Journal of Strength and Conditioning Research, 2006, 20(4), 751–755

䉷 2006 National Strength & Conditioning Association

EFFECTS OF EIGHT WEEKS OF CAFFEINE

SUPPLEMENTATION AND ENDURANCE TRAINING ON
AEROBIC FITNESS AND BODY COMPOSITION
MOH H. MALEK,1 TERRY J. HOUSH,1 JARED W. COBURN,2 TRAVIS W. BECK,1
RICHARD J. SCHMIDT,1 DONA J. HOUSH,3 AND GLEN O. JOHNSON1
1
University of Nebraska-Lincoln Human Performance Laboratory, Department of Nutrition and Health Sciences,
Lincoln, Nebraska 68583; 2California State University-Fullerton, Division of Kinesiology and Health Sciences,
Fullerton, California 92834; 3University of Nebraska Medical Center, College of Dentistry, Department of Oral
Biology, Lincoln, Nebraska 68583.

ABSTRACT. Malek, M.H., T.J. Housh, J.W. Coburn, T.W. Beck, maté, guarana seed extract, and black tea extract each
R.J. Schmidt, D.J. Housh, and G.O. Johnson. Effects of eight contain caffeine, which increases metabolic rate and plas-
weeks of caffeine supplementation and endurance training on ma fatty acid concentrations (19, 22, 25). It has been sug-
aerobic fitness and body composition. J. Strength Cond. Res.
gested that the mechanism of action of these ingredients
20(4):751–755. 2006.—The purpose of this study was to examine
the effects of daily administration of a supplement that con- is related, in part, to stimulation of enzymes that control
tained caffeine in conjunction with 8 weeks of aerobic training lipolysis (2, 3, 20, 22). In addition, catechins in black tea
on V̇O2peak, time to running exhaustion at 90% V̇O2peak, body act synergistically with caffeine to increase metabolic rate
weight, and body composition. Thirty-six college students (14 and stimulate lipolysis (22). Ginger extract also may in-
men and 22 women; mean ⫾ SD, age 22.4 ⫾ 2.9 years) volun- crease metabolic rate directly, as well as indirectly, by
teered for this investigation and were randomized into either a stimulating the release of catecholamines from the adre-
placebo (n ⫽ 18) or supplement group (n ⫽ 18). The subjects nal medulla (9, 22). Schisandra chinensis fruit extract is
ingested 1 dose (3 pills ⫽ 201 mg of caffeine) of the placebo or a Chinese herbal preparation which has been suggested
supplement per day during the study period. In addition, the
to improve work performance and to reduce fatigue (18,
subjects performed treadmill running for 45 minutes at 75% of
the heart rate at V̇O2peak, three times per week for 8 weeks. All 22). Specifically, Hancke et al. (18) found that a single
subjects were tested pretraining and posttraining for V̇O2peak, dose of schisandra extract administered 30 minutes prior
time to running exhaustion (TRE) at 90% V̇O2peak, body weight to exercise reduced plasma lactate concentration levels
(BW), percentage body fat (%FAT), fat weight (FW), and fat-free and respiratory frequency. In preclinical studies, grape
weight (FFW). The results indicated that there were equivalent seed extract has been suggested to have antioxidant prop-
training-induced increases (p ⬍ 0.05) in V̇O2peak and TRE for erties (22). Vitablue, along with vitamin C and the B vi-
the supplement and placebo groups, but no changes (p ⬎ 0.05) tamins niacin, pyridoxine (B6), and pantothenic acid, are
in BW, %FAT, FW, or FFW for either group. These findings in- known to have metabolic as well as antioxidant properties
dicated that chronic use of the caffeine-containing supplement
in the present study, in conjunction with aerobic training, pro-
(22, 27).
vided no ergogenic effects as measured by V̇O2peak and TRE, Typically, the studies that have examined the efficacy
and the supplement was of no benefit for altering body weight of supplements containing these ingredients have tested
or body composition. parameters of aerobic fitness such as V̇ O2peak or time to
exhaustion following ingestion of a single dose of supple-
KEY WORDS. aerobic power, maximal oxygen uptake, nutrition,
supplements, treadmill exercise ment (4, 5, 7, 11, 15, 16). For example, Birnbaum and
Herbst (7) found that 7 mg·kg⫺1 of caffeine administered
1 hour prior to a 30-minute exercise bout corresponding
INTRODUCTION to 70% of V̇O2max improved respiratory efficiency. Col-
affeine is widely consumed in the United States lomp et al. (12), however, found that 5 days of caffeine

