Fluid Balance in Team Sport Athletes and the Effect of Hypohydration on Cognitive, Technical, and Physical Performance

Sweat losses in team sports can be significant due to repeated bursts of high-intensity activity, as well as the large body size of athletes, equipment and uniform requirements, and environmental heat stress often present during training and competition. In this paper we aimed to: (1) describe sweat losses and fluid balance changes reported in team sport athletes, (2) review the literature assessing the impact of hypohydration on cognitive, technical, and physical performance in sports-specific studies, (3) briefly review the potential mechanisms by which hypohydration may impact team sport performance, and (4) discuss considerations for future directions. Significant hypohydration (mean body mass loss (BML) >2%) has been reported most consistently in soccer. Although American Football, rugby, basketball, tennis, and ice hockey have reported high sweating rates, fluid balance disturbances have generally been mild (mean BML <2%), suggesting that drinking opportunities were sufficient for most athletes to offset significant fluid losses. The effect of hydration status on team sport performance has been studied mostly in soccer, basketball, cricket, and baseball, with mixed results. Hypohydration typically impaired performance at higher levels of BML (3–4%) and when the method of dehydration involved heat stress. Increased subjective ratings of fatigue and perceived exertion consistently accompanied hypohydration and could explain, in part, the performance impairments reported in some studies. More research is needed to develop valid, reliable, and sensitive sport-specific protocols and should be used in future studies to determine the effects of hypohydration and modifying factors (e.g., age, sex, athlete caliber) on team sport performance.


Introduction
Body water is lost as a consequence of thermoregulatory sweating, and when fluid intake is insufficient to replace sweat losses, hypohydration (a body water deficit) occurs.
Since evaporation of sweat is the primary avenue of heat loss during exercise, fluid losses and the risk of hypohydration in athletes can be significant. The rate of sweat loss is directly related to exercise intensity (metabolic heat production) [1]. Team sports, which are characterized by intermittent bursts of high-intensity exercise over prolonged periods of time (*1-2 h), can elicit heavy sweat losses [2,3]. Other factors that are associated with increased sweating, such as large body mass [4,5], hot/ humid environments [1], and wearing protective clothing/ equipment [6,7], are also present in many team sports. Thus, it is not surprising that some of the highest sweating rates in athletes have been reported in team sports [3,8]. However, individual sweating rates vary considerably [9,10], as do the fluid intake habits of athletes and the ingame fluid replacement opportunities across sports [11]. Thus, the level of hypohydration incurred in team sport athletes can also vary substantially [10,11]. Many studies have measured fluid balance in team sport athletes; however, few have presented a comprehensive summary of the literature [11]. It would be of interest to compare the levels of hypohydration incurred across sports to determine in which team sport(s) hydration education is potentially of greater concern.
It is well established that hypohydration ([2% body mass loss; BML) can impair endurance performance, particularly in hot/humid environments [9,12]. However, the impact of hypohydration on an athlete's performance during team sport competition is less clear. Performance in many team sports is dependent upon cognitive function (e.g., attention, decision making, memory, and reaction time), the execution of sportspecific technical skills (e.g., shooting, passing, and dribbling in soccer), and high-intensity physical abilities (e.g., sprinting, lateral movement, jumping, intermittent high-intensity running capacity). While studies have investigated the effect of hypohydration on some of these aspects of team sport performance, no papers have reviewed and discussed them collectively. There is a need to better understand the potential impact of hypohydration on sport-specific performance to help inform practical recommendations around fluid balance and team sports performance.
The aims of this paper are to: (1) provide a compilation of the fluid balance changes observed in team sport athletes during training and competition, (2) review the literature assessing the impact of hypohydration on cognitive, technical, and physical performance in team sports, (3) briefly discuss the potential mechanisms by which hypohydration could impact team sport performance, and (4) comment on current study limitations and considerations for future directions.
Although racket sports (e.g., tennis) are typically considered individual sports, they are included in this review because they are team sports when played in ''doubles'' competition (a match between two pairs of players) and because of the intermittent high-intensity physical demands and technical skill requirements. Sports that require skill but do not rely heavily upon intermittent bouts of highintensity running (e.g., golf) are not discussed here. Endurance, strength/power, combat, and esthetic sports are also outside the scope of this review.

Fluid Balance Terminology
Body fluid balance is primarily a function of an individual's fluid intake (i.e., hydration practices) relative to his or her fluid losses (i.e., sweat) during exercise. The term ''euhydration'' refers to maintenance of ''normal'' baseline body water content, while the terms ''hypohydration'' and ''hyperhydration'' refer to body water deficits and excesses beyond euhydration, respectively. The term ''dehydration'' is defined as the process of the dynamic loss of body water or the transition from euhydration to hypohydration. The simplest method to assess an individual's acute change in hydration status (or fluid balance) is to compare his/her body mass to baseline values [13]. For example, 3% hypohydration is defined as a water deficit equal to 3% BML. It is acknowledged that a small portion of body mass   Risk for hypohydration is based upon the criteria specified in Fig. 1 b The ranges include mean values reported from each study as well as sub-groups or separate training sessions/matches within studies, where applicable c Where sweating rates were not reported they were estimated from total sweat loss and exercise duration provided in the original paper (in studies where total exercise duration was not reported, sweating rates were not calculated) d The n represents the number of studies that were conducted during training (T) and competition (C) NA not available, RH relative humidity, temp temperature       loss during exercise occurs due to substrate oxidation, that is, non-water mass loss through expiration of carbon dioxide. The reader is referred to other papers that discuss potential errors in hydration assessment methodologies in greater detail [13,14]. Importantly, the BML method of hydration assessment and related terminology have been used in all relevant studies identified for the aims of this review and, therefore, will be used throughout the discussion that follows. Intuitively, the factors that elevate risk of hypohydration are those that increase thermoregulatory sweat loss (hot/ humid environment and high exercise intensity) or limit fluid replacement (low availability of fluid or opportunity for drink breaks). Sweating rate and/or fluid balance has been researched the most in soccer , followed by American Football [36][37][38][39][40][41][42][43][44][45][46][47][48], tennis [49][50][51][52][53][54][55][56][57][58], basketball [15,[59][60][61][62][63][64][65][66], rugby [67][68][69][70][71][72][73], and ice hockey [74][75][76][77][78]. Australian Rules Football [79,80], cricket [81][82][83], baseball [84,85], futsal [17], Gaelic Football [86], netball [15], beach volleyball [87], court volleyball [88], field hockey [89], badminton [90], and water polo [91] have been studied to a lesser extent. Table 1 provides a summary of the findings by sport organized per level of risk for hypohydration according to factors identified in Fig. 1.

