Effect of Caffeine on Sport-Specific Endurance Performance: A Systematic Review

Abstract

Endurance athletes often ingest caffeine because of its reported ergogenic properties. Although there are a vast number of studies quantifying caffeine's effects, many research studies measure endurance performance using a time-to-exhaustion test (subjects exercise at a fixed intensity to volitional exhaustion). Time-to-exhaustion as a performance measure is not ideal because of the high degree of measurement variability between and within subjects. Also, we are unaware of any endurance sports in which individuals win by going a longer distance or for a longer amount of time than their competitors. Measuring performance with a time-trial test (set distance or time with best effort) has high reproducibility and is more applicable to sport. Therefore, the purpose of this review was to critically and objectively evaluate studies that have examined the effect of caffeine on time-trial endurance (>5 minutes) performance. A literature search revealed 21 studies with a total of 33 identifiable caffeine treatments that measured endurance performance with a time-trial component. Each study was objectively analyzed with the Physiotherapy Evidence Database (PEDro) scale. The mean PEDro rating was 9.3 out of 10, indicating a high quality of research in this topic area. The mean improvement in performance with caffeine ingestion was 3.2 +/- 4.3%; however, this improvement was highly variable between studies (-0.3 to 17.3%). The high degree of variability may be dependent on a number of factors including ingestion timing, ingestion mode/vehicle, and subject habituation. Further research should seek to identify individual factors that mediate the large range of improvements observed with caffeine ingestion. In conclusion, caffeine ingestion can be an effective ergogenic aid for endurance athletes when taken before and/or during exercise in moderate quantities (3-6 mg.kg body mass). Abstaining from caffeine at least 7 days before use will give the greatest chance of optimizing the ergogenic effect.

Full text

EFFECT OF CAFFEINE ON SPORT-SPECIFIC
ENDURANCE PERFORMANCE:ASYSTEMATIC REVIEW
MATTHEW S. GANIO,JENNIFER F. KLAU,DOUGLAS J. CASA,LAWRENCE E. ARMSTRONG,
AND CARL M. MARESH
Human Performance Laboratory, Department of Kinesiology, University of Connecticut, Storrs, Connecticut
ABSTRACT
Ganio, MS, Klau, JF, Casa, DJ, Armstrong, LE, and Maresh, CM.
Effect of caffeine on sport-specific endurance performance:
a systematic review. J Strength Cond Res 23(1): 315–324,
2009—Endurance athletes often ingest caffeine because of its
reported ergogenic properties. Althoughthere are a vast number
of studies quantifying caffeine’s effects, many research studies
measure endurance performance using a time-to-exhaustion
test (subjects exercise at a fixed intensity to volitional
exhaustion). Time-to-exhaustion as a performance measure is
not ideal because of the high degree of measurement variability
between and within subjects. Also, we are unaware of any
endurance sports in which individuals win by going a longer
distance or for a longer amount of time than their competitors.
Measuring performance with a time-trial test (set distance or
time with best effort) has high reproducibility and is more
applicable to sport. Therefore, the purpose of this review was
to critically and objectively evaluate studies that have examined
the effect of caffeine on time-trial endurance (.5 minutes)
performance. A literature search revealed 21 studies with
a total of 33 identifiable caffeine treatments that measured
endurance performance with a time-trial component. Each study
was objectively analyzed with the Physiotherapy Evidence
Database (PEDro) scale. The mean PEDro rating was 9.3 out
of 10, indicating a high quality of research in this topic area. The
mean improvement in performance with caffeine ingestion
was 3.2 64.3%; however, this improvement was highly variable
between studies (20.3 to 17.3%). The high degree of variability
may be dependent on a number of factors including ingestion
timing, ingestion mode/vehicle, and subject habituation.
Further research should seek to identify individual factors that
mediate the large range of improvements observed with caffeine
ingestion. In conclusion, caffeine ingestion can be an effective
ergogenic aid for endurance athletes when taken before and/or
during exercise in moderate quantities (3–6 mg!kg
21
body
mass). Abstaining from caffeine at least 7 days before use will
give the greatest chance of optimizing the ergogenic effect.
KEY WORDS aerobic exercise, competition, ergogenic, racing,
theophylline
INTRODUCTION
Caffeine (CAF) is commonly ingested by athletes
because of its reported ergogenic effects (23).
Caffeine has been proposed to improve physical
performance by acting independently, or concur-
rently, via 3 different mechanisms: 1) an increased mobili-
zation of intracellular calcium, 2) an increase in free fatty acid
oxidation, and 3) serving as an adenosine receptor antagonist
in the central nervous system (59). Early research by Costill
et al. (16) suggested that the ergogenic effect of CAF with
aerobic exercise was related to an increase in fatty acid
oxidation and subsequent sparing of muscle glycogen.
However, recent research and reviews conclude that CAF
affects endurance performance largely through its antagonist
effect on adenosine receptors in the brain (20,40). Acting
through this mechanism, CAF may modulate central fatigue
and influence ratings of perceived exertion, perceived pain,
and levels of vigor, all of which may lead to performance
improvements (39,51).
A recent meta-analysis by Doherty and Smith (23) has
quantified the effect of CAF on endurance time to exhaustion
and on short-term, high-intensity exercise protocols. Caffeine
was found to improve endurance time to exhaustion (effect
size ½ES#= 0.63) to a greater degree than short-term
performance (ES = 0.16). Exercise is generally classified as
endurance activity (vs. short-term) when the majority of
energy is produced through aerobic (vs. anaerobic) pathways
(55). A higher percentage of total energy is produced
aerobically after about 3 minutes of exercise, and this is
independent of exercise mode. Although Doherty and Smith
(23) categorize the different protocols examined, they never
specify their criteria for classifying a protocol as ‘‘endurance’’
or ‘‘short-term.’’ They identify several protocols lasting .5
minutes as ‘‘short-term, high intensity’’ (1,10).
To isolate the effect of CAF on performance, CAF capsules
are commonly used as the mode of delivery (1,6,7), but other
Address correspondence to Matthew S. Ganio, matthew.ganio@uconn.
edu.
