It is well documented that dehydration can reduce exercise capacity (16) and increase thermoregulatory strain, especially in the heat (24). A reduction in aerobic performance is linked to dehydration-induced hypovolemia (21) and is exacerbated in hot environments (9,15). A recent study showed that progressive dehydration induced a considerable cardiovascular, thermoregulatory, and metabolic strain (greater carbohydrate oxidation) even during mild degrees of hypohydration of ~1% body mass loss (25).
Drinking ad libitum during exercise often leads to dehydration ranging from −1% to −3% of body weight regardless of environmental conditions (23,31). This dehydration might be exacerbated because many athletes begin practice or competition in a suboptimal hydration state (1,5,26,30). At the same time, some researchers support the idea that athletes who finish prolonged exercise with large weight losses do not have greater risk for heat illnesses or decrease their performance when drinking ad libitum (29). A study during a marathon race in France found that runners that were more dehydrated finished faster than those who were less dehydrated or euhydrated, emphasizing that the most successful athletes dehydrate more than 3%–4% of their body weight during competition (37). Similarly, Berkulo et al. (11) found that mild hypohydration of ~1% had no detrimental effect on 40-km time trial performance, suggesting that drinking ad libitum had no negative nor positive effects on exercise performance. Lastly, two studies found that hypohydration of even 3% body mass did not impair time trial performance in the heat, even when the subjects were blinded to their hydration state via intravenous infusion (35) and matched for thirst via mouth rinse (14). Although the participants in the previous two studies did not drink to the dictation of thirst, both studies suggested that drinking to thirst is satisfactory and that current hydration guidelines on fluid replacement (34) are inaccurate.
Currently, no study has examined the effectiveness of prescribed versus ad libitum drinking (AD) on cycling performance in the heat. Therefore, the aim of this study was to examine the effect of the prescribed fluid ingestion to match sweat losses compared with AD, on cycling performance and thermoregulatory responses, during a simulated criterium-like cycling test. We hypothesized that AD would lead to dehydration, which would decrease cycling performance possibly via greater thermoregulatory and/or cardiovascular strain.
METHODS
Participants
Ten heat-acclimatized male cyclists (30 ± 5 yr, 76.5 ± 7.2 kg, 1.81 ± 0.07 m, V˙O2peak = 61.3 ± 5.2 mL·min−1·kg−1, body fat = 10.5% ± 3.3%, Powermax = 392 ± 33 W) were recruited to participate in the study during the summer months (June–August) in Greece. All athletes had been racing in national cycling races for 10 ± 3 yr. Participants gave their written consent before the study that was approved by the university's institutional review board. Eligibility criteria for participation included competitive cycling status and absence of any metabolic, cardiovascular, renal disease, and history of heat stroke.
Preliminary Testing
Body composition
Anthropometric characteristics were recorded during the first visit at the laboratory. Body weight (Seca, model 770, Hamburg Germany) and standing height (Seca, model 700, Hamburg, Germany) were measured to the nearest 0.1 kg and 0.005 m, respectively. Body composition was determined via dual-energy x-ray absorptiometry (model DPX; Lunar, Madison, WI).
V˙O2peak test
V˙O2peak was determined in a thermocomfortable environment (21°C) using an incremental resistance exercise test on a computerized mechanically braked cycle ergometer (Monark 839E, Sweden). After standardized warm-up, participants started cycling at 100 W and cycling power increased by 20 W every minute until volitional exhaustion. During the test, expiratory gases were analyzed breath by breath via an online gas analyzer (MedGraphics Ultima Series; Medical Graphics Corporation, Minnesota, MN). At least three of the four following criteria were used to verify the attainment of V˙O2peak: 1) V˙O2plateau with increased workload, 2) respiratory exchange ratio greater than 1.1, 3) HR greater than 90% of age-predicted maximal value (220 − age), and 4) perceived exertion based on the 6–20 Borg scale greater than 17. The highest workload attained was defined as
;)
where Wout is the highest workload completed in 1 min, t is the number of seconds that remained in the final uncompleted workload, and ΔW is the increase of workload.
