Influence of Hydration Status on Thermoregulation and Cycling Hill Climbing : Medicine & Science in Sports & Exercise

Journal Logo

BASIC SCIENCES: Original Investigations

Influence of Hydration Status on Thermoregulation and Cycling Hill Climbing

Ebert, Tammie R.1,4; Martin, David T.1; Bullock, Nicola1; Mujika, Iñigo2; Quod, Marc J.1; Farthing, Lesley A.3; Burke, Louise M.3; Withers, Robert T.4

Author Information
Medicine & Science in Sports & Exercise: February 2007 - Volume 39 - Issue 2 - p 323-329
doi: 10.1249/
  • Free


Various laboratory-based protocols have shown that dehydration can reduce exercise capacity or performance, with impairments generally becoming detectable when the fluid deficit exceeds 2% of body mass (6). To address this concern, guidelines on exercise and fluid replacement have been formulated by various expert groups, including the American College of Sports Medicine (5). These guidelines recommend that athletes consume sufficient fluid during exercise to replace sweat losses, with the further acknowledgement that fluid intake and carbohydrate needs during prolonged exercise can be met simultaneously by consuming sports drinks in volumes of 600-1200 mL·h−1 (5). However, in real life, many athletes who compete in continuous endurance events do not seem to meet these goals. For example, a review of the distance running literature by Cheuvront and Haymes (4) indicates that these athletes consume approximately 400 mL·h−1, which is well below their rates of sweat loss and nutritional guidelines. In addition, professional male cyclists have been reported to incur fluid losses of 2.1-4.5 kg during a road race (1).

A possible explanation for these observed practices by successful athletes is that dehydration does not always cause a serious impairment of exercise capacity. Indeed, one laboratory study reported no decrement in a 1-h time trial with restricted fluid intake compared with normal fluid intake (27). Similarly, McConell et al. (18) observed minimal impact on the work performed during an all-out 15-min cycling effort when fluid consumption was compromised. Furthermore, there are some practical or biomechanical considerations that might favor a lower fluid intake or tolerance for a certain level of dehydration in the field. Possible benefits include reducing the interruption to race pace that might occur when the athlete obtains fluid from an aid station or changes their posture to facilitate fluid intake. Another potential benefit is that a reduction in body mass attributable to fluid loss has a positive effect on lowering the energy cost of movement and/or increasing the power-to-mass ratio (6). Researchers have previously highlighted the importance of body mass during hill-climbing cycling performance by determining appropriate mass exponents, which represent hill-climbing performance (13,22). Coyle (6) termed the theoretical benefit of reduced energy expenditure arising from a reduced body mass "functional dehydration" but noted that data are not available to balance its effect against the risk of hyperthermia.

Cycle races occur over semimountainous and high mountainous terrain, and success in many 1-d and multiday stage events has been shown to correlate with climbing ability (16). Because hill-climbing performance is highly dependent on the cyclist's power-to-mass ratio (16), this scenario provides a real-life juxtaposition of the potential benefits of a reduction in body mass from fluid loss with the possible compromise of thermoregulatory capabilities attributable to dehydration. Accordingly, this study investigated whether a fluid deficit typical of the mismatch between fluid intake and sweat loss observed among professional male cyclists during stage races (1) is associated with a performance advantage attributable to the related decrease in body mass. We hypothesized that a decrease in body mass attributable to conservative fluid intake would reduce the power output required during hill climbing at a fixed speed and grade, thereby extending the time to exhaustion of a simulated cycling hill-climbing effort.



Eight well-trained male cyclists (mean ± SD; age: 28.4 ± 5.7 yr; body mass: 71.0 ± 5.9 kg; height: 176.7 ± 4.7 cm; body fat: 8.8 ± 3.2%; V˙O2peak: 66.2 ± 5.8 mL·kg−1·min−1; maximal aerobic power output (MAP): 355 ± 26 W) participated in this study after providing written informed consent. The study was approved by the Australian Institute of Sport ethics committee in accordance with the Declaration of Helsinki.