23(1)/315–324
Journal of Strength and Conditioning Research
!2009 National Strength and Conditioning Association
VOLUME 23 | NUMBER 1 | JANUARY 2009 | 315
Copyright © . N ational S trength and Conditioning A ssociation. Unauthorized reproduction of this article is prohibited

forms of CAF delivery include gum (49), carbohydrate-
electrolyte (CE) solutions (18,44,57), and coffee (65). There is
the possibility that these substances (e.g., carbohydrate) may
contribute to CAF’s direct ergogenic actions. Despite
isolating the effect of CAF (e.g., conducting a trial with
carbohydrate only and another with carbohydrate plus
CAF), many studies delivering CAF in modes other than
capsules have not been included in reviews and meta-
analyses (23). However, it is important to evaluate the
findings from these studies because CAF used in sport is
commonly consumed in modes other than capsules (21). For
example, in a recent survey of 2005 Ironman Triathlon World
Championship athletes, 78% of the respondents planned to
use caffeinated cola drinks and 42% planned to use
caffeinated gels (21).
The ergogenic effect of CAF is most consistently observed
when performance is measured with a ‘‘test-to-exhaustion’’
protocol (i.e., performance is measured as time to volitional
exhaustion while exercising at a fixed intensity ½%
_
VO
2
max#)
(23). Unfortunately, this type of protocol may not be ideal
because a small change in an individual’s ability to increase
power results in a large change in time to exhaustion and
a high coefficient of variation (33). Therefore, improvements
may be observed despite a large amount of measurement
variability and inflation of type I error (33). A recent study by
Laursen et al. (45) further suggests that there is a greater
coefficient of variation when subjects perform a test to
exhaustion vs. a time trial. Further, a test to exhaustion does
not elucidate the true performance effect of CAF (38)
because we are unaware of any endurance sport in which
individuals win by going a longer distance or for a longer
amount of time than their competitors. Endurance/aerobic
sport requires competitors to complete a set distance or
amount of work in the shortest time possible (time trial) or
a maximal amount of work in a fixed amount of time.
Therefore, studies examining the ergogenic effect of CAF
using a time-trial protocol are more applicable to athletes.
The purpose of this systematic review is to analyze critically
the effect of CAF on performance in practical endurance sport
settings. This includes studies that have measured perfor-
mance with a time-trial protocol $5 minutes in duration. We
have also included studies using different modes of CAF
administration but in which the effect of CAF has been
appropriately isolated.
METHODS
In August 2007, potential studies were identified by searching
proQuest, SportDiscus, MEDLINE, and Scopus using the
following search terms in varying combinations: caffeine,
caffeinated, coffee, theophylline, paraxanthine, theobromine,
endurance, and exercise. Articles were then prescreened and
cross-referenced to identify those containing time-trial
components. Review articles and book chapters published
since 1985 that pertained to CAF and exercise performance
also were cross-referenced for relevant articles (19,22–25,
28,30,40,43,47,52,54,56,59,60).
Experimental Approach to the Problem
Inclusionary Criteria. Human studies of men and women were
included if the effect of CAF administered in any form could
be isolated and if the performance protocol lasted $5 minutes
and had any component of a time trial. This included studies
in which performance was measured as the time to complete
a set distance or maximal amount of work completed in
a fixed amount of time.
Although a recent meta-analysis (23) has examined the
effects of CAF on performance, this review includes studies
published since that meta-analysis (9,18,48,49,61), those that
were conducted in a field setting (7,9,12,61), and those that
used modes of CAF delivery other than capsules (17,18,34–
36,44,48,50,62,64,65). Although it could be argued that other
factors may confound results in these studies, these studies
are applicable to sports in which athletes often ingest CAF
along with other substances (21). Studies using modes of
CAF ingestion other than capsules were only included if the
independent effects of CAF could be distinguished (i.e., a CE
solution was used with and without CAF).
Studies examining the effects of CAF in combination with
factors other than exercise (e.g., altitude, sleep deprivation)
were only used if there was a control condition with placebo
and CAF trials. Dissertations were included only if the same
data were not presented in a peer-reviewed published paper.
Although double-blind studies are ideal, studies were in-
cluded if they at least used single-blind (blind to participants)
CAF administration.
Exclusionary Criteria. Studies in which performance was
measured by time to exhaustion were not included.
Although graded exercise tests (i.e.,
_
VO
2
max tests) last longer
than 5 minutes, these tests are another form of test to
exhaustion (45).
Physiotherapy Evidence Database Scale. A systematic review
was conducted using the Physiotherapy Evidence Database
(PEDro) scale to rate each article. Although other type of
reviews exist (e.g., meta-analysis), some statisticians argue
that they are susceptible to bias and other problems (27,50).
The PEDro scale was developed by the Centre for
Evidence-Based Physiotherapy (53). This rating scale was
deemed acceptable because of its ability to objectively assess
a study’s internal validity. For example, the scale includes
questions related to levels of subject and assessor blinding.
This knowledge can have large implications when evaluating
the effects of a treatment on performance. The PEDro scale
is a 11-item checklist that yields a maximum score of 10
because no points are awarded for meeting the inclusionary
criterion.
Each article was independently analyzed by 2 reviewers
(k= 0.85) and given a PEDro score. One article (17) included
2 studies (2 subject pools) that were analyzed independently
316
Journal of Strength and Conditioning Research
the
TM
Caffeine and Endurance Performance
Copyright © . N ational S trength and Conditioning A ssociation. Unauthorized reproduction of this article is prohibited

and, thus, received 2 PEDro scores. Articles with discrepant
component PEDro scores were analyzed by a third, in-
dependent reviewer. Any PEDro scores ,6 were deemed
unacceptable and were not used in analyses.
RESULTS
The attrition of studies identified, prescreened, and selected
for analysis is outlined in Figure 1. After preliminary
screening, 26 studies were identified. One study published
as 2 articles (one focusing on cognitive, the other on
physiological measures) contained the same performance
data (32,44); thus, the article focusing on cognitive measures
was excluded (32). Another study (3) failed to meet the
PEDro criterion score of 6, and one failed to isolate the effect
of CAF (4). Two studies (48,49) tested the effects of CAF in
combination with sleep deprivation but failed to test the
effects without sleep deprivation and, thus, were excluded
from analysis. One article (17) consisted of 2 studies with
different subjects and analyses; therefore, it received 2 PEDro
scores. Of the 21 remaining studies, 33 identifiable CAF
treatments were employed (e.g., different CAF dosages in
separate trials), as presented in Tables 1 and 2.