Experimental Protocol
All participants performed two experimental criterium-like cycling tests in a counterbalanced manner in the laboratory. The performance test consisted of three sets of a 5-km cycling at 50% of maximum power output followed by a 5-km cycling all out at 3% grade (total 30 km). Before the first experimental trial, participants completed a familiarization session to estimate their individual sweating rate. The familiarization was identical with the two experimental protocols in ambient conditions and exercise intensity. The two experimental trials were identical except the drinking. During the AD trial, cyclists were provided drinking water as much as they desired. During the prescribed drinking (PD) trial, cyclists were ingesting water every 1 km in a rate to match 100% of fluid lost via sweating, based on their familiarization trial. For all trials, participants consumed Volvic bottled water (Danone, France) from a triathlete water bottle with a straw, mounted on the bike handlebar, to facilitate hands-free drinking and simulate racing conditions. Each trial was separated by at least 1 wk, but no more than 2 wk. Participants were asked to abstain from alcohol, caffeine, and training for 24 h before each trial, to maintain the same training regiment, and to abstain from racing between the two experimental trials. The exercise trials were performed on an individual basis in the morning, at the same time of the day on both visits, to avoid diurnal variations (6). During the previous day of each trial, an individualized diet was provided by a dietitian based on their size and activity level, which consisted of 75% carbohydrate, 15% proteins, and 10% fat. To minimize differences in starting muscle glycogen concentrations in between trials, participants recorded their diet 24 h before their first visit. Diet records were copied and returned to the participants with instructions to follow the same diet before the next subsequent visit. On the day of the test, all volunteers consumed a standardized breakfast consisted of two slices of toasted bread with two tablespoons of honey and a glass of water. Urine samples were collected at the end of the 30-km cycling or in between the 5-km bouts in case the subjects had the urge to void. All samples were collected in dark 2-L containers so that participants would not be able to get feedback on their urine volume and color. Also, during body weighing, participants were not able to see the reading.
Familiarization session
Before the two experimental trials, participants completed the 30-km cycling test to get familiarized with the experimental protocol. They were instructed to drink as much as they wanted from the water provided. During this trial, sweating rate was estimated based on the changes of body weight corrected for water intake and urine output. The volume of water ingested during the prescribed trial (PD) was estimated based on the familiarization trial to replace 100% of fluid losses. The protocol of this session was identical with the two experimental trials, except the lack of blood draws and body temperatures measurements.
Experimental Trials
Upon arrival to the laboratory, a urine sample was collected to assess pretrial hydration state and proceeded to testing only when urine specific gravity (USG) was less than 1.020 (34). A blood sample was taken without stasis, after a 30-min rest in the heat, in a seated position. Participants cycled indoors on their own bike mounted on a cycle ergometer (ComputrainerTM Lab Model trainer; Racer-Mate, Seattle, WA). A large fan was placed directly in front of the subjects to provide airflow of 6.4 m·s−1. In both trials, all cyclists used the same bike, cycling shoes with cleats, and cycling clothes. Bike tires were pumped to 110 psi, and bikes were checked to ensure proper functioning. No cycling computer was allowed to be viewed by the participant throughout the experiment. Ambient temperature during the AD and the PD trials were similar (31.4°C ± 0.5°C and 31.7°C ± 0.4°C, P > 0.05).
Thermoregulatory measurements
To record gastrointestinal temperature (TGI), participants ingested a thermosensitive pill (HQ, Inc., Palmetto, FL), 8–10 h before the test. Skin temperatures were measured using skin thermistors (YSI, 4000 A, Dayton, OH) in forearm (TForearm), chest (Tchest), thigh (Tthigh), and calf (Tcalf). Mean weighted skin temperature (Tsk) was calculated using the Ramanathan equation (32): Tsk = 0.3 (Tchest + TForearm) + 0.2 (Tthigh + Tcalf). TGI and Tsk were recorded before and immediately after each 5-km performance bouts. Mean body temperature (Tb) was calculated according to the following equation (17): Tb = 0.79 (TGI) + 0.21 (Tsk). Heat storage (Hs) was calculated according to the following equation (2): Hs = 0.965BWΔTbAD−1, where 0.965 is the specific heat capacity of the body, BW is the mean body mass over duration of trial (kg), ΔTb is the change in body temperature, and AD is the body surface area in square meter. AD was estimated based on the Du Bois and Du Bois equation (19) AD = 0.202 BW0.425 × height0.725. Sweat sensitivity was calculated by dividing the amount of sweat in grams per heat storage in watts per square meter (3,8).
