be influenced by feedback signals from the active muscles. These findings by Nybo (19) concur with the present study in which reductions in VA and pacing strategies were evident during the self-paced exercise bouts; however, the extent to which sleep loss per se affected as opposed to the consequence of reduced glycogen content affect muscle recruitment remains equivocal.
Another possible contributor to the reduction in performance is the elevation of negative mood states and suppression of positive feelings. Sleep loss has been reported to be associated with detrimental effects on mood and perception, primarily assessed with the POMS questionnaire (16). The present study used a modified, shortened POMS questionnaire with the moods included based on applicability for the study design. Despite this shortened version, the results suggested similar findings to Martin and Haney (16) in which participants' moods were affected by sleep loss, despite minimal differences in reported RPE. Given only a small difference in muscle glycogen content was present, it may be suggested that the increased perceptual strain after sleep disruption may have negatively affected pacing strategies and thus reduced intermittent-sprint performance. The effect of negative mood states on exercise performance is supported by Marcora et al. (13), who reported that a state of mental fatigue reduced exercise performance and time to exhaustion during a cycling trial at 80% peak power output. In the present study, despite increased levels of fatigue and tiredness during SDEP2, it seemed that sleep loss had minimal affect on RPE during the ISE. The minimal effect of sleep on RPE may be because of the participants not exercising at the perceived maximal effort at any stage of the exercise protocol, with a progressive increase in RPE during the ISE protocol. The lack of difference in perceived exertion within and between conditions may also be associated with the reduction in work that was completed on day 2 (20). Owing to the self-paced nature of the protocol, participants reduced their workload during SDEP2 to defend a similar rating of perceived exertion to that in CONT2, at the expense of performance.
Because self-paced exercise allows the continual adjustment in exercise intensity, athletes may also have greater control over physiological responses to the given exercise bout through manipulation of pacing strategies. During the present study, HR was significantly lower on day 2 within both experimental trials, which may be due to the slower sprint speeds and reduced distance covered during the self-paced exercise efforts. Vaara et al. (33) also reported that sleep loss reduced HR; however, no exercise protocol was present in this study and was in a rested state, which is difficult to compare with the present study. Oliver et al. (20) reported dissimilar findings to the present study in that core temperature was reduced during 30 min of self-paced treadmill running after sleep deprivation. However, core temperature in the present study was not different between conditions even at rest, suggesting that sleep deprivation had minimal effect on thermoregulatory function during the subsequent day (20). Despite the increased hours of wakefulness, USG measures showed all subjects presented in a euhydrated state before each testing session, excluding differences in hydration as a possible cause of reductions in pacing strategies. Once exercise commenced, sweat rate was greater in CONT2 compared with CONT1 but not different at any other time within or between conditions. These findings are similar to those of Sawka et al. (27), who reported that a reduction in evaporative heat loss in moderate environmental conditions was due to sleep deprivation despite no differences in core temperature. It seems in the present study that because of the self-paced nature of the protocol, decreased physiological responses after sleep loss may be explained by the reduced sprinting speed, distance covered during submaximal bouts, and muscular activation, and in moderate conditions, sleep loss has a minimal effect on thermoregulation.