Subjects participated in a maximal graded exercise test to determine V˙O2peak and MAP, the latter of which was used to determine workloads in the experimental trials. This test was performed on a stationary cycle ergometer (Lode Excalibur Sport, Groningen, The Netherlands) and started at 100 W with 5-min increments of 50 W until volitional exhaustion. MAP was calculated as follows (15):

in which WL is the power output of the last complete workload (W), t is the time (min) for the final incomplete workload, and 5 and 50 are the increments for each workload for time (min) and power output (W), respectively. Subjects' expired air was collected into a customized Douglas bag gas analysis system, which incorporated an automated piston that allowed the volume of air displaced to be quantified, with O2 and CO2 analyzers (AEI Technologies, Pittsburgh, PA). The operation and calibration of this equipment have been described previously (28). Heart rate (HR: Polar S710® series, Polar, Finland), rating of perceived exertion (RPE; 3), and blood lactate (BLa: Radiometer ABL 700 Series, Copenhagen, Denmark) were recorded during the final 30 s of each workload.

Experimental trials.

After two to three sessions of familiarization with treadmill cycling, subjects undertook one habituation trial and, in the following week, two randomized experimental treatments that were separated by 2 d of recovery. Subjects were asked to refrain from vigorous exercise and caffeine intake for 12 and 3 h before each treatment, respectively. They were provided with food parcels before each trial to ensure that they followed a similar diet for the 24 h before both experimental treatments (energy: 15113 ± 1054 kJ; carbohydrate: 574 ± 46 g; fat; 77 ± 4 g; protein: 131 ± 9 g). To ensure that all subjects were adequately hydrated, they were asked to consume 5 mL·kg−1 body mass of water during the 2 h before testing.

Subjects presented to the laboratory at the same time of day for each testing session to control for circadian rhythm variations in core temperature. A midstream urine sample was collected before testing to determine urine specific gravity (UG1 refractometer, Atago, Tokyo, Japan) and to ensure euhydration. Each subject completed two fluid-replacement strategies (outlined below) in a counterbalanced, randomized manner to control for order effects.

The subjects' clothes and shoes were weighed. After this, they inserted a rectal thermistor (Single use Mon-a-therm Thermistor 400 Series, Mallinckrodt, Juárez Chih, Mexico) 10-12 cm beyond the anal sphincter to monitor rectal temperature (Tre). A cannula (Optiva 20G, Ethicon S.p.A, Pomezia, Italy) was inserted into a superficial forearm vein and kept patent during the test with a 0.9% saline solution (Sterisafe, Pharmacia & Upjohn, Perth, Australia). The subjects then entered the environmental chamber, which was maintained at an ambient temperature and relative humidity of 29.3 ± 0.4°C and 36.7 ± 4.6%, respectively. A fan that was set at the same speed (~15 km·h−1) and position (1.5 m from subject's face) for all treatments maintained air circulation. Skin thermistors (YSI 400 Series, YSI, Yellow Springs, OH) were attached to the forearm, chest, midthigh, and posterior calf in accordance with Ramanathan (26) to estimate mean skin temperature (Tsk). The subjects were then weighed, and nude mass was later determined by subtraction of the equipment and clothing masses.

The subjects sat on the stationary cycle ergometer for 5 min to establish postural stability. A preexercise blood sample was then collected, and their first fluid bolus was consumed. They undertook a 2-h submaximal ride (SUBMAX) on the ergometer at 53% MAP while wearing a bicycle helmet to further simulate road conditions. This power output was chosen to replicate one seen in the field during a semimountainous professional male road race (24). During the SUBMAX, they consumed either a HIGH or LOW fluid intake. Once the 2-h SUBMAX was complete, the subjects were towel dried and reweighed. Then, each subject rode his own bicycle, which was fitted with a dynamically calibrated powermeter (SRM professional version powermeter, Schoeberer Rad Messtechnik Training System, Jülich, Germany), until exhaustion on a customized treadmill (Australian Institute of Sport, Canberra, Australia) at 8% gradient and at a speed that elicited 88% MAP. This intensity was selected to result in a time to exhaustion of 10-30 min, which approximates hill-climb durations commonly observed in international cycling road races (T.R. Ebert, unpublished observations). The transition from the stationary ergometer to the treadmill was < 3 min. Before the simulated hill climb, the SRM powermeter was zeroed and the tires were inflated to the same pressure (100 psi). Subjects received similar verbal encouragement throughout all treatments, and all timing devices in view were covered. The subjects indicated when they could no longer continue by verbally communicating with the investigator or by holding onto the treadmill's safety bar, at which time the session was stopped immediately. On cessation, they were again towel dried and weighed, then reweighed after voiding. Subjects did not receive any feedback on their performance until completion of the second time-to-exhaustion trial.

Fluid-replacement strategies.

