The effect of caffeine ingestion on skill maintenance and fatiguein epee fencers

LINDSAY BOTTOMS1, ANDREW GREENHALGH2, & KIM GREGORY3,4

1University of East London, Health and Bioscience, Water Lane, Stratford, E15 4LZ United Kingdom, 2University ofHertfordshire, School of Life Sciences, Hatfield, United Kingdom, 3University of Bath, Sport and Exercise Science, Bath,United Kingdom, and 4West Midlands Deanery, Sport and Exercise Medicine, Birmingham, United Kingdom

(Accepted 1 January 2013)

AbstractThe ergogenic effect of caffeine on sports performance focuses predominantly on endurance sports (Doherty & Smith,2004) with little research on intermittent high intensity sports. This study aimed to explore the effect of caffeine ingestion onskill maintenance following fencing simulated exercise. Eleven competitive fencers participated (four female; seven male;age 33 ± 6.5 years). Following a maximal test to exhaustion, fencers completed two trials assessing accuracy and reactiontimes (Stroop test) before and after a fatiguing protocol designed to simulate the demands of a fencing competition. Skilltesting involved 30 lunges to hit a target. 500 ml placebo or 3 mg · kg−1 caffeine supplemented drink was administered afterthe initial reaction and skill tests in a single-blind crossover design. The fatiguing protocol involved simulating six fights with6-minute rests between each. Fencers rated their perceived exertion (arm, legs, overall) using the Borg scale. There was nooverall effect of caffeine on total skill score (P = 0.40), however there was a tendency for fewer misses with caffeine(P = 0.10). Caffeine had no effect on the Stroop Test. Caffeine produced significantly lower perceived fatigue for overall(P < 0.01). These results provide some support for caffeine producing maintenance of skill and reducing perceived fatigueduring fencing.

Keywords: caffeine, fencing, skill, fatigue

Introduction

Caffeinewas removed from theworld anti-doping asso-ciation (WADA) banned list in 2004 (WADA, 2012)but remains on the monitoring list to detect patterns ofmisuse in sport. Consequently interest in the ergogeniceffects of caffeine on sporting performance hasincreased.The effects of sports drinks on exercise capa-city, performance and recovery have been well docu-mented (Tarnopolsky & Cupido, 2000; Cureton &Warren, 2007). Caffeine has generally not been addedto sports drinks because of a purported diuretic effect(Cureton & Warren, 2007), which could potentiallyhave detrimental thermoregulatory effects in sports(ACSM [American College of Sports Medicine],2009; Cureton & Warren, 2007; Pipe & Ayotte,2002). However, contrary to popular beliefs, a reviewof literature by Armstrong, Casa, Maresh, & Ganio(2007) has shown that caffeine consumption does notresult in water-electrolyte imbalances or hyperthermia.

Fencing is an open-skilled combat sport (Roi &Bianchedi, 2008). Fencing involves frequent short

high intensity bursts of exercise, lunging towards anopponent during an attack; followed by periods oflower intensity exercise, bouncing preparatory move-ments (Paul, Miller, Bottoms, & Beasley, 2011). Acompetition can last between 9 and 11 hours but thisconsists of an effective fight time between 17 and 48minutes. The distance covered by fencers has beenreported to be between 250–1000 metres (Roi &Bianchedi, 2008). A competition consists of a firstround where each fencer has between 5–6 fights upto five hits with a maximum duration of 3 minutesper fight. The fight ends when a fencer reaches fivehits, or when 3 minutes of fencing time has beenreached. If the scores are even after 3 minutes aminute of extra time is given. Following the firstround a series of direct elimination fights is under-taken which consists of 15 hits and each fight can lastup to 9 minutes with 2 × 1 minute breaks. If theyhave not achieved 15 hits and the score is even after9 minutes of fencing time, an extra minute is givento decide the winner. The eventual winner of a

Correspondence: Lindsay Bottoms, University of East London, Health and Bioscience, Water Lane, Stratford, E15 4LZ United Kingdom. E-mail:l.bottoms@uel.ac.uk

Journal of Sports Sciences, 2013Vol. 31, No. 10, 1091–1099, http://dx.doi.org/10.1080/02640414.2013.764466

© 2013 Taylor & Francis

competition will undertake about seven of thesedirect elimination fights. A key factor in performancesuccess is the fencer’s reaction time in response totheir opponent’s actions (Paul et al., 2011).Although there is a general paucity of research infencing, studies have identified that an adequatepsychological condition is required to prevent cen-tral and peripheral fatigue (Daya, Donne, &O’Brien, 2002; Roi & Bianchedi, 2008). Fatigue isone explanation for overt performance reductionsduring competitive sport (Noakes, 2000).Therefore, reducing the negative effect of fatigue,could potentially improve fencers’ responses andthus performance.