The effect of sleep loss on pacing strategies during ISE has not been previously examined; however, previous research unrelated to sleep loss suggests factors such as reduced muscle glycogen content (21) and negative perceptual stress (13,23) to be linked to declines in pacing strategies and exercise performance. In relation to the present study, pacing seemed to be manipulated during the exercise protocol, with a progressive decline in distance covered throughout the protocol in both conditions. Interestingly, however, participants were able to increase exercise intensity during the final self-paced efforts, with less distance covered during this "end-spurt" during SDEP2. These findings were mirrored during the hard running efforts, with a progressive decline in distance covered and a final "end-spurt" by an average of 12.5 ± 2.7 m, with the lowest increase during SDEP2. These findings suggest that sleep disruption and/or associated factors reduced the exercise intensity set during self-paced exercise and reduced the capacity to perform a high-intensity "end-spurt," which is similar to a previous work examining self-paced cycling (10). The determinants that affect the chosen intensity and allow this end-spurt are unclear but may be related to feedback regarding the actual sleep loss, reduction in muscle glycogen concentrations (23), or increased perceptual stress (13). The notion of feedback from the periphery is also supported by participants covering the same distance on day 1 between conditions; however, once changes in muscle glycogen and perceptual stress were present, there seemed to be a reduction in exercise intensity. This reduction in exercise intensity coincided with reductions in maximal VF and activation of the right knee extensors after sleep loss. However, potential limitations of the study is that these findings may not be able to be generalized to other subpopulations because of the sample size, and it is difficult to differentiate whether these changes in exercise intensity were due to the sleep deprivation and/or contributing factors associated with a lack of substantial sleep.
In conclusion, sleep deprivation negatively affects pacing strategies and intermittent-sprint performance, which seems to be associated with integrated feedback from the periphery. Afferent feedback including reduced muscle glycogen concentration and increased perceptual strain may have reduced recruitment of active musculature, which was evident before exercise during SDEP2. The level of contribution of these factors including sleep deprivation itself is, however, unclear. From the findings, it is suggested that team-sport athletes provide themselves with adequate time and conditions conducive to sleep and, if not possible, ensure additional CHO are consumed to counteract the reduction in muscle glycogen content.
This research undertaking acknowledges the funding received from Sport and Recreation New Zealand and the Massey University Research Fund.
Further, the authors thank the participants involved in the study, the assistance of Andrea Short, and Massey's Sleep/Wake Centre for assistance with the actigraphy.
The results from the present study do not constitute endorsement by the American College of Sports Medicine. Finally, this article is dedicated to the memory of the late Dr. Johann (Hans) Edge, a colleague and friend who will be greatly missed.
1. Babault N, Desbrosses K, Fabre MS, Michaut A, Pousson M. Neuromuscular fatigue development during maximal concentric and isometric knee extensions. J Appl Physiol
2. Balsom PD, Seger JY, Ekblom B. Physiological response to maximal intensity intermittent exercise. Eur J Appl Physiol
3. Berger RJ, Phillips NH. Energy conservation and sleep. Behav Brain Res
4. Bishop D. The effects of travel on team performance in the Australian national netball competition. J Sci Med Sport
5. Bulbulian R, Heaney JH, Leake CN, Sucee AA, Sjoholm NT. The effect of sleep deprivation and exercise load on isokinetic leg strength and endurance. Eur J Appl Physiol
6. Burke L, Collier G, Davis PG, Fricker PA, Sanigorski AJ, Hargreaves M. Muscle glycogen storage after prolonged exercise: effect of the frequency of carbohydrate feedings. Am J Clin Nutr