Fluid Balance in Team Sports
In soccer, a broad range in mean sweating rate has been reported (0.3-2.5 L/h). This is due, in part, to the varied . High mean sweating rates have been reported in American Football (0.6-2.9 L/h), with the large body size of players partially responsible [4,8,40,41], as well as the equipment/uniform requirements [6,7]. In addition, all studies in American Football have been conducted during preseason training, which typically occurs in warm-hot summer weather [3]. Despite high sweating rates, the observed disturbances in fluid balance have generally been mild [3], with the exception of one study [36] reporting a mean BML of *2.3-2.4% (SD: 1.1-1.9%) in players practicing in full pads under high levels of heat stress (29-32°C WBGT). Similar results have been reported in rugby and Australian Football, where high sweating rates led to significant hypohydration in some [67,80] but not all studies [67,70]. High mean sweating rates have also been reported in basketball [15,59,64,65], tennis [50,51,58], ice hockey [74][75][76], and beach volleyball [87]; however, mean BML was \2%, suggesting that drinking opportunities were sufficient to provide most athletes with enough fluid to offset sweat losses.
Some studies have tested athletes multiple times to determine intra-individual variability in sweating rates and fluid balance. For example, Australian Rules Footballers exhibited higher sweating rates and accrued greater levels of hypohydration during high-intensity training and game simulations (1.1-1.3 L/h and 3.4-3.5% BML) compared with low-intensity training (0.8 L/h and 2.1% BML) [34]. Similarly, studies in soccer [16,22] and tennis [58] have reported that warm-hot conditions elicited higher sweating rates during match play than cool-temperate environments. While higher rates of sweat loss can lead to greater body mass deficits [22,58], this is not always the case [15,16]. For example, Mohr et al. [16] found that soccer players accrued the same level of hypohydration (1.8 and 1.9% BML, respectively) during indoor (21°C) and outdoor (43°C) match play despite significantly different sweating rates (1.6 vs. 2.5 L/h, respectively). Taken together, it seems that the sports or conditions associated with higher sweating rates are not always associated with the higher levels of hypohydration. Factors related to fluid availability (type and amount), drinking opportunities (per the rules and structure of the game), exercise duration, hydration education, and personal preferences also play an important role in determining fluid balance [11].  [93] used a field study approach and complex data modeling to determine the relative contribution of various factors, including BML, on cognitive performance in male university soccer players. In this study two trials were completed; players drank water ad libitum in the first match and were encouraged to drink a sports drink and/or water in the second match. Matches were played in a hot environment (34°C, 64-65% relative humidity). Before, at halftime, and after each match the players completed a battery of cognitive tests, which included fine motor speed (finger tapping test), visuomotor reaction time (visual sensitivity test), visuospatial working memory (Corsi block test), and working memory simple reaction time (Sternberg test). Bandelow et al. [93] reported that hypohydration (up to 2.5% BML) impaired working memory reaction time, but had no effect on any other measure of cognitive performance. Instead, maintenance of blood glucose (presumably from sports drink consumption) and core temperature changes played more important roles in determining speed and accuracy during the cognitive battery [93].

Basketball
Two basketball studies have tested the impact of hypohydration on cognitive performance, with mixed results [96,97]. Hoffman et al. [96] found no differences in visual reaction time during a hand-eye reaction test (Dynavision Fluid Balance and Team Sport Performance 1967 D2) when female college players drank water versus no fluid to accrue 2.3% BML. By contrast, the number of successful attempts during a lower body reactive agility test (Quick Board TM ) was significantly lower in the nofluid trial [96]. In addition, Baker et al. [97] found that hypohydration was associated with impaired vigilance in male players. In this study, subjects performed the test of variables of attention at baseline, after exercise-heat stress to induce 1-4% hypohydration or maintain euhydration (i.e., pre-game), and then after a simulated basketball game (where target levels of BML were maintained). The players made significantly more errors of omission and commission and had slower response times (by *6-8%) in the 1-4% BML trials than the euhydration trial [97]. In this study there were no differences in vigilance between the graded levels of hypohydration [97].

Field Hockey
The effect of hypohydration on cognitive performance has been tested in one field hockey study by MacLeod and Sunderland [98]. On day 1 of this study, elite female field hockey players underwent 2 h of passive heat stress to stimulate fluid loss. After completion of this dehydration phase, subjects' fluid intake was controlled such that the following morning (day 2) they were either euhydrated or 2% dehydrated. On day 2, players completed a field hockey skill test [99] before and after a 1-h intermittent treadmill protocol in which players drank water ad libitum (maintained BML difference between trials). Decision-making time during the skill test was 7% slower in the 2% hypohydrated versus the euhydration trial, but only before (not after) the intermittent exercise [98]. Thus the effects of hypohydration on cognitive performance seem to be inconsistent in this study [98]. Nonetheless, it is interesting to note that the study by MacLeod and Sunderland [98] has been the only one to employ a cognitive test that is sportspecific as opposed to others (discussed above in the Sects. 4.1.1 and 4.1.2) that used tests originally designed for the general population.

