Effect of caffeine co-ingested with carbohydrate or fat on metabolism and performance in endurance-trained men

Transcription

1 Effect of caffeine co-ingested with carbohydrate or fat on metabolism and performance in endurance-trained men Talia L. Jacobson, Mark A. Febbraio, Melissa J. Arkinstall and John A. Hawley* Exercise Metabolism Group, Department of Human Biology and Movement Science, RMIT University, Bundoora 3083, Victoria and Exercise Physiology and Metabolism Laboratory, Department of Physiology, The University of Melbourne, Parkville 3052, Melbourne, Australia (Manuscript received 6 June 2000; accepted 5 October 2000) We examined the effect of caffeine co-ingested with either carbohydrate or fat on metabolism and performance in eight endurance-trained subjects who performed a random order of four experimental trials consisting of 120 min of steady-state ergometer cycling at 70 % of maximal Oµ uptake (SS) followed by a time trial in which subjects completed a set amount of work (7 kj kg ) as quickly as possible. One hour before SS subjects ingested either 2.6 g kg carbohydrate (CHO); 2.6 g kg CHO + 6 mg kg caffeine (CHO + CAF); 1.2 g kg fat with 2000 U i.v. heparin (FAT); or 1.2 g kg fat with 2000 U i.v. heparin + 6 mg kg caffeine (FAT + CAF). The rate of carbohydrate oxidation was higher (ìmol kg min : CHO, 243 ± 39 and CHO + CAF, 239 ± 30 vs. FAT, 196 ± 48 and FAT + CAF, 191 ± 55; P < 0.05, values are means ± s.d.) and the rate of fat oxidation lower (ìmol kg min : CHO, 19 ± 8 and CHO + CAF, 22 ± 7 vs. FAT, 35 ± 19 and FAT + CAF, 37 ± 17; P < 0.05) with carbohydrate than fat ingestion. Yet despite lower carbohydrate use with fat feeding, the time taken to complete the time trial was less after carbohydrate than after fat ingestion (min: CHO, ± 7.42 and CHO + CAF, ± 5.62 vs. FAT, ± 8.50 and FAT + CAF, ± 7.70; P < 0.05). We conclude that (1) caffeine co-ingested with either carbohydrate or fat meals has no additive effect on substrate utilization or exercise performance and (2) carbohydrate ingestion before exercise improves subsequent time trial performance compared with fat ingestion. Experimental Physiology (2001) 86.1, Over twenty years ago Costill and co-workers reported that caffeine ingestion prolonged exercise time at submaximal work rates (Costill et al. 1978; Ivy et al. 1979; Essig et al. 1980). Since that time, many independent studies have confirmed the ergogenic effect of caffeine on endurance exercise capacity (Graham & Spriet, 1991, 1995; Spriet et al. 1992; Pasman et al. 1995). Several mechanisms have been proposed to explain the ergogenic effect of caffeine. These include an increased reliance on extramuscular free fatty acids (FFA, Erickson et al. 1987; Spriet et al. 1992) andïor intramuscular triacylglycerol (IMTG; Essig et al. 1980) which attenuates the rate of muscle glycogenolysis, and alterations in central neurotransmitter (Waldeck, 1973) andïor neuromuscular function (Kalmar & Cafarelli, 1999). Similar to caffeine, the consumption of carbohydrate electrolyte solutions before (for review see Hawley & Burke, 1997) andïor during (for review see Coggan & Coyle, 1991; Tsintzas & Williams, 1998) prolonged (> 90 min), continuous, submaximal (70 75% of maximal oxygen uptake (ýoµ,max)) exercise delays the onset of fatigue compared with the ingestion of a placebo. The mechanisms underlying the ergogenic effect of carbohydrate ingestion include a decreased rate of muscle glycogen utilization (Tsintzas et al. 1995), a suppression of hepatic glucose production (McConell et al. 1994) andïor the maintenance of high rates of blood glucose oxidation by muscle late in exercise (Coyle et al. 1986). Regardless of the precise mechanism for the enhanced endurance reported with carbohydrate electrolyte feeding, improvements in exercise capacity are often associated with sustained or elevated rates of carbohydrate oxidation rather than any increase in fatty acid (FA) oxidation. Despite some reports showing that the consumption of either medium-chain or long-chain FA before andïor during exercise can improve exercise performance (Van Zyl et al. 1996; Pitsiladis et al. 1999) most evidence suggests that this is not the case (Okano et al. 1996, 1998; Jeukendrup et al. 1998; Whitley et al. 1998; Angus et al. 2000; Hawley et al. 2000). The ingestion of a high-fat meal (plus i.v. heparin) can markedly affect plasma FFA availability (Jeukendrup et al. 1998; Whitley et al. 1998; Hawley et al. 2000). However, the subsequent patterns of fuel utilization are remarkably resistant to alteration during both moderate (Jeukendrup et al. 1996; Whitley et al. 1998) and high-intensity exercise (Horowitz et al. 2000). Caffeine has been shown to increase FA oxidation Publication of The Physiological Society * Corresponding author: john.hawley@rmit.edu.au 2072