Cardiovascular variables
HR was recorded during exercise via a wireless HR monitor (Suunto T6c, Oy, Finland). Blood pressure was measured twice (Standby; Baumanometer, Copiague, NY) at the end of each climbing bout. Mean arterial pressure (MAP) was calculated as MAP = DP + (SP − DP)/3, where DP and SP were diastolic and systolic blood pressure, respectively (22).
Performance and perceptual variables
Cycling cadence, distance, time, and power output were recorded online by the Computrainer software (RacerMate Inc., Seattle, WA), whereas HR data were recorded via an HR monitor mounted underneath the bike saddle. Cyclists could view on the screen the course profile, but they could not get any feedback on their performance or HR. Moreover, participants were asked to rate their effort using Borg's perceived exertion (RPE) (12), their perceived thirst (“how thirsty are you now”), and their stomach fullness (“how full is your stomach now”) after each 5-km bout with visual analog scales (33). The visual analog scale used was consisted of a 180-mm line with an anchor on the left side (0 mm, “not at all”) and a second anchor on the 125-mm mark with the label “extremely.” Values for the visual analog scale are expressed as percent of “extremely thirsty” at 125 mm.
Blood and urine analyses
USG and total plasma proteins were measured using a handheld refractometer (Atago SUR-NE, Tokyo, Japan). Hematocrit (Hct) was determined in triplicate from whole blood using the microcapillary technique, after centrifugation for 5 min at 10,000 rpm. Hemoglobin (Hb) was also measured in triplicate from whole blood via the cyanmethemoglobin technique, using a commercially available kit (Drabkin's reagent; Sigma, Saint Louis, MO). Percent change in plasma volume was calculated using the Dill and Costill equation (18). Plasma and urine osmolality were measured in duplicate via freezing-point depression (3D3 Osmometer; Advanced Instruments Inc., Norwood, MA). Serum potassium and sodium concentrations were determined in duplicate (IL Ilyte 20 Electrolyte Analyzer; Instrumentation Laboratory, Milano, Italy). Whole blood lactate was measured at the end of each 5-km climb with a lactate analyzer (Accutrend Lactate; Roche Diagnostics, Mannheim, Germany).
Statistical Analysis
Normality of data was graphically explored using percentile plots. All variables are presented as mean ± SD because they were normally distributed. Normality was tested using P-P plots. Differences in the mean values or the distributions of parameters between PD and AD (BW, ΔBW, urine osmolality, USG, urine color, plasma osmolality, plasma sodium, plasma potassium, total plasma protein, and change in plasma volume) were assessed using Student's paired t-tests. Generalized estimating equations were fitted to evaluate differences between the two experimental trials (PD and AD) as well as across time points. For the dependent variables (i.e., cycling time, power output, cycling cadence, HR, mean blood pressure, lactate acid, thirst, gastrointestinal comfort questionnaire, RPE, TGI, Tb, Tsk, sweat sensitivity, and sweat rate), the normal distribution was used for fitting generalized estimating equations, with the identity as the link function. The unstructured formation of the correlation matrix was used after comparing various scenarios, using the corresponding quasi-likelihoods under the independence criterion for model's goodness of fit. Independent variables were time (before the first climb, before the second Climb, and before the third Climb) (first climb, second climb, and third Climb) and trial (PD and AD) or kilometer (1–5) and trial (PD and AD). The first-order interactions between time and trial or between kilometers and trial were also applied. Post hoc analysis for comparing mean values between trials across time points, as well as different time points, was applied by using the sequential Bonferroni correction rule to adjust for the inflation of type I error due to multiple comparisons. Using data from a similar study (8), an α of 0.05 and a statistical power of 0.8, it was estimated that 10 subjects would be required to reject the null hypothesis. All statistical analyses were conducted with SPSS 23 for Windows (IBM SPSS, Chicago, IL). A value of P < 0.05 was regarded as statistically significant.