C and is derived from natural sources such as cof-

fee beans and tea leaves (13). In addition to its
traditional use in various foods and drinks, caf-
feine has been used as an ergogenic supplement to in-
crease energy metabolism and to improve exercise and
supplementation (250 mg·d⫺1) increased the catechol-
amine response to exercise, but did not affect cycling time
during a standardized test of V̇O2peak. Except for the
study by Collomp et al. (12), there is little information
available regarding the effects of caffeine in combination
sports performance, as well as to modify body composition with aerobic training on endurance exercise outcomes or
and body weight (14). A number of supplements that in- body composition. Thus, the current investigation was an
clude caffeine or other metabolically active ingredients extension of previous studies of the acute effects of caf-
such as yerba maté (Ilex paraguariensis), guarana seed feine on exercise performance and body composition, as
extract (Paullinia cupana), black tea extract (Camellia si- well as the study by Collomp et al. (12), who examined
nensis), Ginger extract (Zingiber officinale), Schisandra the effects of short-term (i.e., 5 days) administration of
chinensis, grape seed extract (Vitis vinifera), Vitablue caffeine on aerobic performance. Theoretically, chronic
(Vaccininium angustifolium), and various vitamins have caffeine supplementation and aerobic training may result
been examined for their potential effects on energy me- in metabolic and hormonal adaptations, including in-
tabolism and body composition (2, 22). For example, yerba creases in fatty acid utilization, that result in carbohy-

751
752 MALEK, HOUSH, COBURN ET AL.

TABLE 1. Ingredients contained in each of the supplement

tablets.*
Content
(mg)
Ingredient SUPP
Yerba maté (Ilex paraguariensis)
(8% caffeine ⫽ 40 mg) 500
Guarana seed extract (Paullinia cupana)
(36% caffeine ⫽ 152 mg) 422
Black tea extract (Camellia sinensis)
(9% caffeine ⫽ 9 mg) 100
Ginger extract (Zingiber officinale) (5% gingerols) 500
Schisandra chinensis fruit extract 100
Dill weed extract (Anethum graveolens) 5
FIGURE 1. Results of V̇O2peak (mean ⫾ SD) tests prior to Grape seed extract (Vitis vinifera) 1
(pretraining) and following (posttraining) 8 weeks of endurance Vitamin C (as ascorbic acid) 120
training. * p ⬍ 0.05, pretraining ⬍ posttraining for placebo Niacin 20
(PLAC) and supplement (SUPP) groups. Vitamin B6 (as pyridoxine hydrochloride) 2
Pantothenic acid (as calcium d-pantothenate) 10
Vitablue (Vaccininium angustifolium) 50
Cinnamon 25
* Total caffeine content for the supplement (SUPP) group was
201 mg per dose.