There is a growing body of evidence suggestingthat caffeine has an ergogenic effect through bothperipheral and central mechanisms (ACSM, 2009;Cureton & Warren, 2007; Del Coso, Estevez, &Mora-Rodriguez, 2008; Doherty & Smith, 2004;Fredholm, Battig, Homen, Nehlig & Zvartaue,1999; Glastier et al., 2008; Kalmar, 2005; Lopes,Aubier, Jardim, Aranda, & Macklem, 1983; Pipe &Ayotte, 2002; Plaskett & Cafarelli, 2001; Spriet,1995; Tarnopolsky & Cupido, 2000). Caffeine hasbeen demonstrated to have a direct action on skeletalmuscle (Tarnopolsky & Cupido, 2000) during lowstimulation frequency replicating endurance exer-cise, this could potentially lead to lower heart rateat a given workload. This action on the skeletalmuscle demonstrates a peripheral action of caffeine,whereas it has also been shown to act through centralmechanisms such as on the basal ganglia (Spriet,1995) which improves motivation and cognitivefunction and reduces perception of fatigue. Thesecentral and peripheral mechanisms of action of caf-feine may potentially benefit fencing performance byimproving repeated high intensity exercise (DelCoso et al., 2012) and improving reaction time andfinally reducing fatigue consequently preventing theassociated deterioration in performance. However,caffeine can be ergolytic in high doses, producingtremor, mental agitation and tachycardia, all ofwhich could negatively impact performance(Cureton & Warren, 2007).

Evidence from previous research investigating ten-nis and cycling identified physiological and cognitivefatigue characteristics which were demonstrated tonegatively influence skill-performance (Davey,Thorpe, & Williams, 2002; Hornery, Farrow,Mujika, & Young, 2007; Murray, Cook, Werner,Schlege, & Hawkins, 2001; Royal et al., 2006;Vergauwen, Spaepen, Lefevre, & Hespel 1998).Fencing performance which encompasses both phy-siological and cognitive demands may therefore besignificantly influenced by the effects of fatigue.Thus ergogenic aids such as caffeine may have abeneficial effect on fencing performance.

This study aims to investigate the effect of caffeineon skill maintenance, cognitive function, perceivedfatigue and physiological response following an exer-cise protocol developed to reflect the intensity of afencing competition. It was hypothesised that caf-feine ingestion would attenuate changes in skillwithin fencers and improve cognitive function.

Methods

Participants

Eleven healthy adult epee fencers (four female; sevenmale; 33 ± 6.5 years; body mass 76.3 ± 14.3 kg;height 1.77 ± 0.09 m) who were all regularly com-peting in UK national level competitions and hadbeen fencing for over two years (mean ± s11.5 ± 6.0 years) volunteered to participate in thestudy. All participants were given written informa-tion concerning the nature and purpose of the study,completed a pre-participation medical screeningquestionnaire and gave written consent prior to par-ticipation. A post hoc statistical power analysis wasconducted using the Hopkins method, and it wasfound that the sample size was sufficient to providemore than 80% statistical power. University EthicsCommittee approval for the study’s experimentalprocedures was obtained and followed the principlesoutlined in the Declaration of Helsinki.

Procedure

The current investigation incorporated a single blindrandomised cross over design; with all participantsperforming a maximal oxygen uptake test (VO2max)performed on a treadmill prior to data collection.This allowed for identification of the participantsVO2max and their maximum heart rate in order todetermine relative intensity during the experimentaltrials. The participants then completed two experi-mental trials separated by a week, where they eitherconsumed 500 ml of sugar free fruit juice with 3 mg ·kg · bw−1 of caffeine or 500 ml of the same fruit juice(Placebo).