The mean PEDro rating was a 9.3 out of 10. Sixteen of the
studies scored a perfect 10. The most common PEDro item
that studies failed to achieve was ‘‘blinding of therapists’’ and
‘‘blinding of assessors’’ (4 studies for each ½17,34,35,61#). By
failing to double-blind a study that is largely dependent on
subject motivation and assessor encouragement, there is a risk
that results are biased. It is important to note that failing to
identify blinding procedures implies that blinding procedures
were not used.
Twelve trials administered CAF in capsule form (i.e., with
water) (mean 6SD improvement ½mean improvement#= 2.9
64.8%), and 13 trials examined the effects of CAF ingested
along with a CE solution compared with a CE solution alone
(mean improvement = 3.2 63.8%). Four trials used CAF
ingested with other substances (mean improvement = 2.9 6
2.2%), and 4 did not report the mode of CAF delivery (mean
improvement = 6.4 67.4%). Interestingly, among trials in
which CAF was ingested before and during exercise, the
CAF was ingested along with a CE solution in all but 2 trials
(Table 2).
Out of the 33 trials, 21 used cycling (mean improvement =
4.4 65.0%), 6 used running (mean improvement = 0.9 6
0.7%), 4 used rowing (mean improvement = 1.1 60.3%),
1 used swimming (mean im-
provement = 1.7%), and 1 used
cross-country skiing (mean im-
provement = 1.1%) as the mode
of exercise. Sixteen trials used
a set intensity before the time-
trial component (mean im-
provement = 4.7 65.6%); the
other 17 did not (mean im-
provement = 1.8 61.4%).
Fourteen of the articles used
time to complete a set distance
as the performance measure
(mean improvement = 1.1 6
0.5%), 10 used maximum work
produced in a fixed time (mean
improvement = 4.3 64.1%),
and 9 used time to complete
a set number of revolutions (W)
while cycling (mean improve-
ment = 5.2 66.2%). Women
were included in 10 of the 29
trials. Only 1 study (2 trials)
tested only women subjects (1);
the 8 remaining trials with
women did not statistically
analyze differences between
sexes.
Independent of ingestion
timing, the average perfor-
mance improvement with CAF
was 3.2 64.3% over placebo.
Figure 1. Decision tree for identifying, screening, and selecting studies for analyses.
VOLUME 23 | NUMBER 1 | JANUARY 2009 | 317
Journal of Strength and Conditioning Research
the
TM
|
www.nsca-jscr.org
Copyright © . N ational S trength and Conditioning A ssociation. Unauthorized reproduction of this article is prohibited

TABLE 1. Effect of caffeine on performance when ingested before exercise.
Reference
Men
(n)
Women
(n)
Caffeine
delivery
mode
Volume and composition
of fluid ingested
Caffeine time of
administration
(minutes before
exercise)
Total
caffeine
(mg!kg
21
) Protocol
Total
exercise
time
(min)
Improvement
over
placebo
(%)
Pedro
score
Jenkins et al. (37) 13 0 Capsule 450 ml of water before 60 1 15-min time trial after
15 min at 80%
_
VO
2
max{
30 20.7 10
Cohen et al. (12) 5 2 Capsule Ad libitum water throughout 60 9 21-km race§ 89 20.1 9
Anderson et al. (1) 0 8 Capsule 193.2 ml of water before 60 6 2-km time trial
k
8 0.7 10
Jacobson et al. (36) 8 0 ‘‘Fat meal’’ 1.3 L of water throughout 60 6 7-kJ!kg
21
time trial
after 120 min
at 70%
_
VO
2
max{
150 0.7 10
Cohen et al. (12) 5 2 Capsule Ad libitum water throughout 60 5 21-km race§ 89 0.8 9
Bruce et al. (10) 8 0 Capsule 3 ml!kg
21
of water before 60 9 2-km time trial
k
7 1.0* 10
Bridge and Jones (9) 8 0 Capsule NR 60 3 8-km race§ 32 1.2* 10
Anderson et al. (1) 0 8 Capsule 193.2 ml of water before 60 9 2-km time trial
k
8 1.3* 10
Bruce et al. (10) 8 0 Capsule 3 ml!kg
21
of water before 60 6 2-km time trial
k
7 1.3* 10
Wiles et al. (65) 18 0 Coffee 300 ml water before 60 2.5‡1.5-km time trial§ 5 1.4* 10
Bell et al. (6) 10 2 Capsule NR 90 4 10-km time trial†§ 45 1.7 10
Berglund and
Hemmingsson (7)
10 4 Capsule NR 60 6 20-km race# 60 1.7 9
MacIntosh and
Wright (46)
7 4 Capsule Ad libitum water prior 150 6 1.5-km time trial** 21 1.7* 10
Collomp et al. (13) 8 0 Capsule NR 60 6 10-min time trial
after 10 min
at 95%
_
VO
2
max{
20 2.2 10
Jenkins et al. (37) 13 0 Capsule 450 ml of water before 60 3 15-min time trial after
15 min at 80%
_
VO
2
max{
30 2.9* 10
Cox et al. (17) 12 0 Capsule 2.7 L 6% carbohydrate-
electrolyte solution
throughout
60 6 7-kJ!kg
21
time trial after
120 min at 70%
_
VO
2
max{
148 3.4* 6
Jacobson et al. (36) 8 0 Carbohydrate
meal
1.3 L of water throughout 60 6 7-kJ!kg
21
time trial after
120 min at 70%
_
VO
2
max{
150 4.1 10
Jenkins et al. (37) 13 0 Capsule 450 ml of water before 60 2 15-min time trial
after 15 min
at 80%
_
VO
2
max{
30 4.3* 10
Conway et al. (15) 8 0 Capsule NR 60 6 ~30-min time trial after
90 min at 68%
_
VO
2
max{
120 14.5 10
*Significant improvement over placebo trial; †each subject, by wearing a helmet and backpack, donned an additional 11 kg; ‡calculated by authors of the present study; §running;
k
rowing; {cycling; #cross-country skiing; **swimming. NR, not reported.