and in the morning prior to breakfast on nontraining

days. In addition, the subjects were instructed to main-
tain their normal diet during the 8-week period. Dietary
and caffeine intake were not controlled in this study in
order to mimic the situation that typically occurs when
an individual purchases nutritional supplements for per-
sonal use.
Subjects
FIGURE 2. Results of time to running exhaustion (mean ⫾ Thirty-six college students (14 men and 22 women; mean
SD) tests prior to (pretraining) and following (posttraining) 8 ⫾ SD, age ⫽ 22.4 ⫾ 2.9 years) who had been performing
weeks of endurance training. * p ⬍ 0.05, pretraining ⬍ post- less than 4 hours per week of regular, continuous aerobic
training for placebo (PLAC) and Supplement (SUPP) groups. exercise volunteered as subjects for this investigation.
Furthermore, each subject completed a caffeine consump-
tion questionnaire to determine their caffeine intake
drate sparing, and therefore, potentially influence aerobic
(mg·dl⫺1) prior to and following the 8-week study. All pro-
indices such as V̇O2peak and time to exhaustion. The pur-
cedures were approved by the University Institutional
pose of this study was to examine the effects of daily ad-
Review Board for Human Subjects and each subject
ministration of a supplement that contained a combina-
signed an informed consent prior to any testing.
tion of metabolically active ingredients, in conjunction
with 8 weeks of aerobic training on V̇O2peak, time to run- Body Composition Determination
ning exhaustion (TRE) at 90% V̇O2peak, body weight
(BW), and body composition. Body composition was assessed pre- and posttraining by
underwater weighing. The subjects were instructed to
METHODS avoid exercise for at least 24 hours prior to testing, and
each subject indicated that they were normally hydrated
Experimental Approach to the Problem and in a postabsorptive state (at least 4 hours) upon ar-
This study used a randomized, double-blinded, placebo- rival to the laboratory. BW was determined to the nearest
controlled, parallel design. All subjects participated in an 0.11 kg using a state-certified physician’s scale. Body den-
endurance training program that consisted of 45 minutes sity was assessed from underwater weighing, with cor-
of treadmill running at 75% of the heart rate at V̇O2peak rection for residual lung volume using the oxygen dilution
(HRpeak), three times per week for 8 weeks. The depen- method of Wilmore (26). Residual lung volume was de-
dent variables assessed pretraining and posttraining in termined on land with the subject seated in a position
this study were peak oxygen uptake (V̇O2peak), TRE at similar to that assumed during underwater weighing.
90% V̇O2peak, BW, percentage body fat (%FAT), fat The average of similar scores (within 0.1 L) from 2–3 tri-
weight (FW), and fat-free weight (FFW). The subjects als was used as the representative residual lung volume.
were assigned randomly to either the supplement (SUPP; Underwater weight was measured in a submersion tank
n ⫽ 18) or placebo (PLAC; n ⫽ 18) group. Table 1 contains in which a nylon swing seat was suspended from a 10-kg
the ingredients in the SUPP tablets. The placebo tablets Salter scale (model 230; REGO Designs & Patents, UK).
included microcrystalline cellulose. The subjects ingested The average of the 2–3 highest weights from 6–10 trials
1 dose (3 tablets ⫽ 1 dose) of supplement or placebo 60 was used as the representative underwater weight. Per-
minutes prior to their exercise session on training days centage body fat was calculated using the revised formula
 CAFFEINE AND ENDURANCE TRAINING 753

TABLE 2. Characteristics of subjects (mean ⫾ SD).*

PLAC (n ⫽ 18) SUPP (n ⫽ 18)
Variables Pretraining Posttraining ES (d)‡ Pretraining Posttraining ES (d)‡
Age (y) 22.7 ⫾ 3.1 — 22.0 ⫾ 2.7 —
V̇O2peak (ml·kg⫺1·min⫺1) 39.6 ⫾ 6.9† 42.9 ⫾ 7.3 0.47 41.2 ⫾ 7.3* 45.0 ⫾ 8.5 0.52
TRE (s) 681 ⫾ 295† 1,223 ⫾ 759 1.84 552 ⫾ 198* 1,034 ⫾ 542 2.43
BW (kg) 71.4 ⫾ 13.1 71.6 ⫾ 13.9 0.02 65.9 ⫾ 11.6 65.7 ⫾ 11.1 0.02
%FAT 25.2 ⫾ 8.8 24.5 ⫾ 7.2 ⫺0.08 22.9 ⫾ 7.8 22.1 ⫾ 7.3 ⫺0.10
FW (kg) 18.2 ⫾ 7.6 17.8 ⫾ 7.0 ⫺0.05 14.8 ⫾ 5.5 14.2 ⫾ 5.0 ⫺0.11
FFW (kg) 53.2 ⫾ 10.6 53.8 ⫾ 10.6 0.06 51.1 ⫾ 11.6 51.5 ⫾ 11.4 0.03
* PLAC ⫽ placebo group; SUPP ⫽ supplement group; TRE ⫽ time to running exhaustion; BW ⫽ body weight; %FAT ⫽ percentage
body fat; FW ⫽ fat weight; FFW ⫽ fat-free weight.
† p ⬍ 0.05 for pretraining vs. posttraining measures within each group.
‡ ES (d) ⫽ effect size in standard deviation units based on the formula [(Meanpost ⫺ Meanpre) / SDpre] (21).

