Effects of caffeine ingestion on body fluid balance and thermoregulation during exercise BAREKET FALK~ Faculty of Health Sciences, ,McMaster University, Hamilton, Ont., Canada E8N 325
RUTHBURSTEIN, JOSEFRBSENBLUM, AND YAIRSHAPIRB
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by University of Otago on 01/02/15 For personal use only.
Heller Institute of Medical Research, Sheba Medical Centre, Tel-Hashomer, Sackler School of Medicine, Tel-Aviv University, Israel
E . ZYLBER-KATZ Clinical Pharmacology Unit, Hebrew University Medical School, J e m l e r n , Israel AND
N. BASHAN Pediatric Research Laboratory, Soroka Medical Centre, Ben Gurion University, Beer Sheba, Israel Received September 28, 1989 FALK,B., BURSTEIN, R., RBSENBLUM, J., SHAPRIB, Y.,ZYLBBR-KATZ, E., and BASHAN, N. 1990. Effects of caffeine ingestion on body fluid balance and thermoregulation during exercise. Can. J. Physiol. Pharmacol. 68: $89-892. This study investigated the effects of caffeine supplementation on thermoregulation and body fluid balance during prolonged exercise in a thennoneutral environment (25"C, 50% RH). Seven trained male subjects exercised on a treadmill at an intensity of 70 -75 % of maximal oxygen consumption to self-determined exhaustion. Subjects exercised once after caffeine and once after placebo ingestion, given in a double-blind crossover design. Five milligrams per kilogram body weight of caffeine followed by 2.5 mg . kg-I of caffeine were given 2 and 0.5 h before exercise, respectively. Rectal temperature was recorded and venous blood samples were withdrawn every 15 min. Water loss and sweat rate were calculated from the difference between pre- and post-exercise body weight, corrected for liquid intake. Following caffeine ingestion, when compared with placebo, no significant difference in final rectal temperature or in percent change in plasma volume were found. No significant differences were observed in total water loss (1376 f 154 vs. 1141 f 158 mL, respectively), sweat rate (12.4 f 1.1 vs. 10.9 f 0.7 g . m-' . min-', respectively), rise in rectal temperature (2.1 f 0.3 vs. 1.5 f 0.4"C, respectively), nor in the calculated rate of heat storage during exercise (134.4 f 17.7 vs. 93.5 f 22.5 W, respectively). Thus, in spite of the expected rise in oxygen uptake, caffeine ingestion under the conditions of this study does not seem to disturb body fluid balance or affect thennoregulation during exercise performance. Key words: caffeine, physical exercise, thermoregulation, fluid balance, dehydration. R., ~ S E N B L UJ., M SHAPRIB, , Y.,ZYLBER-KATZ, E., et BASHAN, N. 1990. Effects of caffeine ingesFALK,B., BURSTEIN, tion on body fluid balance and thermoregulation during exercise. Can. J. Physiol. Pharmacol. 68 : 889 - 892. Cette Ctude a examint les effets d'un supplCment de cafCine sur la thermorCgulation et l'bquilibre des liquides corporels, durant un exercice prolong6 dans un environnement thermoneutre (25 "C, 50% HR). Sept sujets males se sont entraints sur un tapis roulant 2i une intensit6 de 70-75 % de la consomation maximale d'oxygkne, jusqu'i kpuisement. Les sujets se sont entraines une fois aprks l'absorption de cafCine et une fois ap&s celle d'un placebo, selon un test croist i double inconnue. Cinq milligramme par kilogramme poids corporel de cafkine, suivi de 2,5 mg . kg-' de cafCine, ont 6tC donnCs 2 et 0,5 h avant l'exercice, respectivement. La temperature rectiile a 6tt enregistrk et des Cchantillons sanguins prClevCs toutes les 15 min. La perte d'eau et le taux de sudation ont kt6 calculb i partir de la diffkrence entre le poids corporel avant et aprks l'exercice, corrigk pour l'apport hydrique. L'absorptian de cafkine, compartivement 2i celle du placebo, n'a provoquC aucune diffkrence significative dans la temp6rature rectale finale su dans le pourcentage des variations du volume plasmatique. Aucune difference significative n'a CtC observCe dans la perte d'eau tstale (1376 f 154 vs. 1141 f 158 mL, respectivement), le taux de sudation (12,4 f 1,l vs. 10,9 f 0,7 g . m-' min- respectivement), 1'ClCvation de t e w r a t u r e rectale (2,1 f 0,3 vs. 1,5 f 0,4"C, respectivement), ni dans le taux calcult d'accumulation de chaleur durant I'exercice (134,4 f 17,7 vs. 93,5 f 22,5 W, respectivement). Ainsi, malgrC l'C1Cvation prCvue de la capture d'oxygkne, I'abssrption de cafkine, dans les conditions de cette Ctude, ne semble pas perturber l'tquilibre des liquides eorporels ni affecter Ia thermoregulation durant l'exercice. [Traduit par la revue]
',
Introduction Caffeine is a methylxanthine known to stimulate the cardiovascular, respiratory, and central nervous systems, as well as affect smooth and skeletal muscle (Ritchie 1985). Caffeine may also have metabolic and thermogenic effects in animals and in humans. Wager-Srdar et al. (1983) demonstrated elevated colonic temperature in rats following caffeine ingestion. At rest, caffeine ingestion has Been shown to increase the metabolic rate of obese and lean individuals (Acheson et al. 'Author for correspondence. Printed in Canada I lmprimk au Canda
1980; Jung et al. 1981), as well as trained and untrained subjects (LeBlanc et al. 1985; Poehlman et al. 1985). During exercise, there is an increase in metabolic heat production, leading to an increase in sweat secretion and in body fluid loss. Wynaham and Strydom (1969) demonstrated that the severity of water deficit incurred during prolonged exercise is one of the most important factors influencing the rise in body temperature. In addition, gastric emptying time may be a limiting factor for adequate fluid replacement during exercise (Costill and Saltin 1974). As caffeine is known to relax smooth muscle (Ritchie 1985), it may attenuate gastric emptying. This,
CAN. J. PHYSIBL.
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by University of Otago on 01/02/15 For personal use only.