2 138 T. L. Jacobson, M. A. Febbraio, M. J. Arkinstall and J. A. Hawley Exp. Physiol and spare muscle glycogen (Essig et al. 1980). However, the precise mechanism responsible for this sparing may only be partially related to the increase in FA mobilization, uptake and oxidation by skeletal muscle. To the best of our knowledge, only one study has investigated the effects of the co-ingestion of caffeine and carbohydrate on exercise performance (Kovacs et al. 1998). In that investigation, well-trained subjects completed a set amount of work (equal to 1 h of cycling) significantly faster when caffeine was co-ingested with a carbohydrate electrolyte solution, compared with when carbohydrate was ingested alone (Kovacs et al. 1998). As carbohydrate ingestion would be expected to elevate plasma insulin concentration, suppress lipolysis, and limit fat oxidation during exercise (Horowitz et al. 1997), it is unlikely that the additive effect of caffeine on performance would have been due to any effect on FA metabolism. Unfortunately, no metabolic data were reported in that study (Kovacs et al. 1998), so the potential mechanism(s) underlying the improved performance could not be elucidated. Therefore, the aim of the current investigation was to examine the effects of caffeine when co-ingested with either carbohydrate or fat meals on exercise metabolism during prolonged submaximal cycling and subsequent time trial performance. Based on previous results showing that caffeine increased IMTG utilization during submaximal exercise (Essig et al. 1980), we hypothesized that the co-ingestion of caffeine with a fat meal would increase FA oxidation above that when fat was ingested alone. However, we also postulated that the co-ingestion of carbohydrate and caffeine would result in the highest rates of carbohydrate oxidation (i.e. the lowest rates of FA oxidation) and the fastest overall time trial performance, the latter by mediating a decrease in subjective effort (Cole et al. 1996). METHODS Subjects and preliminary testing Eight highly trained male cyclists and triathletes (age 21.2 ± 3.7 years, mass 64.7 ± 6.3 kg, ýoµ,max 65.2±3.2ml kg min, maximal power output 5.3 W kg ; values are means ± s.d.) who were cycling 366 ± 93 km week and who had a minimum of 3 years prior endurance training were recruited to participate in this study. The subjects were not habitual caffeine users. All subjects were fully informed of the nature and possible risks of the investigation before giving their written consent. The investigation was approved by the Human Research Ethics Committees of the Royal Melbourne Institute of Technology University and the University of Melbourne and was performed in accordance with the Declaration of Helsinki. On their first visit to the laboratory, each subject performed a maximal, incremental test to exhaustion on an electromagnetically braked cycle ergometer (Lode Instruments, Groningen, The Netherlands). The maximal test protocol has been described in detail previously (Hawley & Noakes, 1992). Expired gas was collected in Douglas bags and the Oµ and COµ contents of dried expirate were analysed by Applied Electrochemistry S-3AÏII and CD-3A analysers (Amatek, Pittsburgh, PA, USA). Analysers were calibrated before each maximal test and all subsequently described experimental trials using commercial gases of known composition. The volume of expired air was measured in a Parkinson-Cowan (Manchester, UK) gas meter calibrated against a Tissot spirometer. Rates of oxygen consumption (ýoµ), carbon dioxide production (ýcoµ), and ventilation (ýe) were determined using conventional equations. ýoµ,max was defined as the highest Oµ uptake subjects attained during two consecutive 30 s sampling periods, while peak sustained power output (PPO) was calculated from the last completed work rate, plus the fraction of time spent in the final noncompleted work rate (Hawley & Noakes, 1992). The results of the maximal test were used to establish 63% of PPO ( 70% of ýoµ,max) which was the workload used for all experimental trials. Training and nutritional control Each subject s training and nutritional status were strictly controlled prior to