of Brozek et al. (10), with FW and FFW derived mathe- minute warm-up at 4.83 km·h⫺1 and 0% grade, the sub-
matically. Previous test-retest reliability data for under- jects ran for 45 minutes at 75% of the heart rate at
water weighing from our laboratory indicated that for V̇O2peak attained during the pretraining V̇Opeak test.
young men (n ⫽ 16) measured 24–72 hours apart, the The subjects wore heart rate monitors and were super-
intraclass correlation coefficient (ICC) was 0.98. vised to assure the appropriate training intensity was
maintained throughout each of the 45-minute training
Determination of V̇o2peak and Peak Heart Rate sessions. As the fitness levels of the subjects improved,
Each subject performed an incremental test to exhaustion the treadmill velocities were increased to maintain the
on a motorized treadmill (Precor C962i, Woodinville, WA) desired training heart rates. The subjects were instructed
for the determination of V̇O2peak. Each subject wore a not to perform any additional exercise training for the
nose clip and breathed through a 2-way valve (2700; Hans duration of the study. In addition, subjects were allowed
Rudolph, Kansas City, MO). Expired gas samples were to consume water ad libitum during the training sessions.
collected and were analyzed using a calibrated TrueMax
Statistical Analyses
2400 metabolic cart (Parvo Medics, Sandy, UT) with O2,
CO2, and ventilatory parameters expressed as 20-second V̇O2peak, TRE, BW, %FAT, FW, and FFW were analyzed
averages. The metabolic cart was calibrated prior to each with 6 separate 2 ⫻ 2 (group [SUPP and PLAC] ⫻ time
test. Each subject was fitted with a Polar Heart Watch [pre- and posttraining]) mixed factorial analyses of vari-
system (Polar Electro Inc., Lake Success, NY) to monitor ance (ANOVAs). Statistical significance was set at p ⱕ
heart rate throughout the test. Following a 5-minute 0.05 and the data were analyzed using the SPSS software
warm-up at 4.83 km·h⫺1 and 0% grade, the test began (version 13.0; SPSS, Inc., Chicago, IL).
with the subject walking at 6.44 km·h⫺1 and 0% grade.
The velocity was increased 1.61 km·h⫺1 every 2 minutes RESULTS
to 14.49 km·h⫺1. At 14.49 km·h⫺1, the exercise intensity Descriptive statistics of the subjects are presented in Ta-
was increased by raising the treadmill grade 2% every 2 ble 2. The caffeine consumption for the subjects prior to
minutes until voluntary exhaustion. V̇ O2peak and the study ranged between 0 and 243 mg·d⫺1. In addition,
HRpeak were defined as the highest values recorded dur- there were no significant differences for caffeine con-
ing the last 30 seconds of the test. Assessment of V̇O2peak sumption between the two groups at pre- and posttrain-
have been shown to be highly reliable, with an ICC ⬎ ing. The 2-way mixed factorial ANOVAs for V̇O2peak and
0.80 (23). TRE resulted in no significant time ⫻ group interactions
or main effects for group, but there were significant main
Time to Running Exhaustion effects for time. Therefore, V̇O2peak and TRE increased
With 48 hours of rest following the body composition and similarly as a result of the training for both the SUPP
V̇O2peak tests, each subject completed a constant velocity and PLAC groups (Figures 1 and 2). The 2-way mixed
treadmill run test. Following a 5-minute warm-up at 4.83 factorial ANOVAs for BW, %FAT, FW, and FFW resulted
km·h⫺1 and 0% grade, the subject began running at a ve- in no significant time ⫻ group interactions, main effects
locity that corresponded to 90% of the velocity at V̇O2peak for group, or main effects for time. Thus, there were no
as determined during the pretraining V̇O2peak test. Each training-induced changes in BW, %FAT, FW, and FFW
subject was instructed to run until voluntary exhaustion for either group.
and strong verbal encouragement was given. Billat et al.
(6) reported there were no significant (p ⬍ 0.05) differ- DISCUSSION
ences (test 1 ⫽ 404 ⫾ 101 seconds vs. test 2 ⫽ 402 ⫾ 113 The results of the present study indicated that there were
seconds) between repeated (1 week apart) treadmill runs no differences between the PLAC and SUPP groups for
to exhaustion at the maximal aerobic speed, and they the training-induced increases in V̇O2peak and TRE. A
were highly correlated at r ⫽ 0.864. prominent ingredient in the supplement tested in the pre-
sent study was caffeine (approximately 3.1 mg·kg⫺1), and
Endurance Training Exercise
there are no data available regarding the effects of chron-
All subjects visited the laboratory 3 times per week (Mon- ic caffeine supplementation on the physiological adapta-
day, Wednesday, and Friday) for 8 weeks. Following a 5- tions to aerobic training with which to compare the cur-
754 MALEK, HOUSH, COBURN ET AL.