  1. LOW = 50 mL every 15 min (total = 0.4 L) of water, starting at 0 min and finishing at 105 min. A sport gel (Gu Energy Gel, Sports Street Marketing, Berkeley, CA) was consumed at 0, 15, 45, 60, 90, and 105 min into the 2-h SUBMAX. The sport gels were used to match the CHO content to that of the HIGH treatment, and their total water content was approximately 45 g. The volume of fluid consumption was representative of that which has been observed for some cyclists in a competitive environment (personal observations of the present authors; supported by reviews by other sport scientists working with cyclists (1)).
  2. HIGH = 300 mL every 15 min (total = 2.4 L) of a commercially available 7% carbohydrate (CHO)-electrolyte solution (Gatorade, Pepsico Australia, NSW, Australia). This volume was chosen to represent the upper limits of fluid intake suggested by the ACSM guidelines (5).

No attempt was made to match electrolyte contents, and the subjects could not be blinded to the treatments because of the noticeable differences in fluid volume and sport gel consumption. However, cyclists were aware of both possibilities of performance change: that the LOW fluid treatment could be beneficial because they would be lighter for the uphill climb, or that the LOW fluid trial could result in a compromised performance because they were dehydrated. The fluid, which was allowed to remain at room temperature, was consumed within 1 min. No fluid or CHO were consumed during the time-to-exhaustion hill-climb trials.

Physiological monitoring.

Tre and Tsk, environmental conditions (Kestrel 4000 Pocket Weather Tracker, Nielsen Kellerman, Boothwyn, PA), and HR were monitored every 5 min during SUBMAX and every 2 min during the hill climb. RPE (3), thermal sensation (30), and stomach fullness (0 = empty to 5 = uncomfortably bloated) were recorded every 15 min, immediately before fluid intake during the SUBMAX, after the SUBMAX, and after the hill climb. Subjects were weighed before, at 40 min, at 85 min, after SUBMAX, after the hill climb, and after voiding. For body mass determination at 40 and 85 min, each subject dismounted the ergometer, towel dried, removed his helmet, and was weighed. This disruption to cycling took no more than 30 s. At 20, 50, 80, and 110 min during the SUBMAX, they expired for 5 min into the indirect calorimetry system to determine V˙O2, V˙CO2, minute ventilation, and respiratory exchange ratio (RER). Sweat rates were determined from the decrease in body mass and urine excreted plus the amount of fluid ingested. No corrections were made for the masses of fat and carbohydrate catabolized (25).

Blood analyses.

After the preexercise blood sample, subsequent samples were collected at 5, 60, and 120 min during SUBMAX and after the hill climb. Two-milliliter blood samples were placed in EDTA tubes and immediately analyzed for hematocrit (Hct), hemoglobin (Hb), and full blood count using an ADVIA 120 Hematology System Analyzer (Bayer Diagnostics, Tarrytown, NY). Percent changes in plasma volume (%ΔPV) were assessed using the method of Dill and Costill (7). Capillary blood samples (100 μL) were obtained at the same time from a finger prick to determine BLa, blood glucose (BGl), pH, and bicarbonate ion concentration (Radiometer ABL 700 Series, Copenhagen, Denmark).

Statistical analyses.

Data from the SUBMAX ride were compared using a two-way repeated-measures ANOVA (treatment × time). Dependent t-tests were administered in the event of either a significant main effect for time or an interaction. Various dependent t-tests were conducted on the hill-climbing data. The statistical analyses were performed using Statistica 6.0 (Statsoft Inc, Tulsa, OK). All data are reported as mean ± standard deviation (SD) unless otherwise specified, and statistical significance was recognized when P ≤ 0.05.


Hydration status, body mass changes, and sweat rates.

A comparable hydration status was achieved by all subjects before the two treatments, as indicated by similar pretesting values for body mass (LOW = 70.2 ± 5.6 kg vs HIGH = 70.4 ± 5.6 kg), urine specific gravity (LOW = 1.006 ± 0.006 vs HIGH = 1.005 ± 0.004), Hb (LOW = 15.6 ± 0.6 vs HIGH 15.2 ± 1.0 g·L−1), and Hct (LOW = 44.9 ± 1.2 vs HIGH 43.8 ± 2.3%). An average loss of 1.72 ± 0.28 kg (−2.5 ± 0.5% body mass) and a gain of 0.18 ± 0.28 kg (+0.3 ± 0.4% body mass) were attained at the end of the 2-h SUBMAX for the LOW and HIGH treatments, respectively. On reaching exhaustion from the hill-climbing trial and the previous SUBMAX effort, subjects had lost 3.6 ± 0.6 and 1.3 ± 0.5% body mass for the LOW and HIGH treatments, respectively (P < 0.05). There was no significant difference between the two treatments for sweat rate during the 2-h SUBMAX.