Preliminary trial

The VO2max test was performed at least seven daysprior to the experimental trials. The exercise test wascarried out on a treadmill (H/P/Cosmos, Germany)using an initial running speed of 10 km · h−1 with a1% gradient. Speed increments of 1 km · h−1 weremade every two minutes for the first six minutesand thereafter the gradient increased by 2% everyminute until volitional fatigue (Bottoms, Hunter, &Galloway, 2006; Bottoms, Taylor, Sinclair, Polman,& Fewtrel, 2012). Heart rate (Polar RS800, Finland)

1092 L. Bottoms et al.

and gas analysis measurements (Cosmed K4, UK)were monitored throughout the test period.Following the test the participants undertook afamiliarisation of the experimental procedures.

Experimental trials

Each participant completed two trials each involvinga reaction time test and a fencing specific skills testboth pre and post a fatiguing protocol (Figure 1). Alltesting was carried out on a shock absorbing sportssurface with fencers using their own protectiveequipment and epees. The two trial sessions wereconducted one week apart. Participants were askedto abstain from caffeine, alcohol and heavy trainingfor 48 hours prior to each trial and also to replicatethe same diet in these time periods.

Immediately prior to each trial session body mass(Salter, 9141WH3R, UK), blood pressure (Omron,M4) and resting heart rates were recorded. Baselineblood glucose (Accu-Chek, Aviva, UK) and lactate(Lactate Pro, Bodycare, UK) levels were obtainedbefore participants completed a 15 minute individualwarm up routine as they generally would within afencing competition.

Following the warm up a reaction time test andskill test lasting approximately five minutes, werecompleted before ingesting the intervention drink.A 30-minute rest period was then undertaken bythe participant followed by a fatiguing protocol.The fatiguing protocol consisted of a series of sixsimulated fights representing the first round of afencing competition with duration of 60 minutes.One litre of fruit juice (125 ml fruit juice: 875 mlwater) was ingested by the participants during thefatiguing protocol to simulate fluid ingestion during acompetition. The reaction and skill tests were per-formed after completion of the fatiguing protocol.Ratings of perceived exertion were recorded afterevery fight for the sword arm (RPEarm), legs(RPElegs) and total exertion (RPEoverall) using theBorg scale. Blood lactate and glucose measures were

obtained immediately after completion of the fatiguingprotocol and again at the end of the study. The experi-mental protocol is represented in Figure 1.

Reaction test

The participants performed an online version of theStroop test (http://www.onlinestrooptest.com) whichdisplayed 20 word-colour combinations (a mixtureof congruent and incongruent). The total responsetime for congruent and incongruent correct answerswas recorded.

Skill test

The skill test used within the study focused on thecentral chest as a target. A 5 cm × 5 cm targetgraduated in 1 cm divisions was placed centrally onthe fencing dummy’s chest. The fencer lunged to hitthe target replicating a fencing competitive attack.The position of their front foot was marked fromeach participants ideal lunge distance (i.e. they per-formed a correct lunge technique and did not over orunder extend during the lunge) and they returned tothis position after each lunge. A cumulative scorefor each successful hit was recorded (3 if central, 2if the point hit the inner ring, 1 if the outer ring, 0 iftarget missed) from 30 successive hits. During theskill tests, participants wore shorts and t-shirt whilstalso wearing their own glove, mask and fencingfootwear.

Fatiguing protocol

The fatiguing protocol simulated six poule fightsfrom a previously developed lab based fencing spe-cific protocol (Bottoms, Rome, Gregory, & Price,2009) with a work to rest ratio during a fight of1:0.8 which is comparable to previous publisheddata (Roi & Bianchedi, 2008).

The fatiguing protocol was designed to replicatecompetition fencing to the completion of the first

Figure 1. A schematic diagram representing the main study protocol.

Effect of caffeine on epee fencing performance 1093

round of poules. The participants started by facingthe static fencing dummy opponent in the on-guardposition. A series of bouncing movements (9 sec-onds on 8 seconds off) with a standardised numberof retreats, arm extensions and lunges whilst boun-cing were then performed to replicate competitionfighting attacks. An example of a fight is illustrated inTable I. A six-minute rest was timed between eachfight to reproduce the maximum rest period withinofficial competition.