Multiple Sports
In a study of college lacrosse and American Football players, D'Anci et al. [100] compared the effects of no fluid versus water intake (euhydration) during 60-to 75-min high-intensity practices on performance during a subsequent cognitive test battery. The players accrued 1.8% BML and 1.2% BML in the no-fluid trials of study 1 and study 2, respectively. Vigilance performance was impaired by 3-4% when athletes were 1.8% hypohydrated, but was not impacted by 1.2% hypohydration. None of the other measures of cognition in the test battery (e.g., memory, reaction time, visual perception, math, and map planning) were affected by hydration status in either study.

Summary
Based on the results of seven studies completed to date, the impact of hypohydration (most studies involved *1 to *2.5% BML) on cognitive performance of team sport athletes is equivocal. In four studies, vigilance, decisionmaking time, working-memory reaction time, or reactive agility were impaired with hypohydration; however, across all studies no other measure of cognition (e.g., mental concentration, fine motor speed, visual perception, visuomotor reaction time, math) was affected. This inconsistency is likely due in part to the aspects of cognition measured, types of cognitive tests used, the reliability and sensitivity of these tests, and other factors related to study design. Cognitive performance is difficult to measure, particularly in the context of sports, and very few studies have employed tests directly relevant to team sports performance. More work is needed to develop and validate sport-specific cognitive performance assessments.

Basketball
The potential impact of hypohydration on basketball shooting performance has been assessed in six studies [60,63,96,[101][102][103]. Most of these studies have investigated *2% hypohydration and found mixed results. For instance, Hoffman et al. [96,101] found no impact of fluid restriction (1.9 and 2.3% BML) versus water intake on shooting performance by male youth [101] or female college players [96]. However, it is interesting to note that field goal percentage decreased by 8.1% (not statistically significant) from the first to second half in the no-fluid trial, but was maintained (1.6% increase) when male youth players were allowed to drink water [101]. Similar results were found in a study comparing the effects of no fluid (2.5% BML) versus ad libitum water intake (1.1% BML) in male youth players; Carvalho et al. [60] found no significant differences in shooting performance between conditions, but did report a non-significant 5.8% lower two-point field goal accuracy with no fluid. In another study with male youth players, Dougherty et al. [103] reported that 2% hypohydration was associated with significantly lower shooting percentage (by 8% combined for all shots) compared with euhydration.
In a descriptive study, Brandenburg and Gaetz [63] allowed elite female players unlimited access to the drink of their choice (sports drink and or water) during two international games. In both games the players accrued up to *2.0% BML. The authors reported a significant inverse relation between BML and field goal percentage (r = -0.61) in the second game, but no relation in the first game. While measuring the impact of hydration status on performance during actual game play increases the ecological validity of this study, the interpretation of results is limited due to the potential impact of confounding factors (e.g., concomitant carbohydrate ingestion, additional dietary intake behaviors, and defensive prowess by the opposing team).
The effects of graded levels of hypohydration on basketball shooting performance have been tested in one study of male players. Baker et al. [102] found that, compared with euhydration, increasing levels of hypohydration (2-4%) led to a progressive 7-12% decrease in the total number of shots made during a simulated game. However, there were no differences in shooting percentage between trials. In this study, players were allotted a standardized time to make as many shots as possible during each drill. Thus, the decrease in shots made with hypohydration were partially a result of fewer shot attempts due to slower sprinting and dribbling speeds between shots [102]. Taken together, the results of the six studies in basketball suggest that C2% hypohydration can potentially impact shooting performance, perhaps due to decreasing shooting accuracy and/or slowing the frequency of shot attempts. Both factors can impact the total number of points scored, which plays an important role in dictating the outcome of a basketball game [104].

Soccer
Three studies [95,105,106] have tested the impact of fluid restriction on soccer-specific skills performed before, during, and after a 90-min intermittent protocol (Loughborough intermittent shuttle running test; LIST [107]). Ali et al. [106] reported no impact of fluid restriction (2.2% BML) versus water intake (1.0% BML) on passing performance [108] in female Premier division players. Similarly, Owen et al. [105] found that passing and shooting skills [108] of male semiprofessional players were unaffected by fluid restriction (2.5% BML) compared with ad libitum (1.1% BML) or prescribed (0.3% BML) water intake. By contrast, and also in semiprofessional soccer players, McGregor et al. [95] found a 5% deterioration in dribbling skill during the LIST with no fluid ingestion (2.4% BML) while skill performance was maintained during the fluid intake trial (1.4% BML). Taken together, these results suggest that the effect of hypohydration on soccer performance may be dependent upon the type of skill measured, albeit more research is needed and, in particular, with cohorts varying in competitive level.

Cricket
The effect of hypohydration on skill in well-trained cricket players has been investigated in two studies [83,109]. Devlin et al. [109] compared the effects of fluid restriction (2.8% BML) versus prescribed fluid intake (0.9% BML) during 1-h intermittent exercise-heat stress (28°C) on performance of a subsequent bowling skill test. Compared with prescribed fluid intake, fluid restriction was associated with a *15-16% impairment in bowling accuracy, but had no effect on bowling velocity. Recently, Gamage et al. [83] found that fluid restriction (3.7% BML) during 2-h cricket training in the heat (27-33°C, 66-89% relative humidity) led to a 1-7% decrease in speed and 14-22% decrease in accuracy of bowling [110] and throwing [111] tests, whereas performance during the cricket skill test was maintained in the fluid provision trials (0.9% BML) [83].