RESULTS
Body weight parameters
Cyclist finished the first, the second, and the third hill climbing tests in the AD trial with a water deficit of −0.2 ± 0.2, −0.5 ± 0.4, and −1.4 ± 0.5 kg, respectively (Table 1, P < 0.001). However, during the PD trial, the riders maintained euhydration (0.0 ± 0.0, −0.1 ± 0.2, and −0.4 ± 0.2 kg). The mean values of water intake in the two trials were 2.1 ± 0.4 L and 0.7 ± 0.4 L for the PD and AD, respectively. The percent of the dehydration at the end of the third 5-km hill climbing cycling test for the AD and PD trials was −1.8% ± 0.7% and −0.5% ± 0.3% of body weight, respectively.
TABLE 1:
Blood and urine parameters during the experiment (N = 10).
Cycling performance
Cycling speeds during the first two 5-km hill cycling were not different between the AD trial (30.3 ± 2.3 and 29.2 ± 2.7 km·h−1) and the PD trial (29.8 ± 2.1 and 29.2 ± 2.4 km·h−1, P > 0.05). However, during the third bout of the protocol, the cyclists completed the 5-km climb by 31 ± 30 s (4.7% ± 4.4%) faster in PD compared with AD by cycling at a higher speed (PD = 30.2 ± 2.4 km·h−1, AD = 28.8 ± 2.6 km·h−1, P < 0.05; Fig. 1). Nine of the 10 subjects performed better in the third 5-km bout during the PD trial compared with the AD trial (Fig. 2). Cycling power output was significantly greater during the last 3 km of the third cycling bout in the PD trial versus the AD trial (Fig. 3). Cycling cadence at the first, second, and third 5-km hill race was not different between the AD trial (89 ± 10, 88 ± 10, and 86 ± 10 rpm) and the PD trial (91 ± 9, 91 ± 7, and 90 ± 7 rpm; P > 0.05).
FIGURE 1:
Mean cycling speed of each hill session (mean ± SD). *Statistically significant differences, P ≤ 0.05, between trials at same time point.
FIGURE 2:
Individual performance data during the third 5-km time trial performance test in PD and AD trials. Each point represents a different individual participant.
FIGURE 3:
Mean power output of each kilometer during hill bouts (mean ± SD). *Statistically significant differences, P ≤ 0.05, between trials at same time point.
Thermoregulatory markers
TGI, Tsk, and Tb at the end of the last hill climbing session were greater in the AD trial compared with the PD trial (P < 0.05, Fig. 4). Total whole body sweat rate during the three bouts of hill climbing in the AD trial was not different (0.24 ± 0.07 mg·m−2·s−1 or 2.0 ± 0.4 L·h−1) when compared with the PD trial (0.33 ± 0.05 mg·m−2·s−1 or 2.3 ± 0.3 L·h−1, P > 0.05). Heat storage at the same time was great for the AD trial (69.8 ± 22.3 W·m−2) versus the PD trial (50.8 ± 7.8 W·m−2, P < 0.05). Overall, sweat sensitivity (sweat loss in grams over heat stored in watts per square meter) during the three climbing bouts was lower in the AD (15.6 ± 5.7 g·W−1·m−2) than in the PD trial (22.8 ± 3.4 g·W−1·m−2, P < 0.05).
FIGURE 4:
Gastrointestinal (T GI), mean body temperature (T b), and skin temperature (T sk) at baseline, before, and immediately after circuit course. *Statistically significant differences, P ≤ 0.05, between trials at same time point.
Blood pressure, HR, and lactic acid
MAP at the end of the first, second, and third climbing bouts during the AD trial (129 ± 7, 132 ± 8, and 128 ± 12 mm Hg) did not differ from the PD trial (129 ± 13, 130 ± 12, and 131 ± 11 mm Hg, P > 0.05). HR during the three race-pace hill bouts was near maximal (90% ± 2%) and did not differ significantly between the two drinking interventions (AD = 168 ± 8, 169 ± 8, and 170 ± 9 bpm; PD = 171 ± 9, 171 ± 9, and 173 ± 9 bpm; P > 0.05). Also, blood lactic acid concentration at the end of each performance climbing bout was not different between the AD trial (10.0 ± 1.8, 11.1 ± 1.3, and 10.7 ± 1.8 mmol·L−1) and the PD trial (10.4 ± 2.3, 11.1 ± 1.2, and 11.2 ± 1.7 mmol·L−1, P > 0.05).