Heart rate data was recorded throughout eachsimulated fight to determine minimum, maximumand mean heart rates. Ratings of perceived exertionwere recorded immediately after each simulated fight.

Physiological data

Body mass was recorded pre- and post-trial for eachcondition to calculate sweat loss. Blood glucose andlactate measurements were taken via capillary sam-pling (20 μl) at the finger at the start of the trial,immediately prior to starting the fatiguing protocol,on completion of the fatiguing protocol and at theend of the study for both trials.

Data analysis

The Shapiro-Wilk statistic confirmed that the normaldistribution assumption was met for all variables.Therefore data were analysed using a repeatedmeasures two-way (Treatment X Time) analysis ofvariance (ANOVA; SPSS v20). Appropriate posthoc analyses were conducted using a Bonferroni cor-rection to control for type I error. Partial effect sizeswere calculated using an η2. Data are presented asmean ± standard deviation in tables and figures.Significance was set at P < 0.05.

Results

Participant characteristics and physiological datafrom the preliminary testing are presented inTable II.

The participants were unaware whether they wereingesting caffeine or placebo. No significant differ-ences were observed in the baseline values of anyoutcome measures between trials.

Blood glucose and lactate

Blood glucose concentrations remained constantthroughout the protocol as there was no effect oftime (P = 0.21, η2 = 0.14; Figure 2). There wasalso no effect of caffeine on blood glucose as therewas no significant difference between trials(P = 0.96, η2 = 0.00) with mean concentrations of5.2 (± 1.3) and 5.2 (± 0.8) mmol · l−1 for placeboand caffeine, respectively. There was a significantmain effect for time for blood lactate (P < 0.01,η2 = 0.54; Figure 2). Blood lactate increased fromrest (1.3 ± 1.2 and 1.4 ± 1.0 mmol · l−1 for placeboand caffeine, respectively) to the end of the fatiguingexercise (3.3 ± 1.9 and 3.2 ± 1.6 mmol · l−1 forplacebo and caffeine, respectively). However, caf-feine had no effect on blood lactate as there wereno differences between trials observed (P = 0.99,η2 = 0.14).

Skill test

There was no effect of time on the total skill testscores as can be seen from Figure 3 (P = 0.69,η2 = 0.04), therefore fatigue had no effect on skillperformance. There was also no effect of caffeine onthe skill test (P = 0.40, η2 = 0.13). However,

Table I. Example of a poule fight within the study fatiguing protocol.

On (s) Off (s) No. of Lunges No. of Retreats No. of arm extensions

9 8 1 0 29 8 1 0 29 8 1 1 29 8 1 0 29 8 1 0 39 8 1 0 29 8 1 1 2

Table II. Characteristics of the participants (n = 11).

Age (years) Height (m) Weight (kg)Fencing

Experience (years)VO2max

(ml−1 · min−1 · kg−1)

MaximumHeart Rate

(beats · min−1)

Maximumblood lactate(mmol · l−1)

Mean (± s) 33.0 1.77 76.3 11.5 48.3 200 11.6(± 6.5) (± 0.09) (± 14.3) (± 6.0) (± 6.8) (± 6) (± 0.9)

1094 L. Bottoms et al.

comparing the number of misses between trials therewas a tendency for caffeine to reduce the number ofmisses (P = 0.10, η2 = 0.31). During the placebo trialthe number of ‘0’scores (indicating the fencer missedthe target) after the fatiguing protocol was 7 ± 4 com-pared to 3 ± 3 during the caffeine trial. There were nodifferences in the number of 3s, 2s and 1s scoredbetween trials (P > 0.05; Figure 4). There was alsono difference in pre skill test scores on both trialsindicating the skills test was reliable (P = 0.78).

Reaction time test

Caffeine had no effect on the reaction time test as therewere no significant differences demonstrated betweencaffeine and placebo for total response times through-out this study in relation to congruent conditions

(P = 0.48, η2 = 0.14) or incongruent conditions(P = 0.54, η2 = 0.17). Figure 5 illustrates the totalresponse times for both congruent (e.g. the word bluewritten in blue ink) and incongruent (e.g. the word redwritten in black) questions of the Stroop test.

Body mass

Mean sweat losses were 0.57 ± 0.45 kg during PLand 0.58 ± 0.49 kg during the caffeine trial, with nosignificant difference between trials (P = 0.89).