318
Journal of Strength and Conditioning Research
the
TM
Caffeine and Endurance Performance
Copyright © . N ational S trength and Conditioning A ssociation. Unauthorized reproduction of this article is prohibited

TABLE 2. Effect of caffeine on performance when ingested before and during exercise.
Reference
Men
(n)
Women
(n)
Fluid
composition
Volume (L)
and timing (min)
of fluid ingestion
Amount (mg!kg
21
)
and timing (min)
of caffeine ingestion
Total
caffeine
(mg!kg
21
) Protocol
Total
exercise
time
(min)
Improvement
over
placebo (%)
Pedro
score
Wemple et al. (64) 4 2 6% CES P: 0.53 at 60 min P: 1.9 at 60 min 8.7 500-revolution time
trial after 180 min
at 60%
_
VO
2
max§
186 20.3 10
S: 0.2 S: 0.7
D: 0.2 from 20 to
220 min
D: 0.7 from 20 to
220 min
van Nieuwenhoven
et al. (61)
90 8 7% CES P: 0 P: 0 1.25 18-km race‡78 0.4 8
S: 0.15 S: 0.31
D: 0.31 at 4.5, 9,
and 13.5 km
D: 0.31 at 4.5, 9,
and 13.5 km
Eschbach (26) 11 0 6% CES P: 0 P: 6 at 180 min 9 5-km time trial after
240 min at 55%
_
VO
2
max§
250 0.8 10
S: 0 S: 0
D: 0.25 every 15 min D: 3 at 60 min
Hunter et al. (34) 8 0 7% CES P: 0 P: 6 at 60 min 9.3 100-km time trial with
9 sprints
throughout§
158 1.3 7
S: 0 S: 0
D: 0.15 every 15 min D: 0.33 every 15 min
Kovacs et al. (44) 15 0 7% CES P: 0.58 at 60 min P: 0.7 at 60 min 2.1 60-min time trial§ 60 1.8 10
S: 0 S: 0
D: 0.22 at 20 and
40 min
D: 0.7 at 20 and 40 min
Cox et al. (17) 8 0 6% CES P: 0 P: 0 1.9 7-kJ!kg
21
time trial
after 120 min at
70%
_
VO
2
max§
148 1.8* 7
S: 0 S: 0
D: 0.34 every 20 min D: 0.95 at minutes 80
and 100 during time trial
Cox et al. (17) 8 0 6% CES P: 0 P: 0 1.9 7-kJ!kg
21
time trial
after 120 min at
70%
_
VO
2
max§
148 2.4* 7
S: 0 S: 0
D: 0.34 every 20 min D: 0.95 at minutes 80
and 100 during time trial†
Cox et al. (17) 12 0 6% CES P: 0 P: 0 6 7-kJ!kg
21
time trial
after 120 min at
70%
_
VO
2
max§
148 3.1 7
S: 0 S: 0
D: 0.34 every 20 min D: 1 every 20 min
Ganio (personal
communication,
August 12, 2007)
14 0 6% CES P: 0 P: 0 5.85 15-min time trial after
120 min at 60 and
75%
_
VO
2
max§
135 3.6 10
S: 0.44 S: 1.2
D: 0.22 every 15 min D: 0.5 every 15 min (Continued on next page)
VOLUME 23 | NUMBER 1 | JANUARY 2009 | 319
Journal of Strength and Conditioning Research
the
TM
|
www.nsca-jscr.org
Copyright © . N ational S trength and Conditioning A ssociation. Unauthorized reproduction of this article is prohibited

Kovacs et al. (44) 15 0 7% CES P: 0.58 at 60 min P: 1.1 at 60 min 3.2 60-min time trial§ 60 4.2* 10
S: 0 S: 0
D: 0.22 at 20 and
40 min
D: 1.1 at minutes 20 and 40
Kovacs et al. (44) 15 0 7% CES P: 0.58 at 60 min P: 1.5 at 60 min 4.5 60-min time trial§ 60 4.2* 10
S: 0 S: 0
D: 0.22 at 20 and
40 min
D: 1.5 at minutes 20 and 40
Ivy et al. (35) 7 2 Lemonade NR P: 3.6 at 60 min 7.2 120-min time trial§ 120 5.3* 8
S: 0
D: 0.4 every 15 min
Cureton et al. (18) 16 0 6% CES P: 0 P: 0 5.3 15-min time trial after
120 min at 60 and
75%
_
VO
2
max§
135 15.0* 10
S: 0.44 S: 1.2
D: 0.22 every 15 min D: 0.5 every 15 min
Conway et al. (15) 8 0 Water P: 0 P: 3 at 60 min 6 ~30-min time trial
after 90 min at
68%
_
VO
2
max§
120 17.3 10
S: 0 S: 0
D: 1.4 ad libitum D: 3 at 45 min
*Significant improvement over placebo trial; †6% carbohydrate-electrolyte solution (CES) was substituted for 11% CES at these time points; ‡running; §cycling. NR, not reported;
P, prior to exercise; S, start of exercise; D, during exercise.
320
Journal of Strength and Conditioning Research
the
TM
Caffeine and Endurance Performance
Copyright © . N ational S trength and Conditioning A ssociation. Unauthorized reproduction of this article is prohibited

The median improvement with CAF was 1.7%. Thirty of the
33 trials showed positive improvements in performance with
CAF, but only 15 were statistically significant (p,0.05). One
study with 2 different CAF trials observed large performance
improvements (14.5 and 17.3%) but were not statistically
significant, possibly because of a type II error (15). Another
study (18) observed a large, significant improvement (15%)
over placebo. This improvement may be attributable to the
combination of repeated CAF ingestion throughout the
exercise protocol and/or the unique submaximal protocol in
which subjects alternated between 60 and 75%
_
VO
2
max for
120 minutes before starting a 15-minute time trial.
Caffeine ingested before exercise resulted in a mean
performance improvement of 2.3 63.2% (Figure 2).