rent findings. Most previous studies have examined the dicated that supplementation with yerba maté and gua-
effects of a single administration of caffeine in various rana resulted in a mean loss of BW of 5.1 kg, whereas the
dosages on acute exercise performance (4, 5, 11, 15, 16). placebo group lost a mean of only 0.3 kg of BW. The stud-
For example, Cadarette et al. (11) examined the effects of ies by Boozer et al. (8) and Andersen and Fogh (1) in-
4 dosages of caffeine (i.e., 0, 2.2, 4.4, 8.8 mg·kg⫺1) admin- volved overweight or obese subjects, whereas the subjects
istered 1 hour prior to exercise on time to exhaustion at in the present study ranged in %FAT from 8–41%. In ad-
a running velocity of 80% of V̇O2peak. The investigators dition, the present study combined supplementation with
(11) found that only the 4.4 mg·kg⫺1 dosage significantly exercise training, but the studies by Boozer et al. (8) and
increased time to exhaustion (53.4 vs. 73.4 min). These Andersen and Fogh (1) did not. Perhaps the differences
findings, however, were not consistent with those of Gra- between studies were due to the characteristics of the
ham and Spriet (16), who found that both 3.0 and 6.0 subjects involved, dosages of the various ingredients in
mg·kg⫺1 of caffeine administered 1 hour before exercise the supplements, and the inclusion of exercise training in
significantly increased time to exhaustion by 22% at a conjunction with supplementation.
running velocity of 85% of V̇O2peak. Supplementation
with 9.0 mg·kg⫺1 of caffeine, however, had no effect on PRACTICAL APPLICATION
the time to exhaustion (16). Bell and McLellan (4) com-
pared the effects of caffeine on time to exhaustion during This investigation was the first to examine the effects of
cycle ergometry in caffeine users and nonusers at a power the combination of ingredients in the supplement tested
output of 80% of V̇O2peak at 1, 3, and 6 hours following in the present study, together with 8 weeks of endurance
caffeine (5.0 mg·kg⫺1) or placebo supplementation. The in- training, on aerobic fitness, BW, and body composition.
vestigators (4) found that time to exhaustion was signif- The results indicated that there were no differences be-
icantly greater with caffeine than the placebo at 1, 3, and tween the PLAC and SUPP groups for the training-in-
6 hours for the caffeine non-users. For the caffeine users, duced increases in V̇O2peak and TRE, and that the sup-
however, time to exhaustion was greater following caf- plement had no effect on BW and body composition. These
feine ingestion at 1 and 3 hours only (4). Bell and Mc- findings suggested that chronic use of the supplement in
Lellan (4) concluded that sensitivity to caffeine likely ac- the present study, in conjunction with aerobic training,
counted for the differences in responses between the caf- provided no ergogenic effect as measured by V̇O2peak and
feine users and nonusers. In the present study, our sam- TRE, and the supplement was of no benefit for altering
ple had an average pretraining caffeine intake of 64.8 BW or body composition in the current sample of subjects.
mg·d⫺1, which is below the national average (199 mg·d⫺1) Previous research (1, 8, 9, 17, 19, 22, 24, 25) has sug-
reported for men and women between the ages of 18 to gested that the ingredients in the supplement tested in
34 (13). The questionnaire used in the present study, the present study can positively affect acute exercise per-
however, focused on caffeine-containing beverages only, formance and future studies should consider using higher
whereas the national survey (13) included both food and dosages of the ingredients to determine the effects on aer-
beverage categories. Therefore, it is likely that the survey obic training adaptations, BW, and body composition pa-
used in the present study did not account for the subject’s rameters.
entire daily caffeine intake. Based on the results of the
present study, as well as those from previous studies of REFERENCES
the effects of caffeine ingestion on acute exercise perfor- 1. ANDERSEN, T., AND J. FOGH. Weight loss and delayed gastric
mance, it is recommended that future studies examine emptying following a South American herbal preparation in
the effects of caffeine supplementation at a variety of dos- overweight patients. J. Hum. Nutr. Diet. 14:243–250. 2001.
ages on the physiological adaptations to endurance ex- 2. ANTONIO, J., AND J.R. STOUT. Sports Supplements. Philadel-
ercise training in caffeine users and nonusers. phia: Lippincott Williams & Wilkins, 2001. p. 329.
The results of the present study also indicated that 3. ANTONIO, J., AND J.R. STOUT. Supplements for Endurance Ath-
there were no training-induced changes in BW, %FAT, letes. New York: Human Kinetics, 2002. p. 127.
FW, or FFW for the SUPP or PLAC group. It has been 4. BELL, D.G., AND T.M. MCLELLAN. Exercise endurance 1, 3, and
suggested that the caffeine and other ingredients (Table 6 h after caffeine ingestion in caffeine users and nonusers. J.
Appl. Physiol. 93:1227–1234. 2002.
1) in the supplement tested in the present study may in-
5. BELL, D.G., AND T.M. MCLELLAN. Effect of repeated caffeine
crease energy metabolism and affect body composition ingestion on repeated exhaustive exercise endurance. Med. Sci.
and BW (1, 8, 9, 17, 19, 22, 24, 25). For example, Haller Sports Exerc. 35:1348–1354. 2003.
et al. (17) reported that 200 mg of caffeine increased plas- 6. BILLAT, V., J.C. RENOUX, J. PINOTEAU, B. PETIT, AND J.P. KOR-
ma free fatty acid concentration by approximately four- ALSZTEIN. Reproducibility of running time to exhaustion at
fold in the first 3 hours following ingestion in healthy V̇o2max in subelite runners. Med. Sci. Sports Exerc. 26:254–
young adults. Furthermore, Boozer et al. (8) examined 257. 1994.
the effects of Ma Huang-Guarana (72 mg·d⫺1 ephedrine 7. BIRNBAUM, L.J., AND J.D. HERBST. Physiologic effects of caf-
alkaloids and 240 mg·d⫺1 caffeine) on BW in 67 obese in- feine on cross-country runners. J. Strength Cond. Res. 18:463–
dividuals. The investigators found that the supplement 465. 2004.
group significantly decreased BW (4.0 ⫾ 3.4 kg) and 8. BOOZER, C.N., J.A. NASSER, S.B. HEYMSFIELD, V. WANG, G.
CHEN, AND J.L. SOLOMON. An herbal supplement containing
%FAT (2.1 ⫾ 3.0%), when compared with the placebo
Ma Huang-Guarana for weight loss: A randomized, double-
group (0.8 ⫾ 2.4 kg and 0.2 ⫾ 2.3%) (8). Similarly, An- blind trial. Int. J. Obes. Relat. Metab. Disord. 25:316–324. 2001.
dersen and Fogh (1) examined the effect of a supplement 9. BORDIA, A., S.K. VERMA, AND K.C. SRIVASTAVA. Effect of ginger
that contained both yerba maté (112 mg) and guarana (95 (Zingiber officinale Rosc.) and fenugreek (Trigonella foenum-
mg) on BW loss in overweight subjects. Forty-seven sub- graecum L.) on blood lipids, blood sugar, and platelet aggre-
jects were divided into 2 groups that received the supple- gation in patients with coronary artery disease. Prostaglandins
ment or placebo for a 45-day period. The results (1) in- Leukot. Essent. Fatty Acids 56:379–384. 1997.
 CAFFEINE AND ENDURANCE TRAINING 755