890
PHAWvfACOL. VQL.
coupled with caffeine's diuretic effect (Curaldo and Robertson B983), may further hinder body fluid balance. Gordon et al. (1982) reported no effect of caffeine ingestion on the thermoregulatory response of trained men. However, the above study examined two small groups of subjects in a randomized controlled design, with no crossover of treatments. Thus, this study was undertaken to further investigate the effects of caffeine intake on body fluid balance and thermaregulatory function during exercise performance, using a slightly higher dose of caffeine and a crossover design. It was hypothesized that caffeine ingestion would place an excessive stress on the thermoregulatory mechanism during prolonged exercise.
Materials and methods Subjects Seven trained male volunteers, age 23.8 f 0.9 years, participated in the study. Their mean (f SEM) body weight, height, and body surface area were 73.7 f 2.9 kg, 1.75 f 8.02 m, and 1.89 f 0.04 m2, respectively. Subjects were nonsmokers and not habitual users of caffeine ( 5 8 5 mg . day-l), as determined by a questionnaire. All signed an informed consent and underwent a medical examination prior to testing. Procedure ~ ~ measured ) during a Maximal oxygen consumption ( ~ 0 , was progressive treadmill mnning test at a constant speed of 3.0 -3.6 m . s-I and skpwise grade increments of 2% every 2 min. The test was determined upon self-determined exhaustion or when the subject could not maintain-the mnning speed, in spite of encouragement by the investigator. Vo, was measured by open circuit spirometry, using an automated metabolic. measurement system (MMC Horizon, SensorMedics). The highest Vo, achieved over a period of 15 s was selected as the VO,,x. The sources of caffeine, namely coffee, tea, caffeinated soft drinks, chocolate, and cocoa, were explained to the subjects before participation. Subjects were requested not to change their daily activity and nutritional habits prior to testing. Subjects were also requested to refrain from intake of caffeinated products for 24 h and to fast for 12 h prior to testing days. This was confirmed verbally upon arrival. Four hours pretrial, each subject received a standard low fat meal (0.9 g fats, 35 g carbohydrates, 15 g protein, and 205 kcal) (I cal = 4.1868 J), which aimed to eliminate hunger pains, without supplying lipid energy stores. A treadmill endurance exercise bout was performed following caffeine ingestion and another bout following a placebo ingestion. Caffeine or placebo were given in 100 mL of artificially sweetened drink at a dose of 5 mg kg-' body weight 2 h preexercise and an additional dose of 2.5 mg kg body weight 0.5 h preexercise. Each subject performed both trials at a similar time of day to minimize diurnal variation. Trials were carried out in a double-blind crossover procedure in random order. The exercise consisted of a treadmill walk, at a speed of 1.56 m * s-l, carrying a 22-kg backpack. The grade of elevation was precalculated to achieve work intensity of 70-75% Vo,,, (Pandolf et al. 1977). Exhaustion was determined when subjects could no longer maintain the required walking speed. Ambient temperature was 25 "C with a relative humidity of 50%. Ad libitum water consumption was encouraged and there was no significant difference in water intake between the two trials. B!o& sampki~ag Venous blood samples were drawn before fluid ingestion, preexercise, zund every 15 min during exercise. The last blood sample was drawn upon completion of exercise. Blood was analyzed for caffeine levels (Syva Cs., Palo Alto, CA), hemoglobin (Hb), and hematocrit (Hct) (Coulter S7). Percent change in plasm volume was calculated according to Dill and Costill (1974).
68, 1990
Measurements Rectal temperture (T,) was measured with a rectal thermistor probe (YSI-401), inserted 10 cm beyond the anal sphincter. Heart rate (HR) was monitored continuously with a telemetry system (Siemens Telecust 36). Expiratory gases were analyzed for Vo,, ventilation (V,), and respiratory exchange ratio (RER), as described above. The calculations based on metabolic parameters, considered the values collected following 30 min of walking. Calculations The rate of heat storage (S) was calculated as follows:
-
-
where 0.965 is the specific heat of the body (W kg-' "C- 9,Wt is the body mass (kg), and T, is the rectal temperature ("C). The change in mean skin temperature (Tsk)was unavailable; however, from our past experience with numerous testing protocols at similar environmental conditions, we observed a change in T,, of 1 f 0.2"C. Therefore, we decided to use this value for the purpose of calculation. Values of VO, obtained after 30 min of exercise were applied to calculate the metabolic rate (M), as follows: The external work rate (M,) was calculated as [3] M, = 0.098 G -V (Wt
+ L)
where G is the grade elevation (%), V is the walking velocity (m - s-I), Wt is the body mass (kg), and L is the load carried (kg) (Givoni and Goldman 1972). Metabolic heat production (Mne,)was thus the difference between M and Mw. Total water loss was determined by loss of weight (f 10 g) adjusted for water intake. Mean sweating rate (rh,,) was calculated from total water loss and exercise time. Metabolic and respiratory weight losses were considered negligible and were not taken into account (Mitchell et d. 1972). Statistical analysis Data were analyzed rasing an ANOVA with repeated measures and the paired t-test, where applicable, with significance level at p < 0.05. Data are reported as mean f SEM.
Results Serum caffeine levels were significantly elevated during the caffeine trial from 0.10 0.05 pg mL-I preingestion to 10.74 f 0.98 pg mL- immediately preexercise. After 15 min of exercise, caffeine levels increased to 13.07 f 0.98 pg . mL-l and did not change significantly throughout the remainder of exercise. Time to exhaustion was not significantly different between caffeine and placebo treatments (63.91 f 6.42 vs. 61.60 f 9.99 rnin, respectively). The percent change in plasma volume upon term?nation of exercise was similar following the two treatments (- 1.4 f 1.8 vs. 1.4 f 1.2 %, respectively). or RER were No significant differences in HR, b2, observed between treatments over time. The respective values after 30 min of exercise are reported in Table 1. The response pattern of k2and RER are illustrated in Fig. 1. During the final stages of exercise, VE and JV82 were significantly higher in the caffeine trial when compard with the control (102.0 f 5.6 vs. 84.4 f 5.9 L min-I and 41.0 f 6.7 vs. 36.9 f 1.3 mL kg- I . .in-I, respectively). Mean total water loss, sweat rate, and rectal temperatures were elevated following caffeine ingestion compared with the placebo, although the differences were not statistically significant (Fig. 2, Table 2).