an experimental trial. Subjects refrained from consuming caffeine-containing substances (coffee and soft drinks) for 72 h before an experiment. Twenty-four hours before an experiment subjects completed a standardized 60 min cycling bout at a work rate equivalent to 70% of ýoµ,max. They were then provided with a pre-packed standard diet with an energy content of 0.21 J (kg body mass (BM)), composed of 63% carbohydrate (8 g (kg BM) ), 20% fat and 17% protein. This food was consumed over the next 24 h during which time subjects refrained from any further strenuous physical activity. Such a diet exercise protocol has previously been shown to result in standardized pre-exercise muscle glycogen stores (Flynn et al. 1987). Experimental trials Each subject completed four experimental trials in random order. On the morning of an experiment, subjects reported to the laboratory between and h after a h overnight fast. After sitting quietly for min, a Teflon catheter (Terumo, 20G, Tokyo, Japan) was inserted into a vein in the antecubital fossa and a resting blood sample (10 ml) taken. The catheter was flushed with 2 3 ml of 0.9% sterile saline following this and subsequent sampling to ensure patency. Subjects then ingested one of four test meals, which were presented double-blind. Meals were isoenergetic and contained either 2.6 g (kg BM) of a high glycaemic index carbohydrate (glucose polymer Polyjoule, Sustagen Sport powder, chocolate flavouring and skim milk; CHO); 2.6 g (kg BM) of high glycaemic index carbohydrate plus 6 mg (kg BM) caffeine (CHO + CAF): 1.2 g (kg BM) of saturated fat (heavy whipping cream, low energy chocolate flavouring and full cream milk; FAT); or 1.2 g (kg BM) of saturated fat plus 6 mg (kg BM) caffeine (FAT + CAF). Immediately following the ingestion of the test meal, subjects ingested opaque capsules containing either caffeine or a placebo ( 500 mg sucrose). For the FAT and FAT + CAF trials, heparin (2000 U) was administered intravenously 45 min after meal ingestion, while 2 ml of saline was administered in a similar fashion in the other trials. Blood samples were obtained 30 and 55 min following ingestion. Subjects mounted the cycle ergometer 55 min after meal ingestion and undertook a standardized 5 min incremental warm-up before completing 120 min of steady-state cycling at 63% of PPO (SS). Further blood samples (10 ml) were obtained 20, 40, 60, 80 and 100 min after the start of SS cycling, and 1 min before the completion of SS. Expired gas was collected for 60 s after 19, 39, 59, 79 and 119 min. Water (3.3 ml (kg BM) ) was provided at 20 min intervals throughout SS. Immediately upon completion of SS, the ergometer was adjusted into a pedalling rate-dependent (linear) mode. A linear factor was individualized so that at 100 r.p.m. each subject would be cycling at a workload of 82.5% of PPO (85% of ýoµ,max). This workload was chosen as it represents

3 Exp. Physiol Caffeine and exercise metabolism 139 Table 1. Pulmonary gas exchange measurements during the 120 min steady-state cycling Time (min) Treatment ýoµ (l min ) CHO 2.78 ± 0.28 a,b 2.89± ± ± ± ±0.29 CHO+CAF 2.97± ± ± ± ± ±0.44 FAT 2.99± ± ± ± ± ±0.61 FAT+CAF 3.05± ± ± ± ± ±0.43 ýcoµ (l min ) CHO 2.60 ± ± ± ± ± ± 0.27 CHO + CAF 2.76 ± ± ± ± ± ± 0.37 FAT 2.64± ± ± ± ± ±0.40 FAT+CAF 2.72± ± ± ± ± ±0.24 ýe (l min ) CHO 65.8 ± ± 10.2 a 68.3± ± ± ±9.3 CHO+CAF 70.9± ± ± ± ± ±14.8 FAT 67.9 ± ± ± ± ± ± 12.1 FAT+CAF 71.9± ± ± ± ± ±12.9 CHO, 2.6 g kg carbohydrate; CHO + CAF, 2.6 g kg carbohydrate plus 6 mg kg caffeine; FAT, 1.2 g kg fat with 2000 U i.v.heparin; FAT + CAF, 1.2 g kg fat with 2000 U i.v. heparin plus 6 mg kg a caffeine; ýoµ, oxygen consumption; ýcoµ, carbon dioxide production; ýe, ventilation. CHO different to CHO + CAF (P < 0.05); b CHO different to FAT + CAF (P < 0.05). Values are means ± s.d., n =8. the intensity that subjects can sustain for 30 min after a 2 h preload (J. A. Hawley, unpublished observations). After 3 min rest, subjects commenced a performance ride (TT) equivalent to 7 kj kg. Subjects were provided with 500 ml of water for ad libitum consumption throughout the first TT. At the completion of this TT, the volume of water consumed by each subject was measured and this volume was provided during subsequent TTs. The only feedback given to subjects during the TT was the distance covered after each cumulative 10% of the ride. A financial incentive was provided to all subjects to produce the fastest average TT time. A final blood sample was taken 