Simulated hill-climb performance.

There were no differences in treadmill speed between the treatments (both LOW and HIGH = 16.4 ± 1.4 km·h−1). However, a significant difference occurred for the absolute power output required to maintain this speed (P = 0.003). Subjects in the LOW treatment produced 308 ± 28 W compared with 313 ± 28 W during the HIGH treatment to maintain the same speed. There was a small change in relative power output (W·kg−1) between the LOW (3.92 ± 0.30 W·kg−1) and HIGH (3.88 ± 0.30 W·kg−1; P = 0.008) treatments.

Subjects started the simulated hill climb on the treadmill approximately 1.9 kg lighter (range: 1.4-2.2 kg) after the LOW fluid strategy. During the inclined treadmill ride to exhaustion, subjects in the LOW fluid treatment reached exhaustion 5.6 min earlier than those on the HIGH fluid regimen (Fig. 1). This equated to an average decrement in time to exhaustion of 28.6 ± 13.8% after restricted fluid intake. Despite the counterbalanced design, all subjects demonstrated the same response.

Time to exhaustion on treadmill (8% gradient and set speed) after 2 h of submaximal riding (SUBMAX) and consumption of LOW (0.4 L) vs HIGH (2.4 L) fluid intake (mean ± SD; * P = 0.002).

Tre, Tsk, and HR.

One subject's Tre data were omitted because of technical difficulties during the trial. There was a significant treatment-by-time interaction (P < 0.001) for Tre during the SUBMAX but not for Tsk. Dependent t-tests showed a higher Tre after 50 min of SUBMAX in the LOW treatment, and this persisted until the end of the simulated hill climb (Fig. 2). The rate of rise in Tre from the start to end of SUBMAX was greater for the LOW treatment (0.9°C·h−1) compared with the HIGH treatment (0.6°C·h−1; P = 0.003). Tre at the end of the SUBMAX were 38.9 ± 0.2 and 38.3 ± 0.2°C (P < 0.001) and at exhaustion were 39.5 ± 0.3 and 39.1 ± 0.3°C (P < 0.001) for the LOW and HIGH treatments, respectively. From the end of the SUBMAX to reaching exhaustion, Tre rose 0.5°C in 13.9 min for the LOW treatment and 0.8°C in 19.5 min for the HIGH treatment. There was a significant treatment-by-time interaction for SUBMAX HR (P < 0.001), with the difference starting at 75 min of SUBMAX (except 115 min). Dependent t-tests revealed a difference between the final hill-climb HR for the LOW (187 ± 14 bpm) and HIGH (183 ± 14 bpm; P = 0.02) fluid regimens (Fig. 2).

Changes in heart rate (▪), rectal temperature (▴), and mean skin temperature (•) during 2 h of submaximal riding (SUBMAX) and simulated hill-climb ride to exhaustion (TE) after LOW (0.4 L) and HIGH (2.4 L) fluid treatments. * Significant (P ≤ 0.02) from 50 min onwards; # 115 min, significant (P ≤ 0.05) from 75 min onwards.

Blood glucose and lactate concentrations.

No differences were detected between the LOW and HIGH treatments for BGl, pH, bicarbonate ion concentration, or BLa. After the simulated hill climb, BLa were 11.7 ± 2.7 and 11.6 ± 3.2 mM for the LOW and HIGH treatments, respectively. Despite no difference in end BLa, there was a significant difference in the rate of change in BLa (LOW 0.69 ± 0.26 vs HIGH 0.50 ± 0.23 mM·min−1; P = 0.006).

Oxygen consumption.

There were no differences for V˙O2 (LOW 2.91 ± 0.24 vs HIGH 2.87 ± 0.26 L·min−1), V˙CO2 (LOW 2.65 ± 0.20 vs HIGH 2.66 ± 0.23 L·min−1), RER (LOW 0.91 ± 0.03 vs HIGH 0.92 ± 0.03), or ventilation (STPD: LOW 56.6 ± 3.7 vs HIGH 56.0 ± 4.2 L·min−1) as a result of the fluid strategies during SUBMAX.