Field Hockey
One study has investigated the potential effects of previous day passive heat stress-induced hypohydration on field hockey skill. MacLeod and Sunderland [98] employed a test involving dribbling, passing, and shooting at an illuminated target after a 60-min intermittent treadmill protocol with ad libitum drinking [99] and found no impact of 2% hypohydration versus euhydration on field hockeyspecific skill in elite female players.

Tennis
There are limited data available on the effects of hypohydration on skill performance in racquet sports. One study has reported no differences in post-match tennis shot accuracy during a ball machine test when male and female players drank water (1.1% BML) or no fluid (2.7% BML) during a 2-h simulated tennis match [112].

Summary
The effect of hypohydration on skill performance seems to be inconsistent across sports. Studies suggest that *2-4% hypohydration can impair shooting performance in basketball and bowling/throwing in cricket. By contrast, the balance of studies suggests a minimal impact of *2-3% hypohydration on skill performance in soccer, field hockey, and tennis. Like cognition, skill is difficult to measure, and more work is needed to develop reliable, valid, and sensitive sport-specific tests to use in future studies investigating the impact of hypohydration on skill.

Basketball
The impact of hypohydration on sprint performance has been assessed in three basketball studies [60,102,103]. During a simulated basketball game with male players, increasing levels of hypohydration (2, 3, and 4% BML) were associated with progressively longer (i.e., sprint performance was decreased) total sprint times (by 7, 8, and 16%, respectively) [102]. In a study of youth male players, hypohydration (2% BML) led to a significant 6% longer total and mean sprint time throughout a simulated game compared with euhydration [103]. However, there was no effect of fluid restriction (2.5% BML) versus ad libitum water intake (1.1% BML) on sprint performance after training in another study of youth male basketball players [60].

Soccer
Three studies have tested the impact of hypohydration on 15-m sprint performance in soccer players during the LIST protocol [95,106,113]. Ali et al. [106] reported no difference in mean sprint time between trials in which female Premier division soccer players drank water (1.0% BML) or no fluid (2.2% BML) throughout the LIST protocol. In a similar study design with male university soccer players, Ali and Williams [113] reported no impact of water restriction (3.7% BML) versus water ingestion (2.3% BML) on mean 15-m sprint time. On the other hand, McGregor et al. [95] found that the mean 15-m sprint time of male semiprofessional soccer players was significantly longer in the last 15-min block of the LIST protocol when fluid was withheld (2.4% BML) versus when allowed to drink (1.4% BML) during the 90-min LIST.

Batting Sports
Hypohydration and sprint performance in baseball and cricket have been investigated in two studies [83,114]. In college baseball players, hypohydration by 3% BML (induced by previous day exercise-heat stress) was associated with a significant 3-4% longer mean time to complete 30-m sprints during the latter bouts of an intermittent sprinting protocol [115] compared with euhydration [114]. In a study of male elite cricketers, Gamage et al. [83] also found impaired sprinting performance as a result of hypohydration. Sprint time increased significantly (by 2.2%) when fluid was restricted to 4 ml/kg/h (3.7% BML) throughout 2 h of cricket training, whereas performance was maintained from pre-to post-training when 12-15 ml/ kg/h fluid was provided (0.9% BML) [83].

Summary
The effect of hypohydration on sprint performance has been measured in eight studies across four sports. For soccer, the balance of the literature suggests that *2-4% hypohydration is unlikely to impact mean 15-m sprint performance throughout an entire bout of 90-min intermittent exercise, but may prolong sprint time in the latter stages (e.g., last 15 min of a 90-min session). Results are more consistent in basketball and batting sports, with most studies reporting longer time to complete sprints when athletes are hypohydrated by *2-4% in basketball or *3-4% in baseball and cricket. More research is needed to determine the impact of hypohydration on sprint performance in other team sports.

Basketball
Three studies have employed similar lateral slide drills to simulate defensive movements in basketball [60,102,103]. Baker et al. [102] reported that the time to complete defensive slide drills by male players throughout a simulated game was not impacted by 1-2% hypohydration, but was significantly longer with 3-4% hypohydration compared with euhydration. In another study, Dougherty et al. [103] found that 2% hypohydration was associated with *7% longer defensive slide times in youth male players throughout a simulated game. However, Carvalho et al. [60] found that defensive slide times after training were not different between fluid restriction (2.5% BML) and ad libitum water intake (1.1% BML) trials in male youth players.

Summary
The ability to make quick lateral movements is important for performance in many sports, such as defensive sliding in basketball, fielding in baseball, or returning a groundstroke in tennis. However, the effect of hypohydration on performance of sport-specific lateral movements has only been tested in basketball. The mixed results reported in these studies suggest that the impact of hypohydration, ranging from *1 to 4% BML across studies, is currently unclear. More research on sport-specific lateral movement performance is needed in basketball and other relevant team sports.

Basketball
Four basketball studies have investigated the effects of hypohydration on jumping performance, including maximal jump height [96,[101][102][103], time to complete a set number of jumps [102,103], and peak or mean anaerobic power during repeated jump tests [96,101]. These studies report no impact of hypohydration (*1-4% BML) on maximal jump height [101][102][103]. However, Baker et al. [102] reported significantly longer repeated jump time with 4% hypohydration versus euhydration. In addition, Hoffman et al. [101] found that post-game anaerobic power [116] was 19% lower when fluid was restricted (1.9% BML) versus when water intake was permitted, although this difference did not reach statistical significance.