+
a
89 1
FALK BT AL.
TABLE1. Comparison of the metabolic and cardiorespiratory responses during exercise following caffeine and placebo ingestion (7)
r/s2 (mL . kg-'
min-I)
V, (L min- I ) RER
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by University of Otago on 01/02/15 For personal use only.
HR (beats min-I)
Caffeine
Placebo
39.88+ 1.20 90.38f 5.98 0.92 f 0.02 178f5.2
37.93f 1.19 83.47k6.44 0.91 f 0.01 175f 5.9
0-8
caffeine
@----aplacebo
NOTE:The values presented are those observed at 30 min of exercise.
3 6.5 h --7 -1 7 0 15 30 45 60 1
TIME (rnin) FIG.2. Rectal temperature response following caffeine and placebo ingestion. Data are means f SEM. Numbers in parenthesis signify number of subjects. TABLE 2. Temperature and fluid balance response during exercise fob lowing caffeine and placebo ingestion Caffeine
Placebo
p value
1376f154 12.36+ 1.13 38.8f0.2 2.1 k0.3 2.3 k0.3
1141f 158 10.89+0.72 38.5k0.2 1.5f0.4 2.1 1.6
0.128 0.070 Q.068 0.134 0.443
-
Total water loss (mL) Sweat rate (g . m-' - min-9 Final jp,, ("C) A q e (OC) Ajp, ("C h-I)
+
NOTE:T,, rectal tempemre.
0-0 @-@
4
caffeine placebo
1
0.80
Q
15
38
45
60
TIME (rnin)
FIG.1. Oxygen uptake and respiratory exchange ratio observed following caffeine and placebo ingestion. Data are means k SEM. The data represent only four subjects as we had technical difficulties with some of the measurements. The linlited data s f those subjects where technical difficulties were experienced in one of the trials is consistent with the data of those subjects where no difficulties occurred.
The metabolic rate (997.0 f 41.0 vs. 969.3 f 41.0 W), mechanical work rate (185.2 f 8.7 vs. 184.9 f 8.8 W), and calculated rate of metabolic heat production (8 11.1 33.3 vs. 758.0 f 36.3 W) were similar during the caffeine and placebo trials, respectively. The calculated rate of heat storage was slightly, although not significantly, higher following caffeine ingestion compared with the placebo (134.4 f 17.7 vs. 93.5 22.5 W, respectively), which is reflected by the tendency toward a greater rise in Tre (Table 2). No adverse side effects were observed following caffeine ingestion in any of the subjects.
+
+
Discussion This study investigated the effect of caffeine ingestion on thermoregulation during exercise to exhaustion under thermoneutral ambient conditions. The findings did not demonstrate a significant adverse effect of caffeine on body fluid balance or on the thermoregulatory and metabolic response, although slight differences were observed.
Serum caffeine levels continued to increase after initiating exercise, peaking at 15 min after the start of exercise, and leveling off thereafter. According to the pharmakinetics of caffeine, serum levels should peak 30 - 120 min after ingestion (Bonati and Garattini 1984). The increase in serum caffeine levels observed after initiating exercise may be due to the interaction between the first and second doses, ingested 2 h and 30 min before exercise, respectively. The results sf this study are in agreement with Gordon et al(1982) who found no effect of caffeine ingestion on the rise in T,,, sweat rates, and cardiac function during exercise in healthy, trained individuals. The results are also in accordance with Toner et al. (1982) who found no effect of caffeine ingestion on the cardiovascular and metabolic response to submaximal exercise. The final Vo2 values, which were significantly higher following caffeine intake in our study, along with the greater rise in T, are reflected in the tendency toward a greater production and storage of metabolic heat at the end of exercise and are' consistent with the previously documented calorigenic effect of caffeine (Acheson et diH. 1980; Jung et al. 1981; LeBlanc et d. 1985; Lin et d. 1988; Poehlman et al. 1985; Wager-Srdar et al. 1983). In view of caffeine's known thermgenic and diuretic effect (Curaldo and Robertson 1983), the somewhat increased water loss and sweating rate observed following caffeine ingestion, although statistically insignificant, deserve fmrther attention. It has been suggested that caffeine may have a direct effect on the sweat gland (Sato 1977). The author demonstrated that theophylline markedly enhanced sweating rate of single sweat glands in vitro and suggested that methylxanthines, including caffeine, may directly stimulate sweat gland activity, thus increasing m,, and water loss.
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by University of Otago on 01/02/15 For personal use only.
892
CAN. J. PHYSIOL. PHARMACOL. VOL. 68, 1990
The enhanced ventilation at the final stages sf exercise following caffeine intake are in agreement with previous reports (Powers et al. 1985) and may contribute to a greater respiratory evaporation and, thus, a greater water loss. However, the relatively small amount of water lost through respiration (Mitchell et &* 1972) is not likely to contribute significantly to an increased water loss following caffeine ingestion. ,, respiratory evaporation, Although caffeine may affect & and diuresis, and thus, potentidly affect fluid balance, this study does not demonstrate a significant effect of caffeine on fluid balance or thermoregulatisn during exercise. The possible effect of this does of caffeine (approximately six cups of coffee), however, may be masked by the relatively stronger effect of exercise itself on thermoregulation. It should be noted that the dosage of caffeine ingested in each of the studies discussed above did not exceed 5 mg kg-"ody weight. The effect of this dose on exercise performance has been controversial, with some early studies demonstrating a beneficial ergogenic effect (Costill et &. 1978; Essig et d. 1980; Ivy et d. 1979) and later studies reporting no difference (Casd and Leon 198%; Knapik et d . 1983; McNaughton 1987). McNaughton (1987), on the other hand, experimented with various doses and observed an enhancement in maximal performance following a dose of 15 mg kg-%&y weight. In view of the tendency toward increased water loss and heat storage observed in the present study, it is suggested that the effect of such a high dose of caffeine on the thermoregulatory fbnction during exercise deserves fbrther investigation. In summary, caffeine ingestion, at the dosage administered in this study, while exercising in thermoneutral conditions, does not significantly affect water deficit and sweat loss, the rise in Tr,, or the rate of heat storage in the body. The observed tendency toward an increased water loss following caffeine ingestion leads to the speculation that during exercise, performance under more stressfral environmental conditions, or after a higher dose of caffeine, thermoregulation may be hindered. In view of the common use of caffeine among athletes (Applegale et d . 19891, this possibility warrants frarther research. The suggested mechanisms through which caffeine may affect thermoregullatory fbnction remain speculative and require further research.