3 min after the completion of each TT. Throughout the SS ride subjective ratings of perceived exertion (RPE) were recorded following each gas collection using the modified Borg scale (Borg, 1975). Heart rates (HR) were recorded throughout all experimental trials using telemetry (Polar Sports Tester, Polar Electro OY, Kempele, Finland). Analytical procedures Instantaneous rates of whole-body carbohydrate and fat oxidation were calculated at 20 min intervals during the SS ride from ýcoµ and ýoµ using the non-protein respiratory exchange ratio (RER) values (Jeukendrup et al. 1996). Total rates of substrate oxidation during the four ² 120 min SS exercise bouts were estimated from the area under the CHO and fat vs. time curves for each subject. Two millilitres of whole blood was placed into a tube containing fluoride EDTA, mixed and spun in a centrifuge at 6800 g for 2 min at 4 C. The plasma was analysed in duplicate within 2 h using an automated glucoseïlactate analyser (2300STAT, Yellow Springs Instruments). Five millilitres of whole blood was placed into a tube containing lithium heparin, mixed and spun, and the plasma was stored at 80 C for later analyses of insulin by radioimmunoassay (Incstar, Stillwater, MN, USA). A further aliquot of blood was mixed in a tube containing lithium heparin and spun in a centrifuge. A 500 ìl volume of plasma was placed in a tube containing 500 ìl of ice-cold 3 Ò perchloric acid, mixed vigorously on a vortex mixer and spun. An 800 ìl volume of this supernatant was added to a tube containing 200 ìl of 6 Ò potassium hydroxide, mixed and spun. The resultant supernatant was analysed for glycerol using enzymatic spectrophotometric analysis (Pinter et al. 1967). Blood (1.5 ml) for determination of FFA concentration was placed in tubes containing EGTA and reduced glutathione and spun in a centrifuge at 4 C for 15min at 2550g. The plasma was then stored at 80 C until analysis. Plasma FFA concentrations were measured by an enzymatic colorimetric method (Wako, NEFAC code , Tokyo, Japan). Statistical analyses Data from the four trials were compared using a two-factor (treatment and time) analysis of variance (ANOVA) with repeated measures. Newman-Keuls post hoc tests (with Bonferroni correction) were conducted when ANOVA revealed a significant interaction. Total carbohydrate and fat oxidation between trials, and mean power and TT performance times were compared using oneway ANOVA with Newman-Keuls post hoc test. Significant differences were accepted when P < All data are reported as means ± s.d.. RESULTS Steady-state metabolic data, energy expenditure and rates of substrate oxidation There was a significant main time effect for ýoµ (P = ) but not for ýcoµ between the four treatments (Table 1). There was a main treatment effect for ýe with CHO + CAF being greater than CHO (75.3 ± 2.3 vs ± 0.9 l min ; P = 0.034). The instantaneous RER values determined throughout SS along with the corresponding rates of carbohydrate and fat oxidation are shown in Table 2, while Fig. 1 displays the mean RER and mean rates of carbohydrate and fat oxidation for the 120 min ride. There was a significant main treatment and time effect for RER, with values declining over time throughout SS

4 140 T. L. Jacobson, M. A. Febbraio, M. J. Arkinstall and J. A. Hawley Exp. Physiol Table 2. Respiratory exchange ratio and rates of carbohydrate and fat oxidation during 120 min steady-state cycling Time (min) Treatment Respiratory exchange CHO 0.94 ± 0.04 a,b 0.93±0.03 a,b 0.93±0.03 a,b 0.92±0.02 a,b 0.91±0.02 a,b 0.90±0.03 a,b * ratio CHO + CAF 0.93 ± 0.03 c,d 0.93±0.03 c,d 0.92±0.03 c,d 0.92±0.03 c,d 0.90±0.02 c,d 0.89±0.02 c,d * FAT 0.89± ± ± ± ± ±0.04* FAT+CAF 0.90± ± ± ± ± ±0.05* Carbohydrate CHO 248 ± 44 a 255±43 b 242±35 a,b 252±38 a,b 240±44 a,b 222±29 a,b oxidation CHO + CAF 259 ± 28 c 243±40 244±22 c,d 244±32 c,d 231±34 c,d 209±21 d (ìmol kg min ) FAT 204 ± ± ± ± ± ± 12 FAT+CAF 224±65 198±54 192±44 189±49 175±61 166±56 Fat oxidation CHO 15 ± 9 a,b 16±8 a,b 17±7 a,b 19±6 a,b 21±8 a,b 24 ± 8 a,b (ìmol kg min ) CHO + CAF 17 ± 9 c,d 19±8 c,d 21 ± 7 c,d 23±7 c,d 27±6 d 28±6 c,d FAT 31±17 31±18 35±20 36±21 37±20 42±18 FAT+CAF 28±18 35±18 35±15 36±16 42±16 45±17 CHO, 2.6 g kg carbohydrate; CHO + CAF, 2.6 g kg carbohydrate plus 6 mg kg caffeine; FAT, 1.2 g kg fat with 2000 U i.v. heparin; FAT + CAF, 1.2 g kg fat with 2000 U i.v. heparin plus 6 mg kg caffeine; a CHO different to FAT (P < 0.05); b CHO different to FAT + CAF (P < 0.05); c CHO + CAF