Baseball
One study has measured the effect of graded dehydration on anaerobic power in college baseball players. In a crossover study, Yoshida et al. [85] induced 0.7, 1.7, 2.5, and 3.9% BML in players during a 3.8-h practice in the heat (29°WBGT) by having them drink to replace 80, 60, 40, and 20% of fluid losses, respectively. Maximal anaerobic power during a 10-s cycling test was decreased significantly by *13% from pre-to post-exercise with 3.9% BML, but there was no significant change with the lower levels of hypohydration.

Tennis
One study has compared the effects of ingesting water (1.1% BML) or no fluid (2.7% BML) during a 2-h simulated match on maximal jump height and anaerobic power (Sargent jump test) in male and female tennis players. In this study, Burke and Ekblom [112] found no change in performance from pre-to post-practice with either fluid intake condition.

Summary
Jump height and anaerobic power are critical to performance in many team sports; however, only six studies have measured the potential effects of hypohydration in sportspecific studies. These studies suggest that hypohydration is unlikely to have a negative impact on vertical jump height. However, anaerobic power may be impaired by hypohydration, especially at higher levels of hypohydration (*4% BML). In general, these results are in agreement with recent reviews and meta-analyses on the effect of hypohydration on jumping ability and anaerobic power [12,117]. Nonetheless, more research is needed to understand how hypohydration may impact jump height and anaerobic power in the context of team sport performance. Finally, while a recent meta-analysis concluded that *3% BML may improve body mass-dependent tasks such as vertical jumping ability [117], this has not been found in the team sport studies reviewed here. As demonstrated by Cheuvront et al. [118], the theoretical improvement in jump height associated with a dehydration-induced body mass deficit may be offset by an inability to produce the same degree of contractile force, thus confounding the interpretation of how hypohydration affects body massdependent tasks.

Soccer
Two studies [94,105] have employed the Yo-Yo intermittent recovery test [119,120] to determine the effect of hypohydration on intermittent running capacity in soccer. Owen et al. [105] measured performance during the Yo-Yo test in male semiprofessional soccer players before and after they completed the LIST protocol. There were no differences in Yo-Yo performance between trials in which players drank no fluid (2.5% BML), water ad libitum (1.1% BML), or water to replace *90% of fluid losses (0.3% BML) during the LIST protocol. By contrast, in another study, 13-15% less distance was covered during the Yo-Yo test when male soccer players were 2.1% (water mouth rinse) and 2.4% (no fluid) hypohydrated versus when they were allowed water ingestion (0.7% BML) during 45 min of cycling followed by a 45-min match [94].

Cricket
In a study of male, well-trained bowlers, Devlin et al. [109] compared the effect of fluid restriction (2.8% BML) versus prescribed fluid intake (0.5% BML) during 1 h of intermittent exercise-heat stress on subsequent performance of a maximal multi-stage shuttle run [121]. Intermittent running capacity was significantly impaired when fluid was restricted, as the bowlers completed 7.7% fewer shuttles in the 2.8% BML versus 0.5% BML trial.

Summary
For many team sports, the capacity to sustain high intensity efforts alternated with rest or lower intensity periods throughout a game is critical to the success of an athlete. To date, two out of three studies have found a detrimental effect of 2-3% hypohydration on intermittent running capacity. However, more research is needed, particularly on the sports that are highly dependent upon intermittent running capacity (e.g., soccer, rugby, field hockey, and basketball).

Potential Mechanisms and Modifying Factors
for the Effects of Hypohydration on Performance

Overview of Physiological Effects of Hypohydration During Exercise
Because sweat is hypotonic compared with plasma [122], exercise-induced hypohydration is associated with an increase in plasma osmolality and a decrease in plasma volume (i.e., hyperosmotic hypovolemia). Hypovolemia results in a decrease in stroke volume and a compensatory increase in heart rate to maintain a given cardiac output [123][124][125]. Hypovolemia and hyperosmolality delay the onset and decrease the sensitivity of the sweating and skin blood flow responses to hyperthermia [126][127][128], thus increasing heat storage [125,129]. Consequently, exercise performance that is dependent upon the cardiovascular and thermoregulatory systems, such as aerobic exercise in the heat, can be impaired by hypohydration [9,12]. The physiological mechanisms underlying the effect of hypohydration on aerobic performance have been well studied (for reviews, see Sawka et al. [130,131] and Cheuvront et al. [132]). By contrast, much less is known about the potential mechanisms for the detrimental effects of hypohydration on team sport performance. The next section summarizes the proposed mechanisms by which hypohydration could impair cognition, technical skill, and physical performance related to team sports.

Cognition
The effect of hypohydration on cognition has been widely researched. While decrements in cognitive performance with hypohydration have been reported in some studies of athletes [97,98,100,133], healthy young adults [134][135][136][137][138], and military personnel [139,140], other studies have found no effect of hypohydration [94,95,[141][142][143][144][145][146]. Furthermore, a clear mechanism by which hyperosmolality or hypovolemia per se would impair cognition is currently lacking (for a review, see Cheuvront & Kenefick [12]). In brief, hypohydration has been suggested to mediate decrements in brain function by decreasing cerebral blood flow, reducing brain volume, or increasing blood-brain barrier permeability. However, a consistent effect of hypohydration on these measures of brain function [147][148][149][150][151][152][153] at the levels of BML typically reported in the cognition literature (e.g., 1-4% BML in team sport studies) has not been found. An alternative explanation, previously described by Cheuvront and Kenefick [12], is that symptoms of hypohydration, such as thirst, headache, or negative mood states (e.g., fatigue), may distract subjects during cognitive tasks and subsequently impair performance. Moreover, individual variability in cognitive resiliency (ability to overcome the stressors of hypohydration) may explain, in part, the equivocal findings in the cognition literature [12]. For example, Szinnai et al. [145] found that 2.6% BML induced by water deprivation had no impact on cognitivemotor function, but significantly increased ratings of perceived effort and concentration necessary for test completion. Furthermore, Kempton et al. [147] showed that although mild hypohydration did not impair cognitive performance or cerebral perfusion, higher levels of neuronal activity (as indicated by a greater increase in the fronto-parietal blood oxygen-level-dependent response) were required to perform an executive function task. Thus, it may be that some individuals are better at increasing concentration sufficient to overcome symptomologic distracters of hypohydration and achieve the same level of performance as that of a euhydrated state [12]. In the team sport literature reviewed in Sect. 4.1, hypohydration consistently increased ratings of thirst, perceived exertion, and fatigue, but subsequent effects on cognitive performance were equivocal.