We thank 0. Kaduri, M. Sverdlove, and Y.Zick-Bachar for their expert technicd assistance and Dr. A. Zisholtz for his medical supervision. ACHESON,K. J., ZAHORSKA-MARKIEQUICZ, B., and PITTET,PR. 1980. Caffeine and coffee: their influence on metabolic rate and substrate utilization in normal weight and obese individuals. Am. J. Clin. Nutr. 33: 989-997. APPLEGALE, E. A., O'TOOLE,M. E., and HILLER,W. D. B. 1989. Race day dietary intake during an ultraendurance triathlon. Med. Sci. Sports Exercise, 2PS: S48. (Abstr.) BONATI,M., and GAWATTINI, S. 1984. Interspecies comparison of caffeine disposition. In Caffeine: perspectives from recent research. E i k d by P. B. Dew. Springer-Verlag, Berlin. CASAL,D. C., and LEON,A. S. 1985. Failure of caffeine to affect substrate utilization during prolonged running. Med. Sci. Sports Exercise, 17: 174 - 179. COSTILL,D. E., and SALTIN,B. 1974. Factors limiting gastric emptying during rest and exercise. J. Appl. Physiol. 37: 679 -683.
COSTILL,D. L., DALSKY,G. D., and FINK, J. B. 1978. Effects of caffeine ingestion on metabolism and exercise performance. Med. Sei. Sports, 18: 155- 158. CUWALDO, P8W., and ROBERTSON, D. 1983. The health consequences of caffeine. Ann. Intern. Med. 98: 6-41. DILL, D. B., and COSTILL,B. L. 1974. Calculation of percentage changes in volumes of blood, plasma and red cells in hydration. J. Appl. Physiol. 37: 247 -248. ESSIG,D., COSTILL,D. L., and VANHANDBL,B. B. 1980. Effects of caffeine ingestion on utilization of muscle glycogen and lipid during leg ergometer cycling. Int. J. Sports Med. I: 86 -98. GIVONI,B.,and GOLDMAN, R. E 1972. Predicting rectal temperature response to work, environment and clothing. J. Appl. Physiol. 32: 812-822. GORDON,N. E , MYBURGH, J. L., DRUGBW, P. EE.,KBMPPF,9. G., CILLIERS, J. E, MQOLMAN, J., and GROBLEW, H. C. 1982. Effects of caffeine ingestion on thermoregulatory and myocardial hnction during endurance performance. S. A. Med. J. 62: 644 -647. IVY,J. L., COSTILL,B. L., FINK,W. J., and LOWER,R. W. 1979. Influence of caffeine and carbohydrate feedings on endurance performance. Med. Sci. Sports, PI: 6-11. JUNG,R. T., SHBTTY,P. S., JAMES,W. P* T., BARRAND, M. A., and CALLINGHAM, B. A. 1981. Caffeine: its effect on catecholamines and metabolism in lean and obese humans. Clin. Sci. 60: 527535. KNAPIK,J. J., JONES,B. H., TONER,M. M., DANIELS,We L., and EVANS,W. J. 1983. Influence of caffeine on serum substrate changes during mnning in trained and untrained individuals. Biochem. Exercise, 13: 5 14 -5 19. LEBLANC,J., JOBIN,M., COTE, J., SAMSON,P., and LABRIE,A. 1985. Enhanced metabolic response to caffeine in exercise-trained human subjects. J. Appl. Physiol. 59: 832-837. LIN, M. T., CHANDWA, A., a d LIU, @. @. 1980. The effects of theophylline and caffeine on thermoregulatory functions of rats at different ambient temperatures. J. Pham. P h a m c o l . 32: 2444 208. M C N A U G H ~L. N , 1987. Two levels of caffeine ingestion on blood lactate and free fatty acids responses during incremental exercise. Res. Q. Exercise Sports, 58: 255 -259. MIXHELL, J. We, NADEL,E. R., and S ~ L W I J KJ., A. J. 1972. Respiratory water losses during exercise. J. Agpl. Physiol. 32: 474 -476. PANWLF,K. B., GIVONI,B., and GOLDMAN, R. E 1977. Predicting energy expenditure with loads while standing or walking very slowly. J. Apgl. Bhysiol. 43: 577 -581. POEHLMAN, E. T., DESPWES,J. p, BBSSETTE,H., FONTAINB,E., TREMBLAY, A., and BQWHARD,C. 1985. Influence of caffeine on the resting metabolic rate of exercise-trained and inactive subjects. Med. Sci. Sports Exercise, 17: 689 -694. POWERS,S. K., DODD,S., WOODYARD, J., and MANGUM, M. 1985. Caffeine alters ventilatory and gas exchange kinetics during exercise. Med. Sci. Sports Exercise, IS: 101 - 106. RIKHIE, J. M. 1985. Central nervous system stimulants: the xanthines. In The pharmacological basis of therapeutics. Edited by L. S. Goodman and A. @. Gilman. 7th ed. McMillan Publishing Co., New York. pp. 589-603. S A ~K., 1977. The physiology, pharmacology and biochemistry of the mcrine sweat gland. Rev. Physiol. Biochem. Bharmacol. 79: 51-131. TONER,M. M., KIRKENDALL, D. T., DELIQ,D. J., CHASE,J. M., CLEAWY, P. A., and Fox, E. L. 1982. Metabolic and cardiovascular responses to exercise with caffeine. Ergonomics, 25: 1 175 - 1183. WAGER-SRDAR, S. A., OKEN,M. M., MORLEY,J. E., and LEVIN, A. S. 1983. Thermoregulatory effects of purines and caffeine. Life Sci. 33: 2431 -2438. WNDHAM,C. H., and STRYWM,N. B. 1969. The danger of an inadequate water intake during marathon mnning. S. A. M d . J. 43: 893 -896.