different to FAT (P < 0.05); d CHO + CAF different to FAT + CAF (P < 0.05). *Significantly different to 20 min, P < Values are means ± s.d., n =8. Table 3. Heart rate and ratings of perceived exertion during the 120 min steady-state cycling Time (min) Treatment Heartrate CHO 158±9 158±8 158±9 157±10 b,c 157±12 b,c 161 ± 12* (beats min ) CHO + CAF 154 ± ± ± 12 d,e 157±13 156±16 d,e 158±17 e * FAT 158±8 162±9 162±10 162±12 165±13 164±12* FAT+CAF 161±13 161±13 162±14 163±15 167±16 168±14* Rating of perceived CHO 12.4 ± 1.2 a 13.1 ± 1.5 c 13.3±1.5 b,c 13.2 ± 1.7 b 13.8±1.9 a,b 13.8±2.0 a * exertion CHO + CAF 11.4 ± 0.7 d 11.9±1.4 d,e 12.1 ± 1.7 d 12.2 ± 1.6 d,e 12.3 ± 1.7 d,e 12.5 ± 2.4 d,e * FAT 13.1 ± 1.4 f 13.8±1.3 f 14.1 ± 1.2 f 14.1 ± 1.3 f 14.7 ± 1.7 f 14.4±1.9* FAT+CAF 11.9± ± ± ± ± ±1.9* CHO, 2.6 g kg carbohydrate; CHO + CAF, 2.6 g kg carbohydrate plus 6 mg kg caffeine; FAT, 1.2 g kg fat with 2000 U i.v. heparin; FAT + CAF, 1.2 g kg fat with 2000 U i.v. heparin plus 6 mg kg caffeine; a CHO different to CHO + CAF (P < 0.05); b CHO different to FAT (P < 0.05; c CHO different to FAT + CAF (P < 0.05); d CHO + CAF different to FAT (P < 0.05; e CHO + CAF different to FAT + CAF (P < 0.05); f FAT different to FAT + CAF (P < 0.05); *Significantly different to 20 min, P < Values are means ± s.d., n =8. for all treatments, and being significantly lower after 120 min compared to 20 min (P < 0.05; Table 2). The mean RER throughout SS for both CHO trials was higher than the two fat trials, independent of the co-ingestion of caffeine (CHO, 0.92 ± 0.03 and CHO + CAF, 0.91 ± 0.03 vs. FAT, 0.88 ± 0.05 and CHO + FAT, 0.87 ± 0.05; P < 0.01: Fig. 1A). Accordingly, the mean rate of carbohydrate oxidation was greater (ìmol kg min : CHO, 243 ± 39 and CHO + CAF, 239±30 vs. FAT, 196 ± 48 and FAT + CAF, 191 ± 55, P < 0.05: Fig. 1B) while the mean rate of fat oxidation was lower (ìmol kg min : CHO, 19 ± 8 and CHO + CAF, 22±7vs. FAT, 35 ± 19 and FAT + CAF, 37 ± 17, P < 0.05: Fig. 1C). Blood parameters Plasma glucose concentration (Fig. 2A) was greater 30min after CHO and CHO + CAF ingestion than FAT + CAF (mò: 6.0±1.2 and 5.9±1.6 vs. 4.7±0.7, P < 0.05) but not FAT (4.9±0.5mÒ). However, euglycaemia (> 4.5 mò) was well maintained throughout SS for all treatments. Plasma lactate concentration (Fig. 2B) was 1 mò at rest, and although it was elevated after CHO and CHO + CAF (2.0 ± 0.5 and

5 Exp. Physiol Caffeine and exercise metabolism ±0.7mÒ) compared to FAT and FAT + CAF (1.0 ± 0.4 and 1.2 ± 0.2 mò), these differences failed to reach statistical significance. Plasma lactate concentration was maintained at 2 mò throughout exercise for all treatments. Plasma insulin concentration was greater 30 and 60 min after ingestion of CHO and CHO + CAF than both FAT and FAT + CAF (P < 0.05). At the onset of exercise, plasma insulin concentration declined in all treatments and was similar throughout SS (Fig. 2C). The ingestion of fat (plus i.v. heparin) meals elevated plasma FFA compared to carbohydrate meals, such that 1 h after consumption FFA concentration was greater in FAT and FAT + CAF than both CHO and CHO + CAF (mò: 1.5 ± 0.9 and2.0±1.0 vs. 0.3±0.2and0.3±0.1, P < 0.001; Fig. 3A). At this time FAT + CAF was also greater than FAT (P < 0.05). Throughout SS, plasma FFA concentrations remained greater after FAT and FAT + CAF than CHO and CHO + CAF (mò: 0.9±0.3 and 1.1±0.5 vs. 0.3±0.2 and 0.3±0.2, P < 0.01; Fig. 3A). Despite large differences in plasma FFA concentrations after ingestion of fat compared to carbohydrate meals immediately before exercise (Fig. 3A), there were no Figure 1 Respiratory exchange ratio (RER) and mean rates of carbohydrate and fat oxidation during 120 min of steady-state exercise at 70% of maximal oxygen uptake. CHO, 2 6 g kg carbohydrate; CHO + CAF, 2 6 g kg CHO+6mg kg caffeine; FAT, 1 2 g kg fat with 2000 U i.v. heparin; FAT + CAF, 1 2 g kg fat with 2000 U i.v. heparin + 6 mg kg caffeine. Test meals were ingested 1 h before the start of exercise. Values are means ± s.d. of n =8 subjects. Figure 2 Plasma glucose (A), plasma lactate (B) and plasma insulin (C) concentrations for the four experimental trials. R, rest; 30P, 30 min post-ingestion of test meal. CHO, 2 6 g kg carbohydrate; CHO + CAF, 2 6 g kg CHO + 6 mg kg caffeine; FAT, 1 2 g kg fat with 2000 U i.v. heparin; FAT + CAF, 1 2 g kg fat with 2000 U i.v. heparin + 6 mg kg caffeine. Test meals were ingested 1 h before the start of exercise. Values are means ± s.d. of n =8 subjects. * P<0.05.