Physical Performance
In team sports high-intensity efforts are performed within the context of intermittent exercise over a prolonged period of time (1-2 h). Thus, it is plausible that reductions in aerobic capacity [154][155][156][157] or muscle endurance [117,158] that have been shown to occur with hypohydration, could help explain the impaired physical performance (sprinting, lateral movements, and intermittent running capacity) reported in studies mimicking the demands of team sports training/play. In addition, hypohydration has been shown to result in decreased muscle blood flow [159,160] and alterations in skeletal muscle metabolism (increased lactate, muscle glycogenolysis, and carbohydrate oxidation) [160][161][162][163]. However, to date these findings have mostly been documented in prolonged cycling exercise, which is likely a result of the difficulty in obtaining invasive physiological measurements in team sport athletes on the field of play. To our knowledge, only one team sport performance study has measured the effect of hydration status on markers of muscle metabolism. Ali and colleagues [106] found that blood lactate concentration was significantly higher (7.2 vs. 3.7 mmol/L) when female soccer players drank no fluid (2.2% BML) compared with when they ingested fluid during the 90-min LIST protocol (1.0% BML). However, in this study, sprint and skill performance were not impacted by hydration status [106]. It is also interesting to note that, in the studies reviewed (Tables 2,  3), performance was no more likely to be impaired in sports with high aerobic demands (e.g., soccer) than sports that have more rest opportunities (e.g., basketball) or are lower intensity (e.g., baseball, cricket). This is somewhat surprising given the reported detrimental effects of hypohydration on endurance performance [9,12], but it is likely that study limitations and various modifying factors play a role in the discrepancy in these findings (discussed in more detail in Sect. 5.5).
Another potential mechanism to consider is the hyperthermic effect of hypohydration, as some sport-specific studies have reported higher body core temperatures with fluid restriction versus fluid intake [94,102,103]. Drust et al. [164] reported that elevated core and muscle temperatures during a 40-min intermittent cycling protocol were associated with impaired repeated sprint performance. The authors concluded that the results may be related to the influence of hyperthermia on central nervous system function. Central fatigue, as indicated by an impaired ability to sustain maximal muscle activation during sustained contractions, has been implicated in exercise performance decrements associated with hyperthermia [165][166][167]. It is thought that multiple factors (including core and skin temperature) likely provide afferent inputs for central nervous system integration and reduce motor drive to skeletal muscles. The reader is referred to a review by Nybo et al. [165] for a recent comprehensive discussion on physiological factors governing hyperthermia-induced fatigue. It is important to note that while hyperthermia (increased core temperature) can impair performance during prolonged exercise, increased muscle temperature could enhance certain aspects of physical performance [168]. In particular, improved sprinting performance has been reported in soccer [16,169], perhaps as a result of improved muscle contractile properties and anaerobic power [170,171], in hot versus temperate environments. However, these changes occur irrespective of hypohydration, as Mohr and colleagues [16] found faster peak sprinting speed in hot versus temperate conditions when players accrued similar levels of BML between trials with ad libitum fluid intake (1.9 and 1.8%, respectively). As such, optimal performance strategies may involve both maintenance of muscle temperature as well as the limitation of excessive hypohydration.
Finally, as with cognition, it is possible that psychological factors are also involved in hypohydration-induced decrements in physical performance. Negative mood states and other stressors associated with fluid restriction may distract athletes from giving their full effort toward performing the high-intensity exercise task. In support of this notion, most (10 of 11) team sport studies that measured subjects' perceived exertion or fatigue found that ratings were significantly elevated in conditions of fluid restriction versus fluid intake (see Table 3). In addition, there may be some interplay between familiarization with hypohydration, perceived exertion, and the effect of hypohydration on performance. Although not specific to team sports, Flemming and James [172] reported some support for this concept in recreationally active men. In this study, 2.4% BML impaired 5-km treadmill running performance (by 5.8%) when subjects were unfamiliar with the hypohydration protocol. However, there was an attenuation of subjects' ratings of perceived exertion and the performance decrement (1.2%) after completion of four familiarization sessions designed to habituate subjects with the hypohydration protocol. While these novel data are interesting, more research is needed before definitive conclusions can be made regarding the effect of familiarization with hypohydration on ratings of perceived exertion and performance.