Thirst, gastrointestinal comfort, and RPE
At the end of the first two performance climbing bouts, thirst ratings in the AD trial (24% ± 18% and 30% ± 14%) and in the PD trial (21% ± 18% and 20% ± 19%) were not different (P < 0.05). However, at the end of the third performance bout, thirst was higher in the AD trial (49% ± 25%) compared with the PD trial (20% ± 17%, P < 0.05), indicating subjects being moderately thirsty (<50% of extremely thirsty). Stomach fullness ratings at the end of the first (5–10 km), second (15–20 km), and third (25–30 km) race-pace hill bouts for the AD trial (17% ± 26%, 19% ± 22%, and 19% ± 26%) were not different compared with the PD trial (51% ± 38%, 50% ± 37%, and 23% ± 18%; P > 0.05). RPE at the end of each performance climbing was not different between the experimental trials (AD = 17.0 ± 1.0, 18.0 ± 1.0, 19.0 ± 1.0; PD = 17.0 ± 1.0, 17.0 ± 1.0, 18.0 ± 1.0; P > 0.05).
DISCUSSION
The current investigation examined the effect of prescribed hydration protocol compared with ad libitum intake during a simulated criterium-like cycling event in a hot environment. The results indicated that cyclists during the PD trial performed better during the last 5-km hill climb but not during the first and the second climbing bouts. This could be due to the loss of body fluids accumulated more in the last uphill level in the ad libitum trial. Cyclists at the end of the third climbing bout finished with body greater mass loss (−1.8% of body weight) than during the prescribed fluid intake (−0.5% of body weight). These data indicated that AD led to mild dehydration (<−2% of body weight) and lower performance, probably because of the increased thermoregulatory strain and reduced sweating capacity. Similarly, Walsh et al. (36) examined cyclists' performance in 32°C. They found that the time to exhaustion at 90% of V˙O2max was decreased by 31% when participants started cycling in a mild hypohydrated state of −1.8%. Also, Below et al. (10) showed that when cyclists were hypohydrated by less than −2%, their exercise performance declined by 6.5%. Further, Casa et al. (13) observed that during trail running in the heat, core temperature was elevated by 0.22°C more during the hypohydrated than the euhydrated trial for every 1% of body mass lost.
It is known that exercise increases body core temperature (27), especially in a hot environment. However, hypohydration can play a modulating role on the extent of exercise-induced hyperthermia (28). We found that by the end of the exercise, core temperature, mean body temperature, and skin temperature were greater in the AD trial compared with the PD trial, indicating that even a small degree of hypohydration induced greater thermoregulatory strain. The fact that sweating responses were similar between the trials although the thermoregulatory strain was greater in the AD trial indicated that sweat sensitivity deteriorates because of the water deficit. Because power output was greater in the PD trial, it was expected that euhydrated athletes would have higher core temperature because of greater exercise intensity, as supported by the literature (16). Sweat sensitivity has been used as an index of thermoregulatory efficiency because it represents sweating as a response to a rise in core temperature. Our data indicated that a mild hypohydration induced significantly lower sweat sensitivity during climbing bouts in the AD trial compared with the PD trial. The data also agree with Armstrong et al. (3) who examined the thermoregulatory responses during a fixed 90-min treadmill walking in the heat when participants were euhydrated or hypohydrated, with or without ad libitum water intake during exercise. Both sweating and sweat sensitivity were reduced in the hypohydrated (−3.9% body weight) trial when participants did not drink fluids during exercise, showing that hypohydration modulates thermoregulatory responses. Their data also indicated that lower sweat sensitivity was associated with the degree of hypohydration and changes in plasma osmolality.
Recent studies by Bardis et al. (7,8) investigated the effects of mild hypohydration (−1% BW) during an outdoor cycling climbing trial and during simulated repeated hill climbing bouts in the laboratory. These data suggested that mild hypohydration decreased cycling performance during both the 5-km outdoor hill climbing course and the 30-km simulated criterium racing in the laboratory, probably because of greater heat strain and lower sweat sensitivity. Similarly, in the present study, greater body temperature at the end of the last climbing bout and lower sweat sensitivity were evident with the mild hypohydration (−1.8%).