10. BROZEK, J., F. GRANDE, J.T. ANDERSON, AND A. KEYS. Densi- 20. LAURENT, D., K.E. SCHNEIDER, W.K. PRUSACZYK, C. FRANKLIN,
tometric analysis of body composition: Revision of some quan- S.M. VOGEL, M. KRSSAK, K.F. PETERSEN, H.W. GOFORTH, AND
titative assumptions. Ann. N. Y. Acad. Sci. 110:113–140. 1963. G.I. SHULMAN. Effects of caffeine on muscle glycogen utiliza-
11. CADARETTE, B.S., L. LEVIN, AND C.L. BERUBE. Effects of varied tion and the neuroendocrine axis during exercise. J. Clin. En-
dosages of caffeine on endurance exercise to fatigue. In: Bio- docrinol. Metab. 85:2170–2175. 2000.
chemistry of Exercise. H.G. Knuttgen, J.A. Vogel, and J.R. 21. LIPSEY, M.W., AND D.B. WILSON. Practical Meta-Analysis.
Poortmans, eds. Champaign, IL: Human Kinetics, 1983. p. 926. Thousand Oaks: Sage Publication, 2001. pp. 34–208.
12. COLLOMP, K., C. CAILLAUD, M. AUDRAN, J.L. CHANAL, AND C. 22. MCKENNA, D.J., K. JONES, AND K. HUGHES. Botanical Medi-
PREFAUT. Effect of acute or chronic administration of caffeine cines: The Desk Reference for Major Herbal Supplements (2nd
on performance and on catecholamines during maximal cycle ed.). New York: Haworth Herbal Press, 2002. p. 1138.
ergometer exercise. C. R. Seances Soc. Biol. Fil. 184:87–92. 23. MORROW, J.R., A.W. JACKSON, J.G. DISCH, AND D.P. MOOD.
1990. Measurement and Evaluation in Human Performance (2nd ed.).
13. FRARY, C.D., R.K. JOHNSON, AND M.Q. WANG. Food sources and Champaign, IL: Human Kinetics, 2000. pp. 223–271.
intakes of caffeine in the diets of persons in the United States. 24. MURASE, T., S. HARAMIZU, A. SHIMOTOYODOME, A. NAGASAWA,
J. Am. Diet. Assoc. 105:110–113. 2005. AND I. TOKIMITSU. Green tea extract improves endurance ca-