Plasma volume.

A trend for a larger decline in PV (P = 0.17) in the final stages of SUBMAX and hill climb was evident for the LOW treatment. These PV changes were determined using resting values as the baseline. Analyses were also conducted using the blood sample taken at 5 min into SUBMAX as the baseline. This acknowledges the acute changes in PV attributable to the redistribution of body fluids from vascular to extravascular space and the increase in blood pressure that occurs within 5-10 min of the onset of exercise. These acute changes represent a shift as opposed to actual fluid loss attributable to dehydration (9). Similar PV changes were observed when using either the resting or 5 min into SUBMAX measures as the baseline.

Rating of perceived exertion, thermal sensation, and stomach fullness.

RPE, thermal sensation, and stomach-fullness ratings were not significantly different across time between the LOW and HIGH fluid treatments. On cessation of the hill-climbing effort, there were similar values for RPE (LOW = 19 ± 1 vs HIGH = 19 ± 1), thermal sensation (LOW = 6.5 ± 0.5 vs HIGH = 6 ± 1), and stomach fullness (LOW = 3 ± 1 vs HIGH = 3.5 ± 1).


This unique study required subjects to ride their own bicycle on an inclined treadmill to simulate actual hill climbing in the field following different fluid strategies. The novel finding of this research was the significant reduction in the time to exhaustion (P = 0.002) for this protocol after the intake of 0.4 L (LOW) compared with 2.4 L (HIGH) of fluid during the preceding 2 h of submaximal cycling. This reduction in exercise capacity occurred despite the lower power output required to achieve a given riding speed attributable to the 1.9-kg loss when subjects were on the LOW fluid-intake regimen. This provides evidence, previously identified by Coyle (6) as lacking in the scientific literature, of the impact of dehydration on performance where changes in body mass could, in fact, provide benefit by reducing the energy cost of the exercise task. We found that dehydration-induced hyperthermia outweighed the theoretical benefit of a reduction in body mass on the power-to-mass ratio and energy cost during cycling (functional dehydration). The fluid-replacement strategy adopted for our HIGH fluid treatment met the goal of full replacement of sweat losses and represented an upper limit of the range of fluid intakes suggested within the ACSM guidelines (5), whereas the LOW fluid treatment mimicked that observed in some competitive cyclists.

We acknowledge that our subjects were not blinded to the treatment protocol and may have been influenced by their preconceived beliefs about these treatments. It is possible that they were aware of previous studies that found that dehydration has an adverse effect on cycling time to exhaustion (17,19,29) or time to complete a given amount of work (2). But, they also may have been motivated by the prevailing culture within cycling that prizes the benefits of reducing body mass to improve hill-climbing ability.

All subjects showed impaired performance after the LOW treatment, despite the counterbalanced order. The unique design of our study attempted to simulate a climb in a race or a hill-top finish by having the subjects ride their own bicycles on an inclined treadmill at an intensity similar to that seen in a professional men's cycle race. This approach is different from those of other studies (8,17,19) that used stationary ergometers or cycle trainers. It is important to acknowledge the reliability of time-to-exhaustion tests. Although some researchers report time trials (also referred to as constant-work tests) as highly reliable with a coefficient of variation (CV) for duration of 0.9-4.6% for cycling tasks (14), this mode does not reflect the type of effort that is commonly observed in a mass-start cycling road race involving sustained hill climbing. Road racing over mountainous terrain requires the rider to stay with the peloton for as long as possible while climbing at the intensity dictated by the leaders. When a rider cannot sustain the required pace, he or she loses contact with the group. For these reasons, and to provide ecological validity, we employed a time-to-exhaustion or constant-power test. This type of performance task has a reported reliability for duration of 1.7-48% (14) depending on the intensity of the task. Hopkins et al. (14) highlight that constant-power tests are not an inferior performance measure compared with constant-duration tests, despite having a large CV, because a large change in time to exhaustion will result from only a small change in the power output that can be sustained by the individual. Mean power output during a constant-power test has a low CV for efforts of up to 60 min (14) and, thus, a small change in power output and large change in time to exhaustion, such as the one that resulted from the loss in body mass during our study, can be detected with confidence.