Another potential factor that could explain the impairment of exercise performance during the AD trial in our study was the elevation of Tsk at the end of the third 5-km performance test. Interestingly, Tsk was significantly higher in the AD trial despite cycling at a slower speed, thus producing lower metabolic heat. Similarly, Kenefick et al. (24) examined the interaction between four different environmental and hypohydration conditions during a 30-min (50% V˙O2max) cycling exercise, followed by a 15-min time trial. They found that hypohydration and hyperthermia significantly contribute to impaired performance, whereas Tsk modulated the exercise performance decline.
HR responses during the race-pace climbing bouts of the protocol did not differ between trials, although the exercise induced near-maximal HR in both trials. Although we did not measure cardiac output during exercise, it can be speculated that both stroke volume and cardiac output might have been compromised as a response to hypohydration. This response has been well documented with a greater degree of hypohydration by others (21). Recently, Logan-Sprenger et al. (25) studied nine females during a 120-min cycling exercise without or with fluid replacement to match losses. They concluded that the progressive dehydration induced a considerable cardiovascular, thermoregulatory, and metabolic strain (toward greater carbohydrate oxidation) evident from the hypohydration of −1% of body weight. More specifically, the observed increase in muscle glycogenolysis appeared to be primarily the result of a rise in core and muscle temperature.
Previous studies have suggested that the decline in exercise performance observed in dehydration studies might be associated with extreme thirst (14,36). Berkulo et al. (11) found that thirst ratings were significantly higher when hypohydrated compared with euhydrated; however, no differences were found in average power output during a 40-km time trial. To our surprise, drinking ad libitum not only did not prevent dehydration but also did not prevent thirst stimulation although subjects had access to water via a triathlete bottle with straw, mounted on the bike handlebar for convenient, hands-free access. More specifically at the end of exercise protocol, thirst rating in the PD trial was significantly lower than the AD one. It should be noted that the subjects were only moderately thirsty because the average score was less than 50% of the “extremely thirsty” value.
Recently, two separate studies have attempted to remove the potential effect of thirst on exercise performance and blind the subjects from their hydration state (14,35). Both studies found that 3% of body mass hypohydration had no effect on cycling time trial performance. However, in the study by Wall et al. (35), cyclists exhibited higher core temperatures during the last 13 of the 25-km time trial when hypohydrated compared with the euhydrated trial, despite having similar thirst ratings in all trials. However, in the previous study, plasma volume was restored after rehydration to −3% using isotonic saline, which is an unnatural condition. The dehydration-induced plasma volume contraction is one of the major mechanisms that explains how hypohydration impairs exercise performance.
Similarly, in the study by Cheung et al. (14), hypohydrated cyclists had higher HR and core temperatures during the last 12 of their 20-km time trial, whereas thirst was matched between trials. Both of these studies used intravenous rehydration methods to blind their participants to their hydration state (infusion vs sham infusion). Although Cheung et al. (14) provided mouth rinse to control thirst, both studies did not allow for oral drinking, which has been shown to reduce thirst, to increase performance, and to inhibit vasopressin release, probably via pharyngeal receptor stimulation (4,20). In the present study, drinking in each kilometer of the PD trial might have contributed to better performance via pharyngeal stimulation and lower thirst perception. It has been suggested that the activation of the pharyngeal receptors by swallowing a small amount of water could enhance performance in dehydrated individuals compared with mouth rinse (4). Moreover, although during the PD trial subjects were consuming more water to match sweat rate, no differences were found between trials in the stomach fullness ratings. These data suggest that cyclists might be able to handle large amount of water intake during high-intensity exercise in the heat without any gastrointestinal discomfort.