Performance was improved 4.3 65.3% when CAF was
ingested both before and during exercise (Figure 2). Total
CAF ingestion does not explain these differences (mean = 5.3
mg!kg
21
, Table 1; 5.2 mg!kg
21
, Table 2). Further, the range of
performance improvement is similar when CAF is ingested
before exercise (20.7 to 14.5%, Table 1) and when ingested
before and during exercise (20.3 to 17.3%, Table 2).
DISCUSSION
The purpose of this systematic review was to critically
evaluate studies that examined the effects of CAF on sport-
specific endurance performance. We only reviewed articles
that had time-trial components. This type of performance test
is not only more valid in detecting treatment effects (33), but,
more importantly, it is applicable to sport. For this same
reason, we chose to include articles that were conducted in
field settings or that used the treatment of CAF with other
substances (i.e., CE solutions). Although ingesting CAF via
capsules may prevent any interaction with other ingested
foods or fluid, there is limited availability of ready-made CAF
capsules that an athlete may ingest. There are quite a few
studies examining CAF with CE solutions. This is not
surprising because the American College of Sports Medicine
recommends the use of CE solutions in long-duration
exercise (58), a setting in which CAF is a commonly used
ergogenic aid (23). Some have hypothesized that a high-
carbohydrate diet may attenuate the increase in free fatty
acids observed with CAF ingestion and, thus, modulate the
ergogenic effect of CAF (63). This has not been observed in
performance settings, and it likely has little effect. The
ergogenic effect of CAF, especially in non–glycogen-limiting
exercise, is thought to be more central in nature and not
metabolic (20,40).
We found that CAF was equally ergogenic independent of
delivery mode (mean = 2.9% improvement for water and 3.2%
for CE solutions). Other forms of administration (i.e., gum,
soft drink, coffee) have not been well studied, especially not
with time-trial exercise protocols. The bioavailability of CAF
via the gastrointestinal system is relatively quick (i.e., detected
in the blood within 30 minutes), but the coingestion of CAF
with other compounds may (17) or may not (29) slow down
the appearance of CAF in the blood. However, absorption
of CAF orally (via gum) is quicker than intestinally (via
capsules) because of the presence of buccal mucosa in
the mouth (41). Independent of mode, the breakdown of
CAF is slow (i.e., half-life ap-
proximately 4–6 hours) (41).
Critically examining studies us-
ing CAF delivery through
modes other than capsules is
important because many ath-
letes use other modes of CAF
delivery before and during
competition (21).
The ingestion of CAF is
similarly ergogenic regardless
of ingestion approximately 60
minutes before or ingestion
during exercise. The mean im-
provement is slightly greater
when CAF is ingested both
before and during exercise (4.3
65.3%) compared with only
before exercise (2.3 63.2 %);
this may be attributable to
a different number of trials in
each group (n= 14 and 19,
respectively) and/or a large de-
gree of variability. Regardless of
administration timing, there is a
Figure 2. The effect of caffeine (CAF) on exercise performance when ingesting CAF before exercise (above
horizontal hashed line) or before and during exercise (below horizontal hashed line). Vertical lines represent mean
percent improvement of all studies (2.3 63.2 and 4.3 65.3% for above and below the hashed line, respectively).
*Significantly different from placebo (p,0.05). #Ganio (personal communication, August 12, 2007).
VOLUME 23 | NUMBER 1 | JANUARY 2009 | 321
Journal of Strength and Conditioning Research
the
TM
|
www.nsca-jscr.org
Copyright © . N ational S trength and Conditioning A ssociation. Unauthorized reproduction of this article is prohibited

large range of improvements shown between studies (20.7 to
17.3% improvement, Figure 2). Assuming an equal amount of
total CAF, the timing of ingestion (before or during exercise)
does not seem to be an important factor in eliciting the
ergogenic effects of CAF. This is supported by studies that
have conducted trials varying CAF ingestion timing and
finding no differences in performance (15,17). Shorter races
that do not usually involve rehydration during exercise may
not be as conducive to CAF ingestion during exercise. On the
other hand, CAF ingestion during exercise may be more
feasible in longer-duration exercise when concurrent with
fluid ingestion.
Degree of improvement with CAF does not seem to be
consistently associated with mode of CAF delivery, timing,
total exercise time, or the exercise mode employed (Tables 1
and 2). Total CAF ingested seems to be loosely associated
with degree of improvement. It is generally observed that
quantities above 3 mg!kg
21
body mass are needed for
improvement. Quantities up to 6 mg!kg
21
are most com-
monly used, but this amount and greater amounts do not
always result in performance improvements. Using a time-
trial protocol, Bruce et al. (10) did not observe increased
performance compared with placebo when increasing CAF
ingestion from 6 to 9 mg!kg
21
body mass. In the studies we
reviewed, 9 mg!kg
21
resulted in improvements no greater than
1.5% over placebo (Tables 1 and 2). Because some athletic
governing bodies have restrictions against large amounts of
CAF ingestion (i.e., National Collegiate Athletic Association),
but none completely ban CAF use, it is recommended that
CAF use not exceed 9 mg!kg
21
. Performance improvements
with CAF ingestion are maximized with amounts up to
6mg!kg
21
and are not further improved with 9 mg!kg
21
(10,31). The ergogenicity of CAF up to 6 mg!kg
21
has been
observed in a variety of settings, but factors such as one’s
habitual use of CAF may change the dose needed to elicit an
ergogenic effect.
Although the recommendations for maximizing CAF’s
ergogenic properties presented above are also made else-
where (43,59,60), it is evident that performance improve-
ments with CAF are varied and independent of exercise
mode, duration (when .5 minutes), and protocol (Tables 1
and 2). It should be noted that CAF in moderate
consumption does not impact hydration status or thermo-
regulation in exercising individuals (2). Recent reviews
conclude that CAF’s primary mode of action involves
adenosine receptor antagonism in the central nervous system
(28,43,52). Caffeine is able to cross the blood-brain barrier
and is a powerful antagonist of adenosine receptors in the
central nervous system (8). As a result, CAF counteracts the
inhibitory effects of adenosine on neuroexcitability, neuro-
transmitter release, and arousal (20). Because of the potential
importance of adenosine receptors on central fatigue, it is
important to understand factors that may change or
modulate adenosine receptor number or sensitivity. Changes
in adenosine receptor number or sensitivity may play a role in
the effect that CAF has on exercise performance. Chronic
CAF consumption in animal models results in upregulation
of the number and an increase in the affinity of adenosine
receptors within the central nervous system (11,42). This
may result in an increased amount of CAF needed to have
the same antagonist activity on the receptors (termed
‘‘caffeine habituation’’). It is possible that the varied degree
of improvements observed between studies (Figure 2) may
be attributable to lack of control over subject habituation.