At the end of submaximal riding, LOW fluid intake resulted in a significantly higher Tre compared with HIGH. This difference continued until the cessation of the simulated hill climb. Other researchers (2,8,19) have also found a higher Tre at the end of a prescribed exercise task when subjects were given lower fluid volumes during exercise. McConell et al. (19) used treatments of no fluid or 50% (0.72 ± 0.03 L) or 100% (1.47 ± 0.05 L) replacement of sweat losses during 1 h of cycling (45 min at 80% V˙O2max followed by 15 min all out), and Below et al. (2) had subjects consume 200 or 1330 mL during an approximately 50-min cycling effort (50 min at 80% V˙O2max + time trial). Ganio et al. (8) provided no fluid or full fluid replacement to their subjects during 2 h of submaximal exercise, which was followed by a graded exercise test. Unlike our study, Below et al. (2) measured the time to complete a set amount of work in warm conditions (31°C), and Ganio et al. (8) used a graded exercise test to measure performance/exercise capacity, whereas our subjects rode to exhaustion at a set pace in warm conditions. In constant-work or time-trial protocols, subjects can choose a pacing strategy that maintains their body temperature below a level that is critical for the onset of fatigue (15). Our subjects had no control over the pace (speed of the treadmill), so they had to stop when they could no longer sustain the required intensity. Although the testing methods of Below et al. (2) and Ganio et al. (8) were different from ours, decreased fluid intake had an adverse effect on performance/exercise capacity in these three studies. Despite the different performance tasks, there were similar changes in Tre after 50-60 min of submaximal exercise as a result of the ingested fluid volume. The significant difference in Tre and a higher SUBMAX HR in the second hour of exercise support the findings of Montain and Coyle (20), who demonstrated that Tre and HR remain lower for longer (~60 min) when fluid is consumed at regular intervals from the commencement of exercise. Significant differences in HR may take longer to occur than those for Tre. Just as Marino et al. (17) found no difference in HR after 80 min of exercise in a fluid-restricted state, we did not see a statistically significant difference in HR until 75 min of submaximal exercise in the heat (Fig. 2).

The attainment of a critical Tre has been suggested as a contributor to the termination of exercise in the heat with this point being reached regardless of the initial Tre or rate of heat storage (9). However, the present study and recent research by Marino et al. (17) suggest that factors other than the attainment of a critical Tre may influence fatigue. The latter investigations (17) report that fluid restriction was associated with an accelerated increase in Tre as opposed to the attainment of a critical Tre. In the present study, the LOW treatment resulted in a greater rate of rise in Tre during the submaximal ride compared with the HIGH treatment. This may have influenced the onset of fatigue in the subsequent hill climb. Indeed, during the hill climb, despite starting and finishing at a higher Tre, the LOW treatment resulted in a similar rate of rise in Tre as in the HIGH trial. The between-treatments disparity for final Tre may indicate that in the LOW treatment, subjects attained fatigue on reaching a high Tre, whereas fatigue in the HIGH treatment was associated with reaching the threshold for the rate of change in Tre during the hill climb.

Both hyperthermia and dehydration have been implicated in the development of fatigue during exercise in the heat. Our research model did not allow us to deduce whether dehydration, hyperthermia, or a combination of both resulted in the decrement in performance. Very few studies have been designed to investigate this (11,23). Nybo et al. (23) had subjects complete 2 h of submaximal cycling in a euhydrated and dehydrated state followed by a 5- to 8-min time-to-exhaustion test. They found that hyperthermia contributed significantly to a reduction in V˙O2max. Their subjects performed well when they were dehydrated but not hyperthermic; however, their performance dropped considerably when thermal stress was added. The effects of hyperthermia are further supported by González-Alonso et al. (11), who concluded that the superimposition of dehydration on a hyperthermic subject causes significant cardiovascular alterations. This was recently supported by Ganio et al. (8), who found a 14% lower stroke volume and a 9% reduction in V˙O2peak after 2 h of submaximal cycling and no fluid ingestion. Our subjects were dehydrated by 2.5% body mass compared with the 4% for those of other investigators (11,23), and they reached a final Tre that was similar to that reported by González-Alonso et al. (11) but nearly 0.5°C higher than that reported by Nybo et al. (23). This may indicate that hyperthermia had a greater impact on our subjects' ability to continue exercising, because they were less dehydrated but experienced greater increases in Tre. Further studies should involve subjects completing the same experimental design as ours (fluid strategy and performance task) in cooler conditions to assist in determining the role of hyperthermia and/or dehydration in impairing athletic performance.