RUTHBURSTEIN, JOSEFRBSENBLUM, AND YAIRSHAPIRB
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by University of Otago on 01/02/15 For personal use only.
Heller Institute of Medical Research, Sheba Medical Centre, Tel-Hashomer, Sackler School of Medicine, Tel-Aviv University, Israel
E . ZYLBER-KATZ Clinical Pharmacology Unit, Hebrew University Medical School, J e m l e r n , Israel AND
N. BASHAN Pediatric Research Laboratory, Soroka Medical Centre, Ben Gurion University, Beer Sheba, Israel Received September 28, 1989 FALK,B., BURSTEIN, R., RBSENBLUM, J., SHAPRIB, Y.,ZYLBBR-KATZ, E., and BASHAN, N. 1990. Effects of caffeine ingestion on body fluid balance and thermoregulation during exercise. Can. J. Physiol. Pharmacol. 68: $89-892. This study investigated the effects of caffeine supplementation on thermoregulation and body fluid balance during prolonged exercise in a thennoneutral environment (25"C, 50% RH). Seven trained male subjects exercised on a treadmill at an intensity of 70 -75 % of maximal oxygen consumption to self-determined exhaustion. Subjects exercised once after caffeine and once after placebo ingestion, given in a double-blind crossover design. Five milligrams per kilogram body weight of caffeine followed by 2.5 mg . kg-I of caffeine were given 2 and 0.5 h before exercise, respectively. Rectal temperature was recorded and venous blood samples were withdrawn every 15 min. Water loss and sweat rate were calculated from the difference between pre- and post-exercise body weight, corrected for liquid intake. Following caffeine ingestion, when compared with placebo, no significant difference in final rectal temperature or in percent change in plasma volume were found. No significant differences were observed in total water loss (1376 f 154 vs. 1141 f 158 mL, respectively), sweat rate (12.4 f 1.1 vs. 10.9 f 0.7 g . m-' . min-', respectively), rise in rectal temperature (2.1 f 0.3 vs. 1.5 f 0.4"C, respectively), nor in the calculated rate of heat storage during exercise (134.4 f 17.7 vs. 93.5 f 22.5 W, respectively). Thus, in spite of the expected rise in oxygen uptake, caffeine ingestion under the conditions of this study does not seem to disturb body fluid balance or affect thennoregulation during exercise performance. Key words: caffeine, physical exercise, thermoregulation, fluid balance, dehydration. R., ~ S E N B L UJ., M SHAPRIB, , Y.,ZYLBER-KATZ, E., et BASHAN, N. 1990. Effects of caffeine ingesFALK,B., BURSTEIN, tion on body fluid balance and thermoregulation during exercise. Can. J. Physiol. Pharmacol. 68 : 889 - 892. Cette Ctude a examint les effets d'un supplCment de cafCine sur la thermorCgulation et l'bquilibre des liquides corporels, durant un exercice prolong6 dans un environnement thermoneutre (25 "C, 50% HR). Sept sujets males se sont entraints sur un tapis roulant 2i une intensit6 de 70-75 % de la consomation maximale d'oxygkne, jusqu'i kpuisement. Les sujets se sont entraines une fois aprks l'absorption de cafCine et une fois ap&s celle d'un placebo, selon un test croist i double inconnue. Cinq milligramme par kilogramme poids corporel de cafkine, suivi de 2,5 mg . kg-' de cafCine, ont 6tC donnCs 2 et 0,5 h avant l'exercice, respectivement. La temperature rectiile a 6tt enregistrk et des Cchantillons sanguins prClevCs toutes les 15 min. La perte d'eau et le taux de sudation ont kt6 calculb i partir de la diffkrence entre le poids corporel avant et aprks l'exercice, corrigk pour l'apport hydrique. L'absorptian de cafkine, compartivement 2i celle du placebo, n'a provoquC aucune diffkrence significative dans la temp6rature rectale finale su dans le pourcentage des variations du volume plasmatique. Aucune difference significative n'a CtC observCe dans la perte d'eau tstale (1376 f 154 vs. 1141 f 158 mL, respectivement), le taux de sudation (12,4 f 1,l vs. 10,9 f 0,7 g . m-' min- respectivement), 1'ClCvation de t e w r a t u r e rectale (2,1 f 0,3 vs. 1,5 f 0,4"C, respectivement), ni dans le taux calcult d'accumulation de chaleur durant I'exercice (134,4 f 17,7 vs. 93,5 f 22,5 W, respectivement). Ainsi, malgrC l'C1Cvation prCvue de la capture d'oxygkne, I'abssrption de cafkine, dans les conditions de cette Ctude, ne semble pas perturber l'tquilibre des liquides eorporels ni affecter Ia thermoregulation durant l'exercice. [Traduit par la revue]
',
Introduction Caffeine is a methylxanthine known to stimulate the cardiovascular, respiratory, and central nervous systems, as well as affect smooth and skeletal muscle (Ritchie 1985). Caffeine may also have metabolic and thermogenic effects in animals and in humans. Wager-Srdar et al. (1983) demonstrated elevated colonic temperature in rats following caffeine ingestion. At rest, caffeine ingestion has Been shown to increase the metabolic rate of obese and lean individuals (Acheson et al. 'Author for correspondence. Printed in Canada I lmprimk au Canda
1980; Jung et al. 1981), as well as trained and untrained subjects (LeBlanc et al. 1985; Poehlman et al. 1985). During exercise, there is an increase in metabolic heat production, leading to an increase in sweat secretion and in body fluid loss. Wynaham and Strydom (1969) demonstrated that the severity of water deficit incurred during prolonged exercise is one of the most important factors influencing the rise in body temperature. In addition, gastric emptying time may be a limiting factor for adequate fluid replacement during exercise (Costill and Saltin 1974). As caffeine is known to relax smooth muscle (Ritchie 1985), it may attenuate gastric emptying. This,
CAN. J. PHYSIBL.
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by University of Otago on 01/02/15 For personal use only.