6 142 T. L. Jacobson, M. A. Febbraio, M. J. Arkinstall and J. A. Hawley Exp. Physiol differences in plasma glycerol concentrations between the four treatments at this time (Fig. 3B). With the onset of exercise, plasma glycerol concentration increased in all four trials and was greater in FAT and FAT + CAF than both CHO and CHO + CAF at all time points throughout SS (P < 0.02, Fig. 3B). Heart rate response and ratings of perceived exertion HR was not different over time throughout SS, although there were differences between treatments at several time points during the second half of the 2 h exercise bout (Table 3). There was a significant treatment effect (Fig. 4A) with HR being greater in FAT and FAT + CAF than both CHO and CHO + CAF (beats min : 161 ± 10 and 163 ± 14 vs. 157±10 and 156±13, P < 0.05). There was a significant time effect for RPE (P = 0.004; Table 3). RPE values increased from 12.4 ± 1.2, 11.4 ± 0.7, 13.0 ± 1.4 and 11.9 ± 0.6 units after 20 min to 13.8 ± 2, 12.5 ± 2.4, 14.4 ± 1.9 and 14.5 ± 1.9 units after 120 min (Table 3; P < 0.05). There was also a significant treatment ² time interaction (P < 0.004). Mean RPE determined throughout SS was lower when caffeine was co-ingested with either meal (12.1 ± 1.6 vs ± 1.6 for CHO + CAF and CHO, and 12.9 ± 1.3 vs ± 1.0 units for FAT + CAF and FAT, respectively; P <0.05,Fig. 4B) Time trial performance The time taken to complete the time trial was less after carbohydrate than fat ingestion (min: CHO, ± 7.42 and CHO+CAF, 29.12±5.62 vs. FAT, 33.02±8.50 and FAT + CAF, ± 7.70; P < 0.05). However, there was no additive effect on performance when caffeine was co-ingested with carbohydrate compared to when carbohydrate was consumed alone. DISCUSSION The results from the current investigation demonstrate that caffeine co-ingested with carbohydrate 1 h before prolonged, submaximal cycling had little effect on exercise metabolism or performance compared to the ingestion of carbohydrate alone. Taken collectively, these results as well as those reported by others (Costill et al. 1978; Ivy et al. 1979; Essig et al. 1980; Graham & Spriet, 1991) suggest caffeine only elicits a performance benefit when compared to a fasted control trial. Although caffeine has been demonstrated to exert major effects on lipolysis and IMTG utilization (Essig et al. 1980) our data show that in well-trained men, caffeine did not Figure 3 Plasma free fatty acids (FFA, A) and plasma glycerol (B) concentrations for the four experimental trials. R, rest; 30P, 30 min post-ingestion of test meal. CHO, 2 6 g kg carbohydrate; CHO + CAF, 2 6 g kg CHO + 6 mg kg caffeine; FAT, 1 2 g kg fat with 2000 U i.v. heparin; FAT + CAF, 1 2 g kg fat with 2000 U i.v. heparin + 6 mg kg caffeine. Test meals were ingested 1 h before the start of exercise. Values are means ± s.d. of n =8 subjects. 0, P < 0.001; * P <0.01(A) or 0.02 (B). Figure 4 Mean heart rate and ratings of perceived exertion (RPE) throughout 120 min of steady-state exercise at 70% of maximal oxygen uptake. CHO, 2 6 g kg carbohydrate; CHO + CAF, 2 6 g kg CHO + 6 mg kg caffeine; FAT, 1 2 g kg fat with 2000 U i.v. heparin; FAT + CAF, 1 2 g kg fat with 2000 U i.v. heparin + 6 mg kg caffeine. Test meals were ingested 1 h before the start of exercise. Values are means ± s.d. of n = 8 subjects.

7 Exp. Physiol Caffeine and exercise metabolism 143 increase FA oxidation, even when fat availability was optimized via fat ingestion and heparin infusion. To the best of our knowledge, only one previous study has determined the effects of the co-ingestion of caffeine with carbohydrate. In that investigation, the addition of caffeine (3 4 mg kg ) to a carbohydrate electrolyte solution ingested before and throughout exercise improved the performance of a 1 h cycling time trial compared with carbohydrate ingestion alone (Kovacs et al. 1998). Neither metabolic nor subjective (i.e. ratings of perceived exertion) measures were taken in that study (Kovacs et al. 1998). We have previously reported that during a1htimetrial glycogen availability is not limiting for performance (Hawley et al. 1997). Therefore, it seems unlikely that caffeine would have lead to a performance enhancement in the study of Kovacs et al. (1998) by enhancing substrate availability. Taken collectively, the data of Kovacs et al. (1998) and the results of the present study strongly suggest that caffeine does not affect substrate metabolism when co-ingested with carbohydrate. Importantly, in the present study, the ratings of perception of effort during the submaximal ride were reduced with the co-ingestion of caffeine with carbohydrate, compared with carbohydrate intake alone. This finding is consistent with previous studies that have examined the effect of caffeine ingestion on subjective measures (Costill et al. 1978; Cole et al. 1996). It is possible that the subjects involved in the study of Kovacs et al. (1998) also felt better after caffeine ingestion. Although speculative, it may be that subjective feelings are quantitatively more important during shorter more intense exercise bouts than during longer, moderate intensity exercise. It must be noted, however, that the specific mechanisms for the reduction in subjective perception of effort are not well understood. The reason for the slower time trial performance when subjects were fed fat compared to carbohydrate is difficult to explain, especially