Research that has not separated the effects of dehydration and/or hyperthermia on performance has reported a reduction in muscle endurance when subjects are dehydrated. Montain et al. (20) decreased body mass by 4% and observed a 15% lower muscle endurance, and they also reported no differences in muscle pH after isolated-leg exercise to exhaustion. This is comparable with our findings and with previously published data (23) documenting that postexercise BLa and blood pH are similar after exhaustive exercise in both a hydrated and hypohydrated state, despite a hypohydrated state producing a reduction in performance time. Nybo and coworkers (23) have suggested that a similar final BLa value after exercise of varying duration reflects an increase in glycolytic ATP production when subjects are hyperthermic. Although hydration status seemed not to influence postexercise BLa and pH, there was a noticeable increase in the rate of lactate accumulation during the LOW fluid treatment when subjects worked at approximately 88% MAP. Hargreaves and colleagues (12) found that, when no fluid intake was allowed, muscle lactate levels were higher after 2 h of cycling exercise at 67% V˙O2max than when fluid ingestion was permitted. It was suggested that fluid intake reduced muscle glycogen use via attenuation of the rise in epinephrine. Unfortunately, our data do not allow us to identify the extent to which the increased BLa is explained by a reduced lactate-clearance rate during exercise when dehydrated/hyperthermic.

In summary, exercise-induced dehydration (~2 kg) resulting from a fluid-intake regimen that is commonly observed in the field during cycling races does not seem to improve cycling hill-climbing performance in a warm laboratory environment. Despite the observed advantage of a reduced metabolic cost associated with a lower body mass, subjects performed significantly worse during the simulated hill-climbing effort in the heat. These data suggest that modest dehydration does not provide an advantage for cycling hill-climbing performance and that the thermal and cardiovascular penalties imposed by a water deficit are important. Our findings are also pertinent to other cycling disciplines such as mountain biking, where even fewer opportunities are available for drinking because of the highly technical sections and off-road terrain. Additional research is required to determine whether subtle reductions in total body water are associated with dramatic reductions in hill-climbing performance in the field.

The local cyclists and triathletes involved in this study are thanked for their time and commitment. We are grateful to Dr Jim Martin for his technical comments and insight. Special recognition is also given to Professor Allan Hahn and the entire Department of Physiology at the Australian Institute of Sport for support and advice in areas of experimental design, data analysis, equipment calibration, and data collection.

This study was funded by the Australian Sports Commission, Cycling Australia, and Flinders University.