890
PHAWvfACOL. VQL.
coupled with caffeine's diuretic effect (Curaldo and Robertson B983), may further hinder body fluid balance. Gordon et al. (1982) reported no effect of caffeine ingestion on the thermoregulatory response of trained men. However, the above study examined two small groups of subjects in a randomized controlled design, with no crossover of treatments. Thus, this study was undertaken to further investigate the effects of caffeine intake on body fluid balance and thermaregulatory function during exercise performance, using a slightly higher dose of caffeine and a crossover design. It was hypothesized that caffeine ingestion would place an excessive stress on the thermoregulatory mechanism during prolonged exercise.
Materials and methods Subjects Seven trained male volunteers, age 23.8 f 0.9 years, participated in the study. Their mean (f SEM) body weight, height, and body surface area were 73.7 f 2.9 kg, 1.75 f 8.02 m, and 1.89 f 0.04 m2, respectively. Subjects were nonsmokers and not habitual users of caffeine ( 5 8 5 mg . day-l), as determined by a questionnaire. All signed an informed consent and underwent a medical examination prior to testing. Procedure ~ ~ measured ) during a Maximal oxygen consumption ( ~ 0 , was progressive treadmill mnning test at a constant speed of 3.0 -3.6 m . s-I and skpwise grade increments of 2% every 2 min. The test was determined upon self-determined exhaustion or when the subject could not maintain-the mnning speed, in spite of encouragement by the investigator. Vo, was measured by open circuit spirometry, using an automated metabolic. measurement system (MMC Horizon, SensorMedics). The highest Vo, achieved over a period of 15 s was selected as the VO,,x. The sources of caffeine, namely coffee, tea, caffeinated soft drinks, chocolate, and cocoa, were explained to the subjects before participation. Subjects were requested not to change their daily activity and nutritional habits prior to testing. Subjects were also requested to refrain from intake of caffeinated products for 24 h and to fast for 12 h prior to testing days. This was confirmed verbally upon arrival. Four hours pretrial, each subject received a standard low fat meal (0.9 g fats, 35 g carbohydrates, 15 g protein, and 205 kcal) (I cal = 4.1868 J), which aimed to eliminate hunger pains, without supplying lipid energy stores. A treadmill endurance exercise bout was performed following caffeine ingestion and another bout following a placebo ingestion. Caffeine or placebo were given in 100 mL of artificially sweetened drink at a dose of 5 mg kg-' body weight 2 h preexercise and an additional dose of 2.5 mg kg body weight 0.5 h preexercise. Each subject performed both trials at a similar time of day to minimize diurnal variation. Trials were carried out in a double-blind crossover procedure in random order. The exercise consisted of a treadmill walk, at a speed of 1.56 m * s-l, carrying a 22-kg backpack. The grade of elevation was precalculated to achieve work intensity of 70-75% Vo,,, (Pandolf et al. 1977). Exhaustion was determined when subjects could no longer maintain the required walking speed. Ambient temperature was 25 "C with a relative humidity of 50%. Ad libitum water consumption was encouraged and there was no significant difference in water intake between the two trials. B!o& sampki~ag Venous blood samples were drawn before fluid ingestion, preexercise, zund every 15 min during exercise. The last blood sample was drawn upon completion of exercise. Blood was analyzed for caffeine levels (Syva Cs., Palo Alto, CA), hemoglobin (Hb), and hematocrit (Hct) (Coulter S7). Percent change in plasm volume was calculated according to Dill and Costill (1974).
68, 1990
Measurements Rectal temperture (T,) was measured with a rectal thermistor probe (YSI-401), inserted 10 cm beyond the anal sphincter. Heart rate (HR) was monitored continuously with a telemetry system (Siemens Telecust 36). Expiratory gases were analyzed for Vo,, ventilation (V,), and respiratory exchange ratio (RER), as described above. The calculations based on metabolic parameters, considered the values collected following 30 min of walking. Calculations The rate of heat storage (S) was calculated as follows:
-
-
where 0.965 is the specific heat of the body (W kg-' "C- 9,Wt is the body mass (kg), and T, is the rectal temperature ("C). The change in mean skin temperature (Tsk)was unavailable; however, from our past experience with numerous testing protocols at similar environmental conditions, we observed a change in T,, of 1 f 0.2"C. Therefore, we decided to use this value for the purpose of calculation. Values of VO, obtained after 30 min of exercise were applied to calculate the metabolic rate (M), as follows: The external work rate (M,) was calculated as [3] M, = 0.098 G -V (Wt
+ L)
where G is the grade elevation (%), V is the walking velocity (m - s-I), Wt is the body mass (kg), and L is the load carried (kg) (Givoni and Goldman 1972). Metabolic heat production (Mne,)was thus the difference between M and Mw. Total water loss was determined by loss of weight (f 10 g) adjusted for water intake. Mean sweating rate (rh,,) was calculated from total water loss and exercise time. Metabolic and respiratory weight losses were considered negligible and were not taken into account (Mitchell et d. 1972). Statistical analysis Data were analyzed rasing an ANOVA with repeated measures and the paired t-test, where applicable, with significance level at p < 0.05. Data are reported as mean f SEM.
Results Serum caffeine levels were significantly elevated during the caffeine trial from 0.10 0.05 pg mL-I preingestion to 10.74 f 0.98 pg mL- immediately preexercise. After 15 min of exercise, caffeine levels increased to 13.07 f 0.98 pg . mL-l and did not change significantly throughout the remainder of exercise. Time to exhaustion was not significantly different between caffeine and placebo treatments (63.91 f 6.42 vs. 61.60 f 9.99 rnin, respectively). The percent change in plasma volume upon term?nation of exercise was similar following the two treatments (- 1.4 f 1.8 vs. 1.4 f 1.2 %, respectively). or RER were No significant differences in HR, b2, observed between treatments over time. The respective values after 30 min of exercise are reported in Table 1. The response pattern of k2and RER are illustrated in Fig. 1. During the final stages of exercise, VE and JV82 were significantly higher in the caffeine trial when compard with the control (102.0 f 5.6 vs. 84.4 f 5.9 L min-I and 41.0 f 6.7 vs. 36.9 f 1.3 mL kg- I . .in-I, respectively). Mean total water loss, sweat rate, and rectal temperatures were elevated following caffeine ingestion compared with the placebo, although the differences were not statistically significant (Fig. 2, Table 2).