Exercise intensity

As can be seen from Table III, heart rate was sig-nificantly lower during the caffeine trial compared tothe placebo (P = 0.05, η2 = 0.34). This indicated

Figure 3. Mean (± s) total skill test score pre and post fatiguing exercise for both PL and CAF.

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

Resting Pre protocol Post protocol Final

Blo

od

le

ve

l (m

mo

l ·

l−1)

Time

Placebo-glucose Caffeine-glucose

Placebo-lactate Caffeine-lactate

Figure 2. Mean (± s) blood glucose and lactate concentrations during exercise for both placebo (PL) and caffeine (CAF).

Effect of caffeine on epee fencing performance 1095

that athletes were on average exercising at an inten-sity of 72 ± 6 and 67 ± 8% of their maximum heartrate for placebo and caffeine, respectively. In addi-tion overall ratings of perceived exertion were lowerduring the caffeine trial compared to the placebo

trial (P = 0.03, η2 = 0.37; Table III). There was amain effect for time in all ratings of perceived exer-tion values with post hoc demonstrating an increasein ratings of perceived exertion as the fight numberincreased (P < 0.01, η2 = 0.56). No effect of caffeinewas observed for ratings of perceived exertion at thearm (P = 0.16, η2 = 0.21) or legs (P = 0.35,η2 = 0.09).

Discussion

The aim of this study was to examine the effects3 mg · kg−1 of caffeine on skill and reaction timeperformance, and on perceived fatigue associated

0

2

4

6

8

10

12

14

16

Pre Placebo Post Placebo Pre Caffiene Post Caffiene

Nu

mb

er

of h

its s

co

rin

g 3

,2,1

or 0

Study timepoint in relation to fatiguing protocol

3

2

1

0

*

Figure 4. Mean (± s) number of 3s, 2s, 1s and 0s scored pre and post fatiguing exercise for both PL and CAF. *denotes significantdifference from pre CAF.

0.00

2.00

4.00

6.00

8.00

10.00

12.00

14.00

16.00

18.00

PLACEBO Pre PLACEBO Post CAFFEINE Pre CAFFEINE Post

To

ta

l re

sp

on

se

tim

e (

s)

Time

incongruent time (secs)

congruent time (secs)

Figure 5. Mean (±SD) congruent and incongruent response times to the Stroop test for both PL and CAF.

Table III. Mean (± s) RPE (Borg Scale) and heart rate during allfights for PL and CAF trials for arm, legs and overall.

RPEarm RPElegs RPEoverall

Heart Rate(beats · min−1)

PL 11 (± 2) 13 (± 3) 13 (± 2) 143 (± 15)CAF 10 (± 2) 12 (± 3) 11 (± 2)* 133 (± 17)

*denotes significant difference from PL.

1096 L. Bottoms et al.

with epee fencing. This dose of caffeine reducedperceived fatigue during fencing and also there wasan observed tendency for maintenance of skill fol-lowing simulated fencing, however there was no ben-eficial effect on reaction time.

Epee fencing is a sport, which requires accuratebody movements performed quickly in order todefeat the opponent. Measuring accuracy withinthe fencing skill paradigm is challenging given inepee the whole body is a valid target. The presentstudy demonstrated that caffeine has a tendency toimprove skill maintenance by reducing the numberof non-scoring lunges. This finding is supported byprevious research in other skill based sports, whichhas demonstrated caffeine increases passing accuracyin football (Foskett, Ali, & Gant, 2009) and rugby(Roberts et al., 2010; Stuart, Hopkins, Cook, &Cairns, 2005) and stroke accuracy in tennis(Hornery et al., 2007). Previous literature hasshown that caffeine as low as 3 mg · kg−1 is sufficientto produce an ergogenic effect on performance(Graham & Spriet, 1995), however higher doseshave been shown to produce significantly greaterenhancements (Doherty & Smith, 2004).Therefore, it is possible that greater changes in accu-racy would have been seen if a larger dose of caffeine(e.g. 6 mg · kg−1) was ingested. This is an area forfurther investigation.