7. Chen H. Effects of 30-h sleep loss
on cardiorespiratory functions at rest and in exercise. Med Sci Sports Exerc
8. Driver HS, Taylor SR. Exercise and sleep. Sleep Med Rev
9. Horne JA, Pettitt AN. Sleep deprivation and the physiological response to exercise under steady-state conditions in untrained subjects. Sleep
10. Kay D, Marino FE, Cannon J, St Clair Gibson A, Lambert MI, Noakes TD. Evidence for neuromuscular fatigue during high-intensity cycling in warm, humid conditions. Eur J Appl Physiol
11. Knutson KL, Spiegel K, Penev P, Van Cauter E. The metabolic consequences of sleep deprivation. Sleep Med Rev
12. Krustrup P, Mohr M, Steensberg A, Bencke J, Kjær M, Bangsbo J. Muscle and blood metabolites during a soccer game: implications for sprint performance. Med Sci Sports Exerc
13. Marcora SM, Staiano W, Manning V. Mental fatigue impair physical performance in humans. J Appl Physiol
14. Martin BJ. Effect of sleep deprivation on tolerance of prolonged exercise. Eur J Appl Physiol
15. Martin BJ, Bender PR, Chen H. Stress hormonal response to exercise after sleep loss
. Eur J Appl Physiol
16. Martin BJ, Haney R. Self-selected exercise intensity is unchanged by sleep loss
. Eur J Appl Physiol
17. Mougin F, Bourdin H, Simon-Rigaud ML, Nguyen Nhu U, Kantelip JP, Davenne D. Hormonal responses to exercise after partial sleep deprivation and after hypnotic drug-induced sleep. J Sports Sci
18. Mougin F, Simon-Rigaud ML, Davenne D, et al. Effects of sleep disturbances on subsequent physical performance. Eur J Appl Physiol
19. Nybo L. CNS fatigue and prolonged exercise: effect of glucose supplementation. Med Sci Sports Exerc
20. Oliver SJ, Costa RJS, Laing SJ, Bilzon JLJ, Walsh NP. One night of sleep deprivation decreases treadmill endurance performance. Eur J Appl Physiol
21. Opstad PK, Aakvaag A, Rognum TO. Altered hormonal response to short-term bicycle exercise in young men after prolonged physical strain, caloric deficit, and sleep deprivation. Eur J Appl Physiol
22. Plyley MJ, Shephard RJ, Davis GM, Goode RC. Sleep deprivation and cardiorespiratory function. Influence of intermittent submaximal exercise. Eur J Appl Physiol
23. Rauch HGL, St Clair Gibson A, Lambert EV, Noakes TD. A signalling role for muscle glycogen in the regulation of pace during prolonged exercise. Br J Sports Med
24. Rechtschaffen A. Current perspectives on the function of sleep. Perspect Biol Med
25. Reilly T, Piercy M. The effect of partial sleep deprivation on weight-lifting performance. Ergonomics
26. Richmond L, Dawson B, Hillman DR, Eastwood PR. The effect of interstate travel on the sleep patterns of elite Australian Rules footballers. J Sci Med Sport
27. Sawka MN, Gonzalez RR, Pandolf KB. Effects of sleep deprivation on thermoregulation during exercise. Am J Physiol Regul Integr Comp Physiol
28. Skein M, Duffield R. The effects of fluid ingestion on free-paced intermittent-sprint performance and pacing
strategies in the heat. J Sports Sci
29. Souissi N, Sesboüé B, Gauthier A, Larue J, Davenne D. Effects of one night's sleep deprivation on anaerobic performance the following day. Eur J Appl Physiol
30. Spencer M, Bishop D, Dawson B, Goodman C. Physiological and metabolic responses of repeated-sprint activities. Specific to field-based team sports
. Sports Med
31. St Clair Gibson A, Schabort EJ, Noakes TD. Reduced neuromuscular activity and force generation during prolonged cycling. Am J Physiol Regul Integr Comp Physiol
32. Symons JD, VanHelder T, Myles WS. Physical performance and physiological responses following 60 hours of sleep deprivation. Med Sci Sports Exerc
33. Vaara J, Kyulainen H, Koivu M, Tulppo M, Finni T. The effect of 60-h sleep deprivation on cardiovascular regulation and body temperature. Eur J Appl Physiol
34. Vondra K, Brodan V, Bass A, et al. Effects of sleep deprivation on the activity of selected metabolic enzymes in skeletal muscle. Eur J Appl Physiol
35. Winnick JJ, Davis JM, Welsh RS, Carmichael MD, Murphy EA, Blackmon JA. Carbohydrate feedings during team sport exercise preserve physical and CNS function. Med Sci Sports Exerc
Keywords:© 2011 American College of Sports Medicine
SLEEP LOSS; PACING; TEAM SPORTS; GLYCOGEN CONTENT
Due to inactivity, your session expired and you have been logged out.