Although there are no known studies examining the effects
of CAF habituation on time-trial performance, several studies
have examined CAF habituation using other exercise
performance protocols. Using a time-to-exhaustion protocol,
Van Soeren and Graham (62) measured performance after
subjects abstained from CAF for 0, 2, and 4 days. There was
a trend for greater improvement with CAF ingestion after
abstaining from CAF for 2 and 4 days (vs. 0 days). Similarly,
using a time-to-exhaustion protocol, Bell and McLellan (5)
showed that improvements in performance were greater
for CAF nonusers (,50 mg CAF per day) vs. users ($300
mg of CAF per day). Therefore, it is possible that CAF
habituation may modulate performance improvements with
acute CAF ingestion.
It is not known how many days an endurance athlete
should abstain from CAF to maximize its ergogenic effects,
but animal studies show that increases in adenosine receptor
number and affinity are maximized in 7 days (42). Therefore,
we recommend that athletes abstain from CAF ingestion for
no fewer than 7 days before competition. This should allow
for withdrawal symptoms (which may negatively affect
performance) to subside and allow sufficient time for
adenosine receptor downregulation to occur (42), thus
possibly maximizing the ergogenic effects of CAF. Although
abstaining from CAF before use in an athletic setting is ideal,
some may find it too difficult because of withdrawal
symptoms (e.g., headaches, fatigue, lethargy, flulike symp-
toms) (5,62). Over-the-counter medicine may help alleviate
these symptoms, but the interaction of these substances with
CAF is unknown. When CAF is habitually consumed, one
may improperly conclude that an increase in dosage may be
sufficient to elicit an ergogenic effect similar to that
experienced by a CAF-naı
¨ve individual. Unfortunately, this
specific scenario has not been examined in a performance
setting. Regardless of habituation level, the ingestion of large
amounts of CAF may result in negative side effects (14);
therefore, abstaining from CAF before use possibly will give
the greatest chance of optimizing the ergogenic effect (5).
In conclusion, there are a number of high-quality research
articles that have examined the effects of CAF on time-trial
endurance performance. The expected performance im-
provements with CAF ingestion are varied, but they may
be dependent on a number of factors including timing,
ingestion mode/vehicle, and subject habituation. Given the
available evidence, we recommend that endurance athletes
abstain from CAF use at least 7 days before competition.
322
Journal of Strength and Conditioning Research
the
TM
Caffeine and Endurance Performance
Copyright © . N ational S trength and Conditioning A ssociation. Unauthorized reproduction of this article is prohibited

Acute ingestion should occur no more than 60 minutes before
and, if practical, during competition. The amount of CAF
commonly shown to improve endurance performance is
between 3 and 6 mg!kg
21
body mass; these amounts are
equally effective when combined with a CE solution or water.
Further research should seek to identify specific factors that
mediate the large range of improvements commonly
observed with CAF ingestion.
PRACTICAL APPLICATIONS
Caffeine is a widely used legal drug that has been shown
to improve endurance performance and that, theoretically,
could be used before training sessions when high-intensity
exercise is desired. The degree of performance improvement
is variable and likely influenced by the timing of ingestion,
amount ingested, mode of ingestion, and how often an athlete
consumes CAF on a daily basis (level of habituation). To
maximize performance improvements with CAF ingestion, it
is recommended that athletes consume up to 6 mg!kg
21
body
mass no more than 60 minutes before exercise, but it also
may be consumed during exercise. Caffeine seems to be
equally effective when ingested in combination with CE
solutions (e.g., sports drinks) or other modes of ingestion
(e.g., gum), but other substances in caffeinated-coffee may
counteract CAF’s performance-improving properties. Evi-
dence suggests that consuming CAF every day may dampen
the degree of performance improvement observed when
CAF naı
¨ve. Therefore, we recommend that athletes abstain
from CAF no fewer than 7 days before its use in competition.
Because some individuals react differently to CAF than
others (i.e., CAF sensitive), it is recommended that athletes
try CAF while training before using it in competition.
REFERENCES
1. Anderson, ME, Bruce, CR, Fraser, SF, Stepto, NK, Klein, R,
Hopkins, WG, and Hawley, JA. Improved 2000-meter rowing
performance in competitive oarswomen after caffeine ingestion. Int J
Sport Nutr Exerc Metab 10: 464–475, 2000.
2. Armstrong, LE, Casa, DJ, Maresh, CM, and Ganio, MS. Caffeine,
fluid-electrolyte balance, temperature regulation, and exercise-heat
tolerance. Exerc Sport Sci Rev 35: 135–140, 2007.
3. Asumussen, E and Boje, O. The effect of alcohol and some drugs on
the capacity for work. Acta Physiol Scand 15: 109–113, 1948.
4. Bell, DG and Jacobs, I. Combined caffeine and ephedrine ingestion
improves run times of Canadian Forces Warrior Test. Aviat Space
Environ Med 70: 325–329, 1999.
5. Bell, DG and McLellan, TM. Exercise endurance 1, 3, and 6 h after
caffeine ingestion in caffeine users and nonusers. J Appl Physiol 93:
1227–1234, 2002.
6. Bell, DG, McLellan, TM, and Sabiston, CM. Effect of ingesting
caffeine and ephedrine on 10-km run performance. Med Sci Sports
Exerc 34: 344–349, 2002.
7. Berglund, B and Hemmingsson, P. Effects of caffeine ingestion on
exercise performance at low and high altitudes in cross-country
skiers. Int J Sports Med 3: 234–236, 1982.
8. Biaggioni, I, Paul, S, Puckett, A, and Arzubiaga, C. Caffeine and
theophylline as adenosine receptor antagonists in humans.