Sport-Specific Technical Skills
The execution of sport-specific skill is a complex process, as it is dependent upon several aspects of physical and cognitive function. For example, a successful shot attempt during a basketball game requires a combination of fine motor (ball control) and gross motor (balance and coordination) skills, physical abilities (power, strength, and speed), concentration, and decision-making skills, among other factors. Thus, if hypohydration impairs cognition or physical performance, either directly through hyperosmotic/hypovolemia-induced changes in physiological function or as a byproduct of the distracting symptoms of hypohydration, then these mechanisms could also account for impaired execution of technical skills.
Other proposed mechanisms include changes in vestibular function, as some studies have reported impaired postural balance (increased body sway) with hypohydration after exercise [173][174][175][176]. However, other studies have reported no impact of hypohydration on balance control [133,141,177]. Theoretically, balance is more likely to be impaired when hypohydration is combined with hyperthermia [175] or fatigue from previous exercise [173,174]. However, Seay et al. [177] reported no relation between standing balance and levels of hypovolemia or hyperosmolality. As such, a clear physiological mechanism by which hypohydration could impair postural control has not yet been identified.
Future studies are needed to elucidate which physiological mechanisms or combination thereof may account for the detrimental impact of hypohydration on cognition, technical skill, and physical performance reported in some studies. For more details on potential mechanisms underlying the impact of hypohydration on various aspects of performance the reader is referred to previous reviews [12,130,132].

Modifying Factors
Many methodological differences among studies likely contribute to the inconsistent results reported across the literature. For example, the subject characteristics (e.g., sex, age, caliber of athlete), method of dehydration (e.g., passive heat, exercise, or exercise-heat stress), and/or BML differences between hypohydration and control trials varied considerably among studies. Thus, a relevant question that follows is: Do certain factors modify the impact of hypohydration on performance (i.e., are there interaction effects)? To date, no studies have addressed this question directly. However, when comparing studies across the literature (see Table 4) there seems to be no clear pattern regarding the impact of sex, age, or athlete caliber on the effects of hypohydration on team sport performance. This is due in part to the limited data available, as only six studies have included female subjects [63,96,98,100,106,112] and only three studies have tested youth athletes [60,101,103]. A broad range of athlete calibers have been tested across studies with equivocal results within caliber (see Table 4), suggesting that a particular level of athlete is not more or less likely to be negatively affected by hypohydration based on the currently available data in team sports. However, studies directly comparing team sport athletes with different skill levels are needed.
As shown in Table 4, one of the factors that does seem to modify the impact of hypohydration on performance in team sport athletes is the method of dehydration. When dehydration was induced via exercise in the heat, subsequent performance was usually impaired with respect to cognition, skill, sprinting, lateral movements, jumping/ power, and intermittent running capacity (the only exception was jumping performance in one study [103]). By contrast, when dehydration was induced via exercise alone, subsequent performance was impaired in B 50% of the studies within each of the performance categories. This finding is perhaps not surprising given the well-established deleterious effect of environmental heat stress and subsequent heat strain on aerobic performance [130,155,165,178] and muscle function [158,164,179], as well as mood states and perceived exertion [165,180,181], which may in turn impact aspects of team sport performance. It is important to note that the studies using heat and exercise to induce hypohydration also involved higher levels of BML. In addition, some of these studies included a rest period between the dehydrating exercise/heat protocol and the commencement of performance drills to allow body core temperature to return to baseline values [97,102,103]. Thus, it could be argued that hypohydration per se was responsible for the impaired performance. Nonetheless, in real life it can be difficult to separate out the effects of hypohydration and heat stress, as the two are closely linked when training/competing in warm-hot environments (i.e., exercise in the heat increases sweat rate thereby magnifying fluid losses) [9].
Another factor related to the method of dehydration is the timing of body water loss with respect to completion of the sport-specific protocol and performance tests. Most studies were designed to determine the effects of dehydration, accrued throughout sport-specific training/play, on performance. Mixed results were reported with this methodological approach. By contrast, some studies established hypohydration in the hours before [97,102,103] or in some cases the day before [98,114] the sport-specific tests. A detrimental effect of hypohydration was reported in all five of these studies (albeit most of these studies also involved higher levels of hypohydration ([2% BML) and/or heat stress). This methodological approach allows for a more systematic investigation of the effects of hypohydration (e.g., a standard level of BML) and provides insight on performance effects when athletes begin training/competition in a hypohydrated state (which may be applicable for tournaments or back-to-back training sessions). However, it is not applicable to scenarios where athletes begin exercise in a euhydrated state, and, as such, impacts on the ecological validity of the study findings. It would be interesting for future research to directly compare the effects of previous dehydration versus in-game dehydration on performance.
The level of BML reached in hypohydration trials and the inclusion of a proper euhydration control trial are other important factors to consider. In the studies reviewed, most included a control (fluid intake) trial to compare performance against that of hypohydration (i.e., fluid restriction) trials. However, across studies, varying degrees of BML accrued during the control trials. That is, some studies aimed to replace fluid losses and maintain euhydration (\ 1% BML [9]), while others involved ad libitum intake or prescribed a fixed volume of fluid intake that resulted in mild to moderate hypohydration in the control trials. For example, several studies in soccer dehydrated athletes to *2-3% BML in the fluid restriction trials, but also accrued 1-2% BML in the fluid intake trials [95,105,106,113]. By contrast, in most basketball [97,102,103], baseball [85,114], and cricket [83,109] studies, subjects maintained euhydration (\ 1% BML) in the control trials. The differences in study design are likely due, in part, to attempts to implement ecologically valid fluid intake patterns, which reflect differences in drinking opportunities  Table 4), BML differences between control trials and hypohydration trials were calculated for each study. In this regard, most investigations involved a 1-2% BML difference between control trials and hypohydration trials. In these studies, there were mixed results in how hypohydration impacted cognition, sprinting, lateral movements, and intermittent running capacity, and no or little effect of hypohydration on jumping/power and skill. Only five studies involved a 3-4% BML difference between trials, but all found impaired performance with hypohydration [83,85,97,102,114]. It is important to note that these studies also involved exercise in the heat as the method of dehydration, so it is unclear whether the heat stress also contributed to the performance impairment. Nonetheless, two out of three studies involving graded levels of hypohydration (* 1-4% BML) have indicated that *3-4% BML was more likely to impair performance than *1-2% BML [85,97,102]. Figure 2 shows a Venn diagram illustrating the likelihood of team sport performance impairments with hypohydration. Based on the studies reviewed, decrements in cognitive, technical, or physical performance seem more likely with higher levels of hypohydration and heat stress. However, as discussed in Sects. 5 and 6, other factors may play a role, but currently lack sufficient research in team sports. These include high aerobic demand, hypohydration at baseline, and individual differences in the response to hypohydration (e.g., low cognitive resiliency).