This study’s reaction time protocol used a com-puter Stroop test program as a measure of reactiontimes and thus cognitive function. The centralmechanism of caffeine has been shown to improvecognitive function such as memory and reaction time(Jacobson & Edgley, 1987). The results obtained inthe present study did not show an effect of caffeineon reaction times within the Stroop test, in contrastto other studies demonstrating reduced interferencein response to caffeine ingestion (Kenemans,Wieleman, Zeegers, & Verbaten, 1999). Fatiguedid not negatively affect the reaction time test resultsand this may be the reason why no difference wasobserved with caffeine ingestion.

The movements within fencing are complexrequiring coordination of upper and lower limbs inresponse to information given by their opponent towhich the fencer then has to react. Studies havereported reaction times (some also commenting onaccuracy) to protocols involving isolated touché, iso-lated lunge and sequential touché and lunge inresponse to light triggers (Harmenberg, Ceci,Barvestad, Hjerpe, & Nystrom, 1991; Roi &Bianchedi, 2008; Williams & Warmsley, 2000;Yiou & Do, 2000). These have been able to discri-minate between the advanced and novice fencers.This reflects prior knowledge that elite fencers initi-ate a lunge by arm extension rather than foot move-ment and hit their target before the lead foot strikes

the ground (Harmenberg et al., 1991; Roi &Bianchedi, 2008). However, as all the fencers in thecurrent investigation were experienced fencers, ageneric reaction time test was deemed to be suffi-cient to determine effects of fatigue and thus subse-quent effects of caffeine. Visual information alsoplays a role in fencing success as fencers respond tomovements of opponents’ weapons and postures toanticipate feints and counterfeits during bouts, againsupporting the concept of a generic visual reactiontime test such as the Stroop test. Rapid reactiontimes to stimuli are critical for fencing performance.However, as no differences were observed eitherbefore and after exercise or between trials, it maybe more beneficial to undertake a fencing specificreaction time test. In future the authors recommenda sports specific reaction time test should beimplemented.

This study demonstrated a significant effect oftime on ratings of perceived exertion and caffeinesignificantly reduced overall body ratings. This pro-vides support for the fatiguing protocol used in thestudy (Bottoms et al., 2009). Heart rate is depen-dent on intensity of a fencing bout and previousstudies (Roi & Bianchedi, 2008) have indicatedthat heart rates during competition can reach 70%of maximum heart rate which was achieved in thepresent study. Caffeine acts as a stimulant and mayheighten cardiovascular responses; well-known sideeffects include palpitations if subjects are sensitivereflecting variation in caffeine metabolism. Share,Sanders, and Kemp (2009) reported no effect onheart rate however; in this study caffeine signifi-cantly reduced heart rates during the fatiguing pro-tocol. It may be that in reducing ratings ofperceived exertion caffeine also influenced cardio-vascular parameters either through reduced painperception or induced metabolic changes. As pre-viously mentioned, the peripheral actions of caf-feine have been demonstrated to potentiate lowfrequency skeletal muscle force (Tarnopolsky &Cupido, 2000), the lower heart rate could be aresult of this peripheral action. Lower stimulationof the muscle could be required for the same exer-cise output, consequently reducing heart rate. Thisis an area for further research, but could be impor-tant for fencing performance.

The central mechanisms for caffeine as an ergo-genic aid to sport arise, in that the observed reducedoverall body ratings of perceived exertion may con-tribute to enhanced performance despite rising lac-tate levels. Blunting of pain combined with perceivedreduced effort may enable prolonged duration ofexercise resulting in increased lactate level (Davis &Green, 2009). Furthermore ergogenic effects in rela-tion to substrate availability, in particular glycogen,are less limiting to performance at this exercise

Effect of caffeine on epee fencing performance 1097

intensity and duration (Bottoms et al., 2006; Davis& Green, 2009; Roberts et al., 2010).

Conclusion

The present study provides further limited supportfor caffeine demonstrating a positive effect on skillmaintenance and reducing perceived levels of fatiguein sport characterised by intermittent high intensityexercise. A reduction in heart rate, potentially aresult of a peripheral action of caffeine, appears tobe supported from the results as a primary mechan-ism. Sport-specific research both in and out of com-petition is needed to increase applicability of results.The scope for using fencing as a sport in which tofurther this area of research is vast given the strongassociation between the physical, perceptual andpsychological demands of the sport and both centraland peripheral fatigue.

Acknowledgements

The authors would like to thank all the fencers whotook part as well as Andy Smith and Anthony Tillsfor helping with some of the data collection.