1. Atkinson, G., R. Davison, A. Jeukendrup, and L. Passfield. Science and cycling: current knowledge and future directions for research. J. Sports Sci. 21:767-787, 2003.
2. Below, P. R., R. Mora-Rodríguez, J. González-Alonso, and E. F. Coyle. Fluid and carbohydrate ingestion independently improve performance during 1 h of intense exercise. Med. Sci. Sports Exerc. 27:200-210, 1995.
3. Borg, G. Perceived exertion as an indicator of somatic stress. Scand. J. Rehab. Med. 23:92-98, 1970.
4. Cheuvront, S. N., and E. M. Haymes. Thermoregulation and marathon running: biological and environmental influences. Sports Med. 31:743-762, 2001.
5. Convertino, V. A., L. E. Armstrong, E. F. Coyle, et al. Position stand on exercise and fluid replacement. Med. Sci. Sports Exerc. 28:i-vii, 1996.
6. Coyle, E. F. Fluid and fuel intake during exercise. J. Sports Sci. 22:39-55, 2004.
7. Dill, D. B., and D. L. Costill. Calculations of percentage changes in volumes of blood, plasma, and red cells in dehydration. J. Appl. Physiol. 37:247-248, 1974.
8. Ganio, M. S., J. E. Wingo, C. E. Carroll, M. K. Thomas, and K. J. Cureton. Fluid ingestion attenuates the decline in V˙O2peak associated with cardiovascular drift. Med. Sci. Sports. Exerc. 38:901-909, 2006.
9. Gisolfi, C. V., and D. R. Lamb. Fluid homeostasis during exercise. In: Perspectives in Exercise Science and Sports Medicine, C. V. Gisolfi and D. R. Lamb (Eds.). Carmel, IN: Benchmark Press, 1990, pp. 285-297.
10. González-Alonso, J., R. Mora-Rodríguez, P. R. Below, and E. F. Coyle. Dehydration reduces cardiac output and increases systemic and cutaneous vascular resistance during exercise. J. Appl. Physiol. 79:1487-1496, 1995.
11. González-Alonso, J., R. Mora-Rodríguez, P. R. Below, and E. F. Coyle. Dehydration markedly impairs cardiovascular function in hyperthermic endurance athletes during exercise. J. Appl. Physiol. 82:1229-1236, 1997.
12. Hargreaves, M., P. Dillo, D. Angus, and M. Febbraio. Effect of fluid ingestion on muscle metabolism during prolonged exercise. J. Appl. Physiol. 80:363-366, 1996.
13. Heil, D. P., O. F. Murphy, A. R. Mattingly, and B. K. Higginson. Prediction of uphill time-trial bicycling performance in humans with a scaling-derived protocol. Eur. J. Appl. Physiol. 85:374-382, 2001.
14. Hopkins, W. G., E. J. Schabort, and J. A. Hawley. Reliability of power in physical performance tests. Sports Med. 31:211-234, 2001.
15. Kuipers, H., F. T. J. Verstappen, H. A. Keizer, and P. Guerten. Variability of aerobic performance in the laboratory and its physiological correlates. Int. J. Sports Med. 6:197-201, 1985.
16. Lucía, A., H. Joyos, and J. L. Chicharro. Physiological response to professional road cycling: climbers vs time trialists. Int. J. Sports Med. 21:505-512, 2000.
17. Marino, F. E., D. Kay, and N. Serwach. Exercise time to fatigue and the critical limiting temperature: effect of hydration. J. Thermal Biol. 20:21-29, 2004.
18. McConell, G. K., C. M. Burge, S. L. Skinner, and M. Hargreaves. Influence of ingested fluid volume on physiological responses during prolonged exercise. Acta Physiol. Scand. 160:149-156, 1997.
19. McConell, G. K., T. J. Stephens, and B. J. Canny. Fluid ingestion does not influence intense 1-h exercise performance in a mild environment. Med. Sci. Sports Exerc. 31:386-392, 1999.
20. Montain, S., S. Smith, R. Mattot, G. Zientara, F. Jolesz, and M. Sawka. Hypohydration effects on skeletal muscle performance and metabolism. J. Appl. Physiol. 84:1889-1894, 1998.
21. Montain, S. J., and E. F. Coyle. Influence of the timing of fluid ingestion on temperature regulation during exercise. J. Appl. Physiol. 75:688-695, 1993.
22. Nevill, A. M., S. A. Jobson, R. C. R. Davison, and A. E. Jeukendrup. Optimal power-to-mass ratios when predicting flat and hill-climbing time-trial cycling. Eur. J. Appl. Physiol. 97:424-431, 2006.
23. Nybo, L., T. Jensen, B. Nielsen, and J. González-Alonso. Effects of marked hyperthermia with and without dehydration on V˙O2 kinetics during intense exercise. J. Appl. Physiol. 90:1057-1064, 2001.
24. Padilla, S., I. Mujika, J. Orbañanos, J. Santisteban, F. Angulo, and J. J. Goiriena. Exercise intensity and load during mass-start stage races in professional road cycling. Med. Sci. Sports Exerc. 33:796-802, 2001.
25. Pugh, L. G. C. E., J. L. Corbett, and R. H. Johnson. Rectal temperatures, weight losses, and sweat rates in marathon running. J. Appl. Physiol. 23:347-352, 1967.
26. Ramanathan, N. L. A new weighting system for mean surface temperature of the human body. J. Appl. Physiol. 19:531-533, 1964.
27. Robinson, T. A., J. A. Hawley, G. S. Palmer, et al. Water ingestion does not improve 1-h cycling performance in moderate ambient temperatures. Eur. J. Appl. Physiol. 71:153-160, 1995.
28. Russell, G., C. J. Gore, M. J. Ashenden, R. Parisotto, and A. G. Hahn. Effects of prolonged low doses of recombinant human erythropoietin during submaximal and maximal exercise. Eur. J. Appl. Physiol. 86:442-449, 2002.
29. Walsh, R. M., T. D. Noakes, J. A. Hawley, and S. C. Dennis. Impaired high-intensity cycling performance time at low levels of dehydration. Int. J. Sports Med. 15:392-398, 1994.
30. Young, A. J., M. N. Sawka, Y. Epstein, B. Decristofano, and K. B. Pandolf. Cooling different body surfaces during upper and lower body exercise. J. Appl. Physiol. 63:1218-1223, 1987.


©2007The American College of Sports Medicine