J Pharmacol Exp Ther 258: 588–593, 1991.
9. Bridge, CA and Jones, MA. The effect of caffeine ingestion on
8 km run performance in a field setting. J Sports Sci 24: 433–439,
2006.
10. Bruce, CR, Anderson, ME, Fraser, SF, Stepto, NK, Klein, R,
Hopkins, WG, and Hawley, JA. Enhancement of 2000-m rowing
performance after caffeine ingestion. Med Sci Sports Exerc 32: 1958–
1963, 2000.
11. Chou, DT, Khan, S, Forde, J, and Hirsh, KR. Caffeine tolerance:
behavioral, electrophysiological and neurochemical evidence. Life
Sci 36: 2347–2358, 1985.
12. Cohen, BS, Nelson, AG, Prevost, MC, Thompson, GD, Marx, BD,
and Morris, GS. Effects of caffeine ingestion on endurance racing in
heat and humidity. Eur J Appl Physiol 73: 358–363, 1996.
13. Collomp, K, Candau, R, Millet, G, Mucci, P, Borrani, F, Prefaut, C,
and De Ceaurriz, J. Effects of salbutamol and caffeine ingestion
on exercise metabolism and performance. Int J Sports Med 23: 549–
554, 2002.
14. Conlee, RK. Amphetamine, caffeine and cocaine. In: Ergogenics:
Enhancement of Performance in Exercise and Sport. D. R. Lamb and
M. H. Williams, eds. Indianapolis: Brown and Benchmark, 1991.
pp. 285–330.
15. Conway, KJ, Orr, R, and Stannard, SR. Effect of a divided caffeine
dose on endurance cycling performance, postexercise urinary
caffeine concentration, and plasma paraxanthine. J Appl Physiol 94:
1557–1562, 2003.
16. Costill, DL, Dalsky, GP, and Fink, WJ. Effects of caffeine ingestion
on metabolism and exercise performance. Med Sci Sports 10:
155–158, 1978.
17. Cox, GR, Desbrow, B, Montgomery, PG, Anderson, ME, Bruce, CR,
Macrides, TA, Martin, DT, Moquin, A, Roberts, A, Hawley, JA, and
Burke, LM. Effect of different protocols of caffeine intake on
metabolism and endurance performance. J Appl Physiol 93: 990–
999, 2002.
18. Cureton, KJ, Warren, GL, Millard-Stafford, ML, Wingo, JE, Trilk, J,
and Buyckx, M. Caffeinated sports drink: ergogenic effects and
possible mechanisms. Int J Sport Nutr Exerc Metab 17: 35–55, 2007.
19. Daly, JW and Fredholm, BB. Caffeine—an atypical drug of
dependence. Drug Alcohol Depend 51: 199–206, 1998.
20. Davis, JM, Zhao, Z, Stock, HS, Mehl, KA, Buggy, J, and Hand, GA.
Central nervous system effects of caffeine and adenosine on fatigue.
Am J Physiol Regul Integr Comp Physiol 284: R399–R404, 2003.
21. Desbrow B and Leveritt, M. Awareness and use of caffeine by
athletes competing at the 2005 Ironman Triathlon World Cham-
pionships. Int J Sport Nutr Exerc Metab 16: 545–558, 2006.
22. Dodd, SL, Herb, RA, and Powers, SK. Caffeine and exercise
performance. An update. Sports Med 15: 14–23, 1993.
23. Doherty, M and Smith, PM. Effects of caffeine ingestion on
exercise testing: a meta-analysis. Int J Sport Nutr Exerc Metab 14:
626–646, 2004.
24. Doherty, M and Smith, PM. Effects of caffeine ingestion on rating of
perceived exertion during and after exercise: a meta-analysis. Scand J
Med Sci Sports 15: 69–78, 2005.
25. Ellender, L and Linder, MM. Sports pharmacology and ergogenic
aids. Prim Care 32: 277–292, 2005.
26. Eschbach, LC. The effect of caffeine ingestion on metabolism and
performance during prolonged cycling. Dissertation, University of
Southern Mississippi, School of Human Performance & Recreation,
Hattiesburg, 2001.
27. Eysenck, HJ. Meta-analysis and its problems. Br Med J 309: 789–
792, 1994.
28. Graham, TE. Caffeine and exercise: metabolism, endurance and
performance. Sports Med 31: 785–807, 2001.
29. Graham, TE, Hibbert, E, and Sathasivam, P. Metabolic and exercise
endurance effects of coffee and caffeine ingestion. J Appl Physiol 85:
883–889, 1998.
VOLUME 23 | NUMBER 1 | JANUARY 2009 | 323
Journal of Strength and Conditioning Research
the
TM
|
www.nsca-jscr.org
Copyright © . N ational S trength and Conditioning A ssociation. Unauthorized reproduction of this article is prohibited

30. Graham, TE, Rush, JW, and van Soeren, MH. Caffeine and exercise:
metabolism and performance. Can J Appl Physiol 19: 111–138, 1994.
31. Graham, TE and Spriet, LL. Metabolic, catecholamine, and exercise
performance responses to various doses of caffeine. J Appl Physiol 78:
867–874, 1995.
32. Hogervorst, E, Riedel, WJ, Kovacs, E, Brouns, F, and Jolles, J.
Caffeine improves cognitive performance after strenuous physical
exercise. Int J Sports Med 20: 354–361, 1999.
33. Hopkins, WG, Schabort, EJ, and Hawley, JA. Reliability of power in
physical performance tests. Sports Med 31: 211–234, 2001.
34. Hunter, AM, St Clair Gibson, A, Collins, M, Lambert, M, and
Noakes, TD. Caffeine ingestion does not alter performance during
a 100-km cycling time-trial performance. Int J Sport Nutr Exerc
Metab 12: 438–452, 2002.
35. Ivy, JL, Costill, DL, Fink, WJ, and Lower, RW. Influence of caffeine
and carbohydrate feedings on endurance performance. Med Sci
Sports 11: 6–11, 1979.
36. Jacobson, TL, Febbraio, MA, Arkinstall, MJ, and Hawley, JA. Effect
of caffeine co-ingested with carbohydrate or fat on metabolism
and performance in endurance-trained men. Exp Physiol 86: 137–
144, 2001.