+
a
89 1
FALK BT AL.
TABLE1. Comparison of the metabolic and cardiorespiratory responses during exercise following caffeine and placebo ingestion (7)
r/s2 (mL . kg-'
min-I)
V, (L min- I ) RER
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by University of Otago on 01/02/15 For personal use only.
HR (beats min-I)
Caffeine
Placebo
39.88+ 1.20 90.38f 5.98 0.92 f 0.02 178f5.2
37.93f 1.19 83.47k6.44 0.91 f 0.01 175f 5.9
0-8
caffeine
@----aplacebo
NOTE:The values presented are those observed at 30 min of exercise.
3 6.5 h --7 -1 7 0 15 30 45 60 1
TIME (rnin) FIG.2. Rectal temperature response following caffeine and placebo ingestion. Data are means f SEM. Numbers in parenthesis signify number of subjects. TABLE 2. Temperature and fluid balance response during exercise fob lowing caffeine and placebo ingestion Caffeine
Placebo
p value
1376f154 12.36+ 1.13 38.8f0.2 2.1 k0.3 2.3 k0.3
1141f 158 10.89+0.72 38.5k0.2 1.5f0.4 2.1 1.6
0.128 0.070 Q.068 0.134 0.443
-
Total water loss (mL) Sweat rate (g . m-' - min-9 Final jp,, ("C) A q e (OC) Ajp, ("C h-I)
+
NOTE:T,, rectal tempemre.
0-0 @-@
4
caffeine placebo
1
0.80
Q
15
38
45
60
TIME (rnin)
FIG.1. Oxygen uptake and respiratory exchange ratio observed following caffeine and placebo ingestion. Data are means k SEM. The data represent only four subjects as we had technical difficulties with some of the measurements. The linlited data s f those subjects where technical difficulties were experienced in one of the trials is consistent with the data of those subjects where no difficulties occurred.
The metabolic rate (997.0 f 41.0 vs. 969.3 f 41.0 W), mechanical work rate (185.2 f 8.7 vs. 184.9 f 8.8 W), and calculated rate of metabolic heat production (8 11.1 33.3 vs. 758.0 f 36.3 W) were similar during the caffeine and placebo trials, respectively. The calculated rate of heat storage was slightly, although not significantly, higher following caffeine ingestion compared with the placebo (134.4 f 17.7 vs. 93.5 22.5 W, respectively), which is reflected by the tendency toward a greater rise in Tre (Table 2). No adverse side effects were observed following caffeine ingestion in any of the subjects.
+
+
Discussion This study investigated the effect of caffeine ingestion on thermoregulation during exercise to exhaustion under thermoneutral ambient conditions. The findings did not demonstrate a significant adverse effect of caffeine on body fluid balance or on the thermoregulatory and metabolic response, although slight differences were observed.
Serum caffeine levels continued to increase after initiating exercise, peaking at 15 min after the start of exercise, and leveling off thereafter. According to the pharmakinetics of caffeine, serum levels should peak 30 - 120 min after ingestion (Bonati and Garattini 1984). The increase in serum caffeine levels observed after initiating exercise may be due to the interaction between the first and second doses, ingested 2 h and 30 min before exercise, respectively. The results sf this study are in agreement with Gordon et al(1982) who found no effect of caffeine ingestion on the rise in T,,, sweat rates, and cardiac function during exercise in healthy, trained individuals. The results are also in accordance with Toner et al. (1982) who found no effect of caffeine ingestion on the cardiovascular and metabolic response to submaximal exercise. The final Vo2 values, which were significantly higher following caffeine intake in our study, along with the greater rise in T, are reflected in the tendency toward a greater production and storage of metabolic heat at the end of exercise and are' consistent with the previously documented calorigenic effect of caffeine (Acheson et diH. 1980; Jung et al. 1981; LeBlanc et d. 1985; Lin et d. 1988; Poehlman et al. 1985; Wager-Srdar et al. 1983). In view of caffeine's known thermgenic and diuretic effect (Curaldo and Robertson 1983), the somewhat increased water loss and sweating rate observed following caffeine ingestion, although statistically insignificant, deserve fmrther attention. It has been suggested that caffeine may have a direct effect on the sweat gland (Sato 1977). The author demonstrated that theophylline markedly enhanced sweating rate of single sweat glands in vitro and suggested that methylxanthines, including caffeine, may directly stimulate sweat gland activity, thus increasing m,, and water loss.
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by University of Otago on 01/02/15 For personal use only.
892
CAN. J. PHYSIOL. PHARMACOL. VOL. 68, 1990
The enhanced ventilation at the final stages sf exercise following caffeine intake are in agreement with previous reports (Powers et al. 1985) and may contribute to a greater respiratory evaporation and, thus, a greater water loss. However, the relatively small amount of water lost through respiration (Mitchell et &* 1972) is not likely to contribute significantly to an increased water loss following caffeine ingestion. ,, respiratory evaporation, Although caffeine may affect & and diuresis, and thus, potentidly affect fluid balance, this study does not demonstrate a significant effect of caffeine on fluid balance or thermoregulatisn during exercise. The possible effect of this does of caffeine (approximately six cups of coffee), however, may be masked by the relatively stronger effect of exercise itself on thermoregulation. It should be noted that the dosage of caffeine ingested in each of the studies discussed above did not exceed 5 mg kg-"ody weight. The effect of this dose on exercise performance has been controversial, with some early studies demonstrating a beneficial ergogenic effect (Costill et &. 1978; Essig et d. 1980; Ivy et d. 1979) and later studies reporting no difference (Casd and Leon 198%; Knapik et d . 1983; McNaughton 1987). McNaughton (1987), on the other hand, experimented with various doses and observed an enhancement in maximal performance following a dose of 15 mg kg-%&y weight. In view of the tendency toward increased water loss and heat storage observed in the present study, it is suggested that the effect of such a high dose of caffeine on the thermoregulatory fbnction during exercise deserves fbrther investigation. In summary, caffeine ingestion, at the dosage administered in this study, while exercising in thermoneutral conditions, does not significantly affect water deficit and sweat loss, the rise in Tr,, or the rate of heat storage in the body. The observed tendency toward an increased water loss following caffeine ingestion leads to the speculation that during exercise, performance under more stressfral environmental conditions, or after a higher dose of caffeine, thermoregulation may be hindered. In view of the common use of caffeine among athletes (Applegale et d . 19891, this possibility warrants frarther research. The suggested mechanisms through which caffeine may affect thermoregullatory fbnction remain speculative and require further research.