when a significant amount of endogenous carbohydrate was spared with fat feeding. We have previously reported that when plasma FFA concentration was elevated almost fourfold by fat feedings, there was a reduction in carbohydrate oxidation during 30 min of intense (80% of ýoµ,max) exercise, but no subsequent performance benefit to a 30 min time trial when compared to carbohydrate ingestion (Hawley et al. 2000). It is difficult to estimate the magnitude of a glycogen sparing effect required to result in a meaningful performance change. However, using the original data of Spriet et al. (1992) it can be calculated that it would take at least a 20% reduction in glycogenolysis to increase sustainable power output during cycling at 80% of ýoµ,max by 1.5% (see Hopkins et al. 1999). Based on previous data (Essig et al. 1980), we hypothesized that caffeine would increase fat oxidation (i.e. IMTG oxidation), even when fat availability was maximized by fat ingestion and heparin administration. Essig et al. (1980) reported caffeine (5 mg kg ) ingested 1h before 30min of cycling resulted in a 150% increase in IMTG utilization and a 42% decrease in muscle glycogenolysis compared to a control trial. The ingestion of fat and intravenous heparin was effective in markedly elevating both plasma FFA and glycerol concentration compared with carbohydrate ingestion (Fig. 3). Although plasma FFA concentrations were consistently higher after FAT + CAF than FAT ingestion (Fig. 3), these values were only statistically different immediately before exercise. Hence caffeine had little effect on augmenting fat oxidation in a situation where lipolysis was already high. More importantly, however, was the observation that caffeine had no effect on increasing fat oxidation when fat availability was maximized because both RER values (Fig. 1A) and the mean rates of fat oxidation (Fig. 1B and C) were remarkably similar when comparing the two fat ingestion trials. Therefore, we conclude that in well-trained men, caffeine does not increase fat oxidation (via an increase in IMTG use) when plasma fat availability is already maximized. Apart from influencing rates of perceived exertion, the only other effects of caffeine were to elevate ýe (Table 1) and heart rate (Fig. 4). Spriet et al. (1992) have previously reported that ýe during intense (80% of ýoµ,max) cycling was increased 13% after the ingestion of caffeine (9 mg kg ) compared to placebo; in the current study we found a more modest (8%) rise during moderate intensity exercise. There is general agreement that caffeine ingestion increases exercise heart rate because of its powerful effect on the sympathetic nervous system (Spriet et al. 1992; Jackman et al. 1996). However, these small physiological changes were not important in terms of an ergogenic performance benefit. In conclusion, the results of this study show that when caffeine was co-ingested with either a carbohydrate or fat meal, there were no additive effects on circulating blood metabolites, nor rates of substrate oxidation during a subsequent bout of prolonged, moderate intensity cycling. Compared with carbohydrate ingestion, consumption of a high fat meal resulted in a significant reduction in carbohydrate utilization during exercise. However, this sparing of endogenous carbohydrate did not improve subsequent time trial performance. Indeed, when sufficient (2.6 g kg ) carbohydrate was ingested before exercise, caffeine had no additional performance benefit. Angus, D. J., Hargreaves, M., Dancey, J. & Febbraio, M. A. (2000). Effect of carbohydrate or carbohydrate plus medium-chain triglyceride ingestion on cycling time trial performance. Journal of Applied Physiology 88, Borg, G. (1975). Simple rating method for estimation of perceived exertion. In Physical Work and Effort, ed. Borg, G., pp Pergamon, New York. Coggan, A. R. & Coyle, E. F. (1991). Carbohydrate ingestion during prolonged exercise: effects on metabolism and performance. In Exercise and Sport Sciences Reviews, ed. Holloszy, J. O., pp Williams and Wilkins, Baltimore. Cole, K. J., Costill, D. L., Starling, R. D., Goodpaster, B. H., Trappe, S. W. & Fink, W. J. (1996). Effect of caffeine on perception of effort and subsequent work production. International Journal of Sports Nutrition 6,

8 144 T. L. Jacobson, M. A. Febbraio, M. J. Arkinstall and J. A. Hawley Exp. Physiol Costill, D. L., Dalsky, G. P. & Fink, W. J. (1978). Effects of caffeine ingestion on metabolism and exercise performance. Medicine and Science in Sports 10, Coyle, E. F., Coggan, A. R., Hemmert, M. K. & Ivy, J. L. (1986). Muscle glycogen utilisation utilization during prolonged strenuous exercise when fed carbohydrate. Journal of Applied Physiology 61, Erickson, M. A., Schwarzkopf, R. J. & McKenzie, R. D. (1987). Effects of caffeine, fructose, and glucose ingestion on muscle glycogen utilization during exercise. Medicine and Science in Sports and Exercise 19, Essig, D., Costill, D. L. & Van Handel, P. J. (1980). Effects of caffeine ingestion on utilisation of muscle glycogen and lipid during leg ergometer exercise. International Journal of Sports Medicine 1, Flynn, M. G., Costill, D. L., Hawley, J. A., Fink, W. J., Neufer, P. D., Fielding, R. A. & Sleeper, M. D. (1987). Influence of selected carbohydrate drinks on cycling performance and glycogen use. Medicine and Science in Sports and Exercise 