References

ACSM American College of Sports Medicine/American DieteticAssociation/Dieticians of Canada. (2009). Nutrition and ath-letic performance. Joint position statement. Medicine andScience in Sports and Exercise, 41(3), 709–731.

Armstrong, L. A., Casa, D. J., Maresh, C. M., & Ganio, M. S.(2007). Caffeine, fluid-electrolyte balance, temperature regula-tion, and exercise-heat tolerance. Exercise and Sports ScienceReviews, 35(4), 135–140.

Bottoms, L. M., Hunter, A. M., & Galloway, S. D. R. (2006).Effects of carbohydrate ingestion on skill maintenance in squashplayers. European Journal of Sport Science, 6(3), 187–195.

Bottoms, L., Rome, P., Gregory, K., & Price, M. (2009).Development of a lab based fencing protocol for male epeefencers. Journal of Sports Sciences, 27(S2). s3–s135

Bottoms, L., Taylor, K., Sinclair, J., Polman, R., & Fewtrel, D.(2012). The effects of carbohydrate ingestion on the badmintonserve following fatiguing exercise. Journal of Sports Sciences,30(3), 285–293.

Cureton, K. J., & Warren, G. L. (2007). Caffeinated sports drinks:Ergogenic effects and possible mechanisms. InternationalJournal of Sport Nutrition and exercise Metabolism, 17, 35–55.

Davey, P. R., Thorpe, R. D. & Williams, C. (2002). Fatiguedecreases skilled tennis players. Journal of Sports Science, 20(4), 318–8.

Davis, J. K., & Green, J. M. (2009). Caffeine and anaerobicperformance: Ergogenic value and mechanisms of action.Sports Medicine, 10, 813–832.

Daya, A., Donne, B., & O’Brien, M. (2002). Newly designedfencing face mask: Effects on cardiorespiratory costs and sub-maximal performance. British Journal of Sports Medicine, 36,doi:10.1136/bjsm

Del Coso, J., Estevez, E., & Mora-Rodriguez, R. (2008). Caffeineeffects on short term performance during prolonged exercisein the heat. Medicine and Science in Sports and Exercise, 40(4),744–751.

Del Coso, J., Muñoz-Fernández, V. E., Muñoz, G., Fernández-Elías, V. E., Ortega, J. F., Hamouti, N., … Muñoz-Guerra, J.(2012). Effects of a caffeine-containing energy drink on simu-lated soccer performance. PLoS One, 7(2), doi:10.1371/journal.pone.0031380

Doherty, M., & Smith, P. M. (2004). Effects of caffeine ingestionon exercise testing: A meta-analysis. International Journal ofSports Nutrition and Exercise Metabolism, 14, 626–646.

Foskett, A., Ali, A., & Gant, N. (2009). Caffeine enhances cog-nitive function and skill performance during simulated socceractivity. International Journal of Sports Nutrition and ExerciseMetabolism, 19, 410–423.

Fredholm, B. B., Battig, K., Holmen, J., Nehlig, A., & Zvartau,E. E. (1999). Actions of caffeine in the brain with specialreference to factors that contribute to its widespread use.Pharmacological Review, 51(1), 83–133.

Glastier, M., Howatson, G., Abraham, C., Lockey, R. A.Goodwin, J. E., Foley, P., & McInnes, G. (2008). Caffeinesupplementation and multiple sprint running performance.Medicine and Science in Sports and Exercise, 40(10), 1835–1840.

Graham, T. E., & Spriet, L. L. (1995). Metabolic, catecholamine,and exercise performance responses to various doses of caf-feine. Journal of Applied Physiology, 78, 867–874.

Harmenberg, J., Ceci, R., Barvestad, P., Hjerpe, K., &Nystrom, J. (1991). Comparison of different tests of fencingperformance. International Journal of Sports Medicine, 12,573–576.

Hornery, D. J., Farrow, D., Mujika, I., & Young, W. B. (2007).Caffeine, carbohydrate and cooling use during prolonged simu-lated tennis. International Journal of Sports Physiology andPerformance, 2, 423–438.

Jacobson, B. H., & Edgley, B. M. (1987). Effects of caffeine onsimple reaction time and movement time. Aviation, Space andEnvironmental Medicine, 58(12), 1153–1156.