A possible limitation to this study was that the act of drinking alone could have provided a motivational, ergogenic effect independent of oropharyngeal stimulation. This limitation could have risen from fact that the PD received more fluid and were required to drink more. Another potential limitation could be related to the nature of laboratory testing. Although the subjects were cycling on their own bikes, they were mounted on a stationary cycle ergometer where drinking was easy and moving out of optimal aerodynamic position would not impair performance. Carrying additional weight of water bottles for PD could result in a mechanical disadvantage especially during hill climbing. Lastly, a limitation of the study was the fact that the water temperature provided to the cyclists was lower than body temperature. Although the temperature of the drinks was not recorded, it can be assumed that the drinks were very close to room temperature (31°C) because they were stored in the testing area before the experiment. During the PD trial, subjects were drinking on average 1.4 L more water than that in the AD trial (2.1 ± 0.4 vs 0.7 ± 0.4 L), and the temperature difference between body (37°C–38°C) and water (31°C) may have had a cooling effect.
In conclusion, the data showed that PD to match fluid losses during cycling exercise in the heat provided a performance advantage by maintaining better hydration state. This benefit seems to be associated with lower thermoregulatory strain because of lower skin and core temperatures. By contrast, mild dehydration associated with AD induced performance impairments. Further research is needed to evaluate the effect of PD compared with ad libitum in different sports of varying intensities and environments.
The authors would like to thank George Stais, Konstantinos Danias, Eleni Kosti, Eleni Samara, and Giannis Arnaoutis for their help during data collection. SAK is a scientific consultant for Quest Diagnostics and has active grants from Danone Research, France. The results of the present study do not constitute endorsement by the American College of Sports Medicine.
REFERENCES
1. Adams JD, Kavouras SA, Robillard JI, et al. Fluid balance of adolescent swimmers during training.
J Strength Cond Res. 2016;30(3):621–5.
2. Adams WC, Mack GW, Langhans GW, et al. Effects of varied air velocity on sweating and evaporative rates during exercise.
J Appl Physiol (1985). 1992;73(6):2668–74.
3. Armstrong LE, Maresh CM, Gabaree CV, et al. Thermal and circulatory responses during exercise: effects of hypohydration, dehydration, and water intake.
J Appl Physiol (1985). 1997;82(6):2028–35.
4. Arnaoutis G, Kavouras SA, Christaki I, et al. Water ingestion improves performance compared with mouth rinse in dehydrated subjects.
Med Sci Sports Exerc. 2012;44(1):175–9.
5. Arnaoutis G, Kavouras SA, Kotsis YP, et al. Ad libitum fluid intake does not prevent dehydration in suboptimally hydrated young soccer players during a training session of a summer camp.
Int J Sport Nutr Exerc Metab. 2013;23(3):245–51.
6. Atkinson G, Todd C, Reilly T, et al. Diurnal variation in cycling performance: influence of warm-up.
J Sports Sci. 2005;23(3):321–9.
7. Bardis CN, Kavouras SA, Arnaoutis G, et al. Mild dehydration and cycling performance during 5-kilometer hill climbing.
J Athl Train. 2013;48(6):741–7.
8. Bardis CN, Kavouras SA, Kosti L, et al. Mild hypohydration decreases cycling performance in the heat.
Med Sci Sports Exerc. 2013;45(9):1782–9.
9. Barr SI. Effects of dehydration on exercise performance.
Can J Appl Physiol. 1999;24(2):164–72.
10. Below PR, Mora-Rodríguez R, González-Alonso J, et al. Fluid and carbohydrate ingestion independently improve performance during 1 h of intense exercise.
Med Sci Sports Exerc. 1995;27(2):200–10.
11. Berkulo MA, Bol S, Levels K, et al. Ad-libitum drinking and performance during a 40-km cycling time trial in the heat.
Eur J Sport Sci. 2016;16(2):213–20.
12. Borg GA. Psychophysical bases of perceived exertion.
Med Sci Sports Exerc. 1982;14(5):377–81.
13. Casa DJ, Stearns RL, Lopez RM, et al. Influence of hydration on physiological function and performance during trail running in the heat.
J Athl Train. 2010;45(2):147–56.
14. Cheung SS, McGarr GW, Mallette MM, et al. Separate and combined effects of dehydration and thirst sensation on exercise performance in the heat.
Scand J Med Sci Sports. 2015;25(1 Suppl):104–11.