19, Graham, T. E. & Spriet, L. L. (1991). Performance and metabolic responses to a high caffeine dose during prolonged exercise. Journal of Applied Physiology 71, Graham, T. E. & Spriet, L. L. (1995). Metabolic, catecholamine, and exercise performance responses to various doses of caffeine. Journal of Applied Physiology 78, Hawley, J. A. & Burke, L. M. (1997). Effect of meal frequency and timing on physical performance. British Journal of Nutrition 77, S Hawley, J. A., Burke, L. M., Angus, D. J., Fallon, K. E., Martin, D. T. & Febbraio, M. A. (2000). Effect of altering substrate availability on metabolism and performance during intense exercise. British Journal of Nutrition 84, Hawley, J. A. & Noakes, T. D. (1992). Peak sustained power output predicts VO2max and performance time in trained cyclists. European Journal of Applied Physiology and Occupational Physiology 65, Hawley, J. A., Palmer, G. S. & Noakes, T. D. (1997). Effect of 3 days of carbohydrate supplementation on muscle glycogen content and utilisation during a 1-h cycling performance. European Journal of Applied Physiology 75, Hopkins, W. G., Burke, L. M. & Hawley, J. A. (1999). Design and analysis of research on sport performance enhancement. Medicine and Science in Sports and Exercise 31, Horowitz, J. F., Mora-Rodriguez, R., Byerley, L. O. & Coyle, E. F. (1997). Lipolytic suppression following carbohydrate ingestion limits fat oxidation during exercise. American Journal of Physiology 273, E Horowitz, J. F., Mora-Rodriguez, R., Byerley, L. O. & Coyle, E. F. (2000). Preexercise medium-chain triglyceride ingestion does not alter muscle glycogen use during exercise. Journal of Applied Physiology 88, Ivy, J. L., Costill, D. L., Fink, W. J. & Lower, R. W. (1979). Influence of caffeine and carbohydrate feedings on endurance performance. Medicine and Science in Sports 11, Jackman, M., Wendling, P., Friars, D. & Graham, T. E (1996). Metabolic, catecholamine, and endurance responses to caffeine ingestion during intense exercise. Journal of Applied Physiology 81, Jeukendrup, A. E., Saris, W. H. M., Van Diesen, R., Brouns, F. & Wagenmakers, A. J. M. (1996). Effect of endogenous carbohydrate availability on oral medium-chain triglyceride oxidation during prolonged exercise. Journal of Applied Physiology 80, Jeukendrup, A. E., Thielen, J. J. H. C., Wagenmakers, A. J. M., Brouns, F. & Saris, W. H. M. (1998). Effect of medium-chain triacylglycerol and carbohydrate ingestion during exercise on substrate utilization and subsequent cycling performance. American Journal of Clinical Nutrition 67, Kalmar, J. M. & Cafarelli, E. (1999). Effects of caffeine on neuromuscular function. Journal of Applied Physiology 87, Kovacs, E. M. R., Stegen, J. H. C. H. & Brouns, F. (1998). Effect of caffeinated drinks on substrate metabolism, caffeine excretion, and performance. Journal of Applied Physiology 85, McConell, G., Fabris, S., Proietto, J. & Hargreaves, M. (1994). Effect of carbohydrate ingestion on glucose kinetics during exercise. Journal of Applied Physiology 77, Okano, G., Sato, Y. & Murata, Y. (1998). Effect of elevated blood FFA levels on endurance performance after a single fat meal ingestion. Medicine and Science in Sports and Exercise 30, Okano, G., Sato, Y., Takumi, Y. & Sugawara, M. (1996). Effect of 4h pre-exercise high carbohydrate and high fat meal ingestion on endurance performance and metabolism. International Journal of Sports Medicine 17, Pasman, W. J., van Baak, M. A., Jeukendrup, A. E. & de Haan, A. (1995). The effect of different dosages of caffeine on endurance performance time. International Journal of Sports Medicine 16, Pinter, J. K., Hayashi, J. A. & Watson, J. A. (1967). Enzymatic assay of glycerol, dihydroxyacetone and glyceraldehyde. Archives of Biochemistry and Biophysics 121, 404. Pitsiladis, Y. P., Smith, I. & Maughan, R. J. (1999). Increased fat availability enhances the capacity of trained individuals to perform prolonged exercise. Medicine and Science in Sports and Exercise 31, Spriet, L. L., Maclean, D. A., Dyck, D. J., Hultman, E., Cederblad, G. & Graham, T. E. (1992). Caffeine ingestion and muscle metabolism during prolonged exercise in humans. American Journal of Physiology 262, E Tsintzas, O. K. & Williams, C. (1998). Human muscle glycogen metabolism during exercise: effect of carbohydrate supplementation. Sports Medicine 25, Tsintzas, O. K., Williams, C., Boobis, L. & Greenhaff, P. (1995). Carbohydrate ingestion and glycogen utilization on different muscle fibre types in man. Journal of Physiology 489, Van Zyl, C. G., Lambert, E. V., Hawley, J. A., Noakes, T. D. & Dennis, S. C. (1996). Effects of medium chain triglyceride ingestion on fuel metabolism and cycling performance. Journal of Applied Physiology 80, Waldeck, B. (1973). Sensitisation by caffeine of central catecholamine receptors. Journal of Neural Transmission 34, Whitley, H. A., Humphreys, S. M., Cambpell, I. T., Keegan, M. A., Jayanetti, T. D., Sperry, D. A., MacLaren, D. P., Reilly, T. & Frayn, K. N. (1998). Metabolic and performance responses during endurance exercise after high-fat and high-carbohydrate meals. Journal of Applied Physiology 85, Acknowledgements We would like to thank Damien J. Angus, Dr Louise Burke, Vasili Nikolopoulos and Nigel K. Stepto for excellent technical assistance. We also thank the subjects for their commitment and co-operation during this study.