We thank 0. Kaduri, M. Sverdlove, and Y.Zick-Bachar for their expert technicd assistance and Dr. A. Zisholtz for his medical supervision. ACHESON,K. J., ZAHORSKA-MARKIEQUICZ, B., and PITTET,PR. 1980. Caffeine and coffee: their influence on metabolic rate and substrate utilization in normal weight and obese individuals. Am. J. Clin. Nutr. 33: 989-997. APPLEGALE, E. A., O'TOOLE,M. E., and HILLER,W. D. B. 1989. Race day dietary intake during an ultraendurance triathlon. Med. Sci. Sports Exercise, 2PS: S48. (Abstr.) BONATI,M., and GAWATTINI, S. 1984. Interspecies comparison of caffeine disposition. In Caffeine: perspectives from recent research. E i k d by P. B. Dew. Springer-Verlag, Berlin. CASAL,D. C., and LEON,A. S. 1985. Failure of caffeine to affect substrate utilization during prolonged running. Med. Sci. Sports Exercise, 17: 174 - 179. COSTILL,D. E., and SALTIN,B. 1974. Factors limiting gastric emptying during rest and exercise. J. Appl. Physiol. 37: 679 -683.
COSTILL,D. L., DALSKY,G. D., and FINK, J. B. 1978. Effects of caffeine ingestion on metabolism and exercise performance. Med. Sei. Sports, 18: 155- 158. CUWALDO, P8W., and ROBERTSON, D. 1983. The health consequences of caffeine. Ann. Intern. Med. 98: 6-41. DILL, D. B., and COSTILL,B. L. 1974. Calculation of percentage changes in volumes of blood, plasma and red cells in hydration. J. Appl. Physiol. 37: 247 -248. ESSIG,D., COSTILL,D. L., and VANHANDBL,B. B. 1980. Effects of caffeine ingestion on utilization of muscle glycogen and lipid during leg ergometer cycling. Int. J. Sports Med. I: 86 -98. GIVONI,B.,and GOLDMAN, R. E 1972. Predicting rectal temperature response to work, environment and clothing. J. Appl. Physiol. 32: 812-822. GORDON,N. E , MYBURGH, J. L., DRUGBW, P. EE.,KBMPPF,9. G., CILLIERS, J. E, MQOLMAN, J., and GROBLEW, H. C. 1982. Effects of caffeine ingestion on thermoregulatory and myocardial hnction during endurance performance. S. A. Med. J. 62: 644 -647. IVY,J. L., COSTILL,B. L., FINK,W. J., and LOWER,R. W. 1979. Influence of caffeine and carbohydrate feedings on endurance performance. Med. Sci. Sports, PI: 6-11. JUNG,R. T., SHBTTY,P. S., JAMES,W. P* T., BARRAND, M. A., and CALLINGHAM, B. A. 1981. Caffeine: its effect on catecholamines and metabolism in lean and obese humans. Clin. Sci. 60: 527535. KNAPIK,J. J., JONES,B. H., TONER,M. M., DANIELS,We L., and EVANS,W. J. 1983. Influence of caffeine on serum substrate changes during mnning in trained and untrained individuals. Biochem. Exercise, 13: 5 14 -5 19. LEBLANC,J., JOBIN,M., COTE, J., SAMSON,P., and LABRIE,A. 1985. Enhanced metabolic response to caffeine in exercise-trained human subjects. J. Appl. Physiol. 59: 832-837. LIN, M. T., CHANDWA, A., a d LIU, @. @. 1980. The effects of theophylline and caffeine on thermoregulatory functions of rats at different ambient temperatures. J. Pham. P h a m c o l . 32: 2444 208. M C N A U G H ~L. N , 1987. Two levels of caffeine ingestion on blood lactate and free fatty acids responses during incremental exercise. Res. Q. Exercise Sports, 58: 255 -259. MIXHELL, J. We, NADEL,E. R., and S ~ L W I J KJ., A. J. 1972. Respiratory water losses during exercise. J. Agpl. Physiol. 32: 474 -476. PANWLF,K. B., GIVONI,B., and GOLDMAN, R. E 1977. Predicting energy expenditure with loads while standing or walking very slowly. J. Apgl. Bhysiol. 43: 577 -581. POEHLMAN, E. T., DESPWES,J. p, BBSSETTE,H., FONTAINB,E., TREMBLAY, A., and BQWHARD,C. 1985. Influence of caffeine on the resting metabolic rate of exercise-trained and inactive subjects. Med. Sci. Sports Exercise, 17: 689 -694. POWERS,S. K., DODD,S., WOODYARD, J., and MANGUM, M. 1985. Caffeine alters ventilatory and gas exchange kinetics during exercise. Med. Sci. Sports Exercise, IS: 101 - 106. RIKHIE, J. M. 1985. Central nervous system stimulants: the xanthines. In The pharmacological basis of therapeutics. Edited by L. S. Goodman and A. @. Gilman. 7th ed. McMillan Publishing Co., New York. pp. 589-603. S A ~K., 1977. The physiology, pharmacology and biochemistry of the mcrine sweat gland. Rev. Physiol. Biochem. Bharmacol. 79: 51-131. TONER,M. M., KIRKENDALL, D. T., DELIQ,D. J., CHASE,J. M., CLEAWY, P. A., and Fox, E. L. 1982. Metabolic and cardiovascular responses to exercise with caffeine. Ergonomics, 25: 1 175 - 1183. WAGER-SRDAR, S. A., OKEN,M. M., MORLEY,J. E., and LEVIN, A. S. 1983. Thermoregulatory effects of purines and caffeine. Life Sci. 33: 2431 -2438. WNDHAM,C. H., and STRYWM,N. B. 1969. The danger of an inadequate water intake during marathon mnning. S. A. M d . J. 43: 893 -896.