37. Jenkins, NT, Trilk, JL, Singhal, A, O’Connor, PJ, and Cureton, KJ.
Ergogenic effects of low doses of caffeine on cycling performance.
Int J Sport Nutr Exerc Metab 18:328-342, 2008.
38. Jeukendrup, AE and Currell, K. Should time trial performance be
predicted from three serial time-to-exhaustion tests? Med Sci Sports
Exerc 37: 1820, 2005.
39. Judelson, DA, Armstrong, LE, Sokmen, B, Roti, MW, Casa, DJ, and
Kellogg, MD. Effect of chronic caffeine intake on choice reaction
time, mood, and visual vigilance. Physiol Behav 85: 629–634, 2005.
40. Kalmar, JM and Cafarelli, E. Caffeine: a valuable tool to study central
fatigue in humans? Exerc Sport Sci Rev 32: 143–147, 2004.
41. Kamimori, GH, Karyekar, CS, Otterstetter, R, Cox, DS, Balkin, TJ,
Belenky, GL, and Eddington, ND. The rate of absorption and relative
bioavailability of caffeine administered in chewing gum versus capsules
to normal healthy volunteers. Int J Pharm 234: 159–167, 2002.
42. Kaplan, GB, Greenblatt, DJ, Kent, MA, and Cotreau-Bibbo, MM.
Caffeine treatment and withdrawal in mice: relationships between
dosage, concentrations, locomotor activity and A1 adenosine
receptor binding. J Pharmacol Exp Ther 266: 1563–1572, 1993.
43. Keisler, BD and Armsey, TD 2nd. Caffeine as an ergogenic aid. Curr
Sports Med Rep 5: 215–219, 2006.
44. Kovacs, EM, Stegen, J, and Brouns, F. Effect of caffeinated drinks on
substrate metabolism, caffeine excretion, and performance. J Appl
Physiol 85: 709–715, 1998.
45. Laursen, PB, Francis, GT, Abbiss, CR, Newton, MJ, and Nosaka, K.
Reliability of time-to-exhaustion versus time-trial running tests in
runners. Med Sci Sports Exerc 39: 1374–1379, 2007.
46. MacIntosh, BR and Wright, BM. Caffeine ingestion and perfor-
mance of a 1,500-metre swim. Can J Appl Physiol 20: 168–177, 1995.
47. Mangus, BC and Trowbridge, CA. Will caffeine work as an ergogenic
aid? The latest research. Athl Ther Today 10: 57–62, 2005.
48. McLellan, TM, Bell, DG, and Kamimori, GH. Caffeine improves
physical performance during 24 h of active wakefulness. Aviat Space
Environ Med 75: 666–672, 2004.
49. McLellan, TM, Kamimori, GH, Voss, DM, Bell, DG, Cole, KG, and
Johnson, D. Caffeine maintains vigilance and improves run times
during night operations for Special Forces. Aviat Space Environ Med
76: 647–654, 2005.
50. Mulrow, CD and Cook, DJ. Systematic Reviews: Synthesis of Best
Evidence for Health Care Decisions. Philadelphia: American College
Physicians, 1998.
51. O’Connor, PJ, Motl, RW, Broglio, SP, and Ely, MR. Dose-
dependent effect of caffeine on reducing leg muscle pain during
cycling exercise is unrelated to systolic blood pressure. Pain 109:
291–298, 2004.
52. Paluska, SA. Caffeine and exercise. Curr Sports Med Rep 2: 213–
219, 2003.
53. PEDro Scale. Available at: http://www.pedro.fhs.usyd.edu.au.
Accessed September 18, 2007.
54. Powers, SK and Dodd, S. Caffeine and endurance performance.
Sports Med 2: 165–174, 1985.
55. Powers, SK and Howley, ET. Exercise Physiology (4th ed.). Boston:
McGraw Hill, 2001. pp. 22–45.
56. Reents, S. Caffeine. In: Sport and Exercise Pharmacology. Champaign:
Human Kinetics, 2000. pp. 277–295.
57. Sasaki, H, Maeda, J, Usui, S, and Ishiko, T. Effect of sucrose and
caffeine ingestion on performance of prolonged strenuous running.
Int J Sports Med 8: 261–265, 1987.
58. Sawka, MN, Burke, LM, Eichner, ER, Maughan, RJ, Montain, SJ,
and Stachenfeld, NS. American College of Sports Medicine position
stand. Exercise and fluid replacement. Med Sci Sports Exerc 39: 377–
390, 2007.
59. Spriet, LL. Caffeine. In: Performance-Enhancing Substances in Sport and
Exercise. M. S. Bahrke and C. E. Yesalis, eds. New York: Human
Kinetics, 2002. pp. 267–278.
60. Tarnopolsky, MA. Caffeine and endurance performance. Sports Med
18: 109–125, 1994.
61. Van Nieuwenhoven, MA, Brouns, F, and Kovacs, EM. The effect of
two sports drinks and water on GI complaints and performance
during an 18-km run. Int J Sports Med 26: 281–285, 2005.
62. Van Soeren, MH and Graham, TE. Effect of caffeine on metabolism,
exercise endurance, and catecholamine responses after withdrawal.
J Appl Physiol 85: 1493–1501, 1998.
63. Weir, J, Noakes, TD, Myburgh, K, and Adams, B. A high
carbohydrate diet negates the metabolic effects of caffeine during
exercise. Med Sci Sports Exerc 19: 100–105, 1987.
64. Wemple, RD, Lamb, DR, and McKeever, KH. Caffeine vs caffeine-
free sports drinks: effects on urine production at rest and during
prolonged exercise. Int J Sports Med 18: 40–46, 1997.
65. Wiles, JD, Bird, SR, Hopkins, J, and Riley, M. Effect of caffeinated
coffee on running speed, respiratory factors, blood lactate and
perceived exertion during 1500-m treadmill running. Br J Sports Med
26: 116–120, 1992.
324
Journal of Strength and Conditioning Research
the
TM
Caffeine and Endurance Performance
Copyright © . N ational S trength and Conditioning A ssociation. Unauthorized reproduction of this article is prohibited