15. Cheuvront SN, Carter R 3rd, Sawka MN. Fluid balance and endurance exercise performance.
Curr Sports Med Rep. 2003;2(4):202–8.
16. Cheuvront SN, Kenefick RW. Dehydration: physiology, assessment, and performance effects.
Compr Physiol. 2014;4(1):257–85.
17. Colin J, Timbal J, Houdas Y, et al. Computation of mean body temperature from rectal and skin temperatures.
J Appl Physiol. 1971;31(3):484–9.
18. Dill DB, Costill DL. Calculation of percentage changes in volumes of blood, plasma, and red cells in dehydration.
J Appl Physiol. 1974;37(2):247–8.
19. Du Bois D, Du Bois E. A formula to estimate the approximate surface area if height and weight be known.
Arch Intern Med. 1916;17:863–71.
20. Figaro MK, Mack GW. Regulation of fluid intake in dehydrated humans: role of oropharyngeal stimulation.
Am J Physiol. 1997;272(6 Pt 2):R1740–6.
21. González-Alonso J, Mora-Rodríguez R, Below PR, et al. Dehydration markedly impairs cardiovascular function in hyperthermic endurance athletes during exercise.
J Appl Physiol (1985). 1997;82(4):1229–36.
22. Gonzalez-Alonso J, Mora-Rodriguez R, Coyle EF. Supine exercise restores arterial blood pressure and skin blood flow despite dehydration and hyperthermia.
Am J Physiol. 1999;277(2 Pt 2):H576–83.
23. Greenleaf JE. Problem: thirst, drinking behavior, and involuntary dehydration.
Med Sci Sports Exerc. 1992;24(6):645–56.
24. Kenefick RW, Cheuvront SN, Palombo LJ, et al. Skin temperature modifies the impact of hypohydration on aerobic performance.
J Appl Physiol (1985). 2010;109(1):79–86.
25. Logan-Sprenger HM, Heigenhauser GJ, Killian KJ, et al. Effects of dehydration during cycling on skeletal muscle metabolism in females.
Med Sci Sports Exerc. 2012;44(10):1949–57.
26. Magal M, Cain RJ, Long JC. Pre-practice hydration status and the effects of hydration regimen on collegiate division III male athletes.
J Sports Sci Med. 2015;14(1):23–8.
27. Maughan RJ, Leiper JB, Thompson J. Rectal temperature after marathon running.
Br J Sports Med. 1985;19(4):192–5.
28. Montain SJ, Coyle EF. Influence of graded dehydration on hyperthermia and cardiovascular drift during exercise.
J Appl Physiol (1985). 1992;73(4):1340–50.
29. Noakes TD. The central governor model of exercise regulation applied to the marathon.
Sports Med. 2007;37(4–5):374–7.
30. Osterberg KL, Horswill CA, Baker LB. Pregame urine specific gravity and fluid intake by National Basketball Association players during competition.
J Athl Train. 2009;44(1):53–7.
31. Passe D, Horn M, Stofan J, et al. Voluntary dehydration in runners despite favorable conditions for fluid intake.
Int J Sport Nutr Exerc Metab. 2007;17(3):284–95.
32. Ramanathan NL. A new weighting system for mean surface temperature of the human body.
J Appl Physiol. 1964;19:531–3.
33. Rolls BJ, Wood RJ, Rolls ET, et al. Thirst following water deprivation in humans.
Am J Physiol. 1980;239(5):R476–82.
34. Sawka MN, Burke LM, Eichner ER, et al. American College of Sports Medicine Position Stand: exercise and fluid replacement.
Med Sci Sports Exerc. 2007;39(2):377–90.
35. Wall BA, Watson G, Peiffer JJ, et al. Current hydration guidelines are erroneous: dehydration does not impair exercise performance in the heat.
Br J Sports Med. 2015;49(16):1077–83.
36. Walsh RM, Noakes TD, Hawley JA, et al. Impaired high-intensity cycling performance time at low levels of dehydration.
Int J Sports Med. 1994;15(7):392–8.
37. Zouhal H, Groussard C, Minter G, et al. Inverse relationship between percentage body weight change and finishing time in 643 forty-two-kilometre marathon runners.
Br J Sports Med. 2011;45(14):1101–5.