Kalmar, J. M. (2005). The influence of caffeine on voluntarymuscle activation. Medicine and Science in Sports and Exercise,37(12), 2113–2119.

Kenemans, J. L., Wieleman, J. S., Zeegers, M., & Verbaten,M. N. (1999). Caffeine and Stroop interference. PharmacologyBiochemistry and Behaviour, 63(4), 589–598.

Lopes, J. M., Aubier, M., Jardim, J., Aranda, J. V. & Macklem,P. T. (1983). Effect of caffeine on skeletal muscle function beforeand after fatigue. Journal of Applied Physiology, 54(5), 1303–5.

Murray, T. A., Cook, T. D., Werner, S. L., Schlegel, T. F., &Hawkins, R. J. (2001). The effects of extended play on profes-sional baseball pitchers. American Journal of Sports Medicine,29(2), 137–142.

Noakes, T. D. (2000). Physiological models to understand exer-cise fatigue and the adaptations that predict or enhance athleticperformance. Scandinavian Journal of Sports Medicine, 10,123–145.

Paul, S., Miller, W., Bottoms, L., Beasley, P. (2011). Epee fencing:A step-by-step guide to achieving Olympic gold with no guaranteeyou’ll get anywhere near it. Buxton, UK Wellard Publishing.

Pipe, A., & Ayotte, C. (2002). Nutritional supplements and dop-ing. Clinical Journal of Sports Medicine, 12(4), 245–249.

Plaskett, C. J., & Cafarelli, E. (2001). Caffeine increases enduranceand attenuates force sensation during submaximal isometriccontractions. Journal of Applied Physiology, 91, 1535–1544.

Roberts, S. P., Stokes, K. A., Trewartha, G., Doyle, J., Hogben,P., & Thompson, D. (2010). Effects of carbohydrate and caf-feine ingestion on performance during a rugby union simula-tion protocol. Journal of Sports Sciences, 28(8), 833–842.

Roi, G., & Bianchedi, D. (2008). The science of fencing:Implications for performance and injury prevention. SportsMedicine, 38(6), 465–481.

Royal, K., Farrow, D., Mujika, I., Halson, S., Pyne, D., &Abernethy, B. (2006). The effects of fatigue on decision

1098 L. Bottoms et al.

making and shooting skill performance in water polo players.Journal of Sports Sciences, 24(8), 807–815.

Share, B., Sanders, N., & Kemp, J. (2009). Caffeine and perfor-mance in clay target shooting. Journal of Sports Sciences, 27(6),661–666.

Spriet, L. L. (1995). Caffeine and performance. InternationalJournal of Sport Nutrition, 5, S84–99.

Stuart, G. R., Hopkins, W. G., Cook, C., & Cairns, S. P. (2005).Multiple effects of caffeine on simulated high-intensity team-sport performance. Medicine and Science in Sports and Exercise,37(11), 1998–2005.

Tarnopolsky, M. A., & Cupido, C. (2000). Caffeine potentiateslow frequency skeletal muscle force in habitual and nonhabitualcaffeine consumers. Journal of Applied Physiology, 89,1719–1724.

Vergauwen, L., Spaepen, A. J., Lefevre, J., & Hespel, P. (1998).Evaluation of stroke performance in tennis. Medicine andScience in Sports and Exercise, 30(8), 1281–1288.

WADA. (2012).World anti-doping association banned list. Retrievedfrom http://www.wada-ama.org/en/World-Anti-Doping-Program/Sports-and-Anti-Doping-Organizations/International-Standards/Prohibited-List/QA-on-2011-Prohibited-List/

Williams, L. R. T., & Warmsley, A. (2000). Response timing andmuscular coordination in fencing: A comparison of elite andnovice fencers. Journal of Science and Medicine in Sport, 3(4),460–475.

Yiou, E., & Do, M. C. (2000). In fencing does intensive practiceequally improve the speed performance of the touche when it isperformed alone and in combination with the lunge?International Journal of Sports Medicine, 21, 122–126.

Effect of caffeine on epee fencing performance 1099

Copyright of Journal of Sports Sciences is the property of Taylor & Francis Ltd and its content may not be

copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written

permission. However, users may print, download, or email articles for individual use.