Study Limitations
The potential limitations of individual studies reviewed in this paper are described in Tables 2 and 3. One of the most common potential limitations is the inherent difficulty in blinding subjects to the fact that they are dehydrating (e.g., no fluid or strict fluid restriction) versus rehydrating (e.g., prescribed or ad libitum fluid intake) during a given trial. Because the awareness of being dehydrated may confound performance results, attempts should be made to disguise experimental conditions. For example, small volumes of fluid should be provided during dehydration trials and the subjects' body mass, fluid intake, and urine volumes should be concealed [182]. Ganio  Based on the studies reviewed, the likelihood of performance impairment seems to increase with higher levels of hypohydration and heat stress.
The circle with the dashed line represents other factors that may play a role, but require more research in team sports; these include high aerobic demand, hypohydration at baseline, and individual differences in the response to hypohydration (e.g., low cognitive resiliency). BML body mass loss hypohydration without hyperthermia in men and found that 1.6% BML decreased vigilance and working memory and increased tension/anxiety and fatigue. While these masking techniques can be helpful, it remains difficult to effectively blind subjects to experimental conditions when attempting to induce higher levels of hypohydration (e.g., 3-4%). Another common study limitation is related to the type of test used to measure the effect of hypohydration on performance. Many different tests have been used and, in some studies, limited information about the test was reported. Protocols and tests used to measure performance in team sports should be sport-specific (i.e., mimic the actual physical demands and skills required of the sport) and subjects should be familiarized with the methods prior to the start of experimental trials. The tests should also be valid, reliable, and sensitive [183]. Of the studies reviewed, most tests were sport-specific for skill, sprinting, jumping/power, lateral movements, and intermittent running capacity, but not for cognition (with the exception of one field hockey study [98]). Most cognitive tests were, however, valid and reliable standardized tests. The validity and reliability of skill tests were not reported in the basketball, tennis, and cricket studies reviewed, but were reported in the soccer and field hockey studies. Interestingly, the effect of hypohydration on skill performance was mixed whether validity and reliability were reported or not (see Table 2).

Considerations for Future Directions
From the discussion above it is clear that more research is needed to address several remaining questions regarding the potential impact of hypohydration on team sport performance. First, valid, reliable, and sensitive sport-specific protocols should be developed and used in future studies to ensure that tests are able to detect small but meaningful differences in performance. In general, valid/reliable sportspecific tests to measure cognition and skill are currently limited in most sports.
Most studies have tested the effect of low-moderate levels of hypohydration on performance. In future studies, it would be helpful to include higher levels of hypohydration, perhaps in a graded manner. In addition, studies directly comparing the effect of hypohydration on different cohorts, such as male versus female, youth versus adults, or low-versus high-caliber athletes, would be helpful in determining who may be more susceptible to the detrimental effects of hypohydration, from both a physiological (heat safety) and a performance perspective. In most studies of the current literature the amount and pattern of fluid intake is controlled. However, in real life athletes often drink ad libitum. Thus, more studies should include an ad libitum fluid intake trial to compare against the effects of no fluid and prescribed intake to better understand the scenarios in which ad libitum may be sufficient versus when prescribed intake is warranted to maintain performance.
For all of the aforementioned research questions, it is particularly important that future studies focus on the sports that are associated with a moderate or high risk of developing significant hypohydration. Some examples of these sports include soccer, rugby, American Football, Australian Rules Football, field hockey, ice hockey, and tennis. By contrast, sports in which sweating rates are expected to be low and/or fluid replacement opportunities are adequate (e.g., baseball) probably warrant less investigation. Still, there may be certain players in low risk sports that have increased risk of developing hypohydration due to equipment requirements and/or the physical demands of the position (e.g., baseball catcher).

Conclusion
Significant hypohydration ([2%) has been reported most consistently in soccer. Although other sports (e.g., American Football, rugby, basketball, tennis, and ice hockey) have reported high sweating rates, fluid balance disturbances have generally been mild, suggesting that drinking opportunities were sufficient to provide most athletes with enough fluid to offset losses. The effect of hydration status on team sport performance has been mixed. However, it seems that hypohydration is more likely to impair cognition, technical skill, and physical performance at higher levels of BML (3-4% difference between trials), which are not routinely observed in team sport athletes. Detriments to performance are also more likely when the method of dehydration involves heat stress. Although exact mechanisms are unclear, increased subjective ratings of fatigue and perceived exertion consistently accompany hypohydration in team sport studies and could explain, in part, the performance impairments reported in some studies. More research is needed to develop ecologically valid, reliable, and sensitive sport-specific protocols and should be used in future studies to determine the effects of hypohydration and modifying factors (e.g., age, sex, athlete caliber) on team sport performance.

Compliance with Ethical Standards
Funding The preparation of this review was funded by the Gatorade Sports Science Institute, a division of PepsiCo, Inc. Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http:// creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.