14. GRAHAM, T.E. Caffeine and exercise: Metabolism, endurance, pacity and increases muscle lipid oxidation in mice. Am. J. Phy-
and performance. Sports Med. 31:785–807. 2001. siol. Regul. Integr. Comp. Physiol. 288:R708–R715. 2004.
25. SCHINELLA, G.R., G. TROIANI, V. DAVILA, P.M. DEBUSHIAZZO,
15. GRAHAM, T.E., E. HIBBERT, AND P. SATHASIVAM. Metabolic and
AND H.A. TOURNIER. Antioxidant effects of an aqueous extract
exercise endurance effects of coffee and caffeine ingestion. J.
of Ilex paraguariensis. Biochem. Biophys. Res. Commun. 269:
Appl. Physiol. 85:883–889. 1998.
357–360. 2000.
16. GRAHAM, T.E., AND L.L. SPRIET. Metabolic, catecholamine, and
26. WILMORE, J.H. A simplified method for determination of resid-
exercise performance responses to various doses of caffeine. J. ual lung volumes. J. Appl. Physiol. 27:96–100. 1969.
Appl. Physiol. 78:867–874. 1995. 27. WU, X., G. CAO, AND R.L. PRIOR. Absorption and metabolism
17. HALLER, C.A., P. JACOB III, AND N.L. BENOWITZ. Enhanced of anthocyanins in elderly women after consumption of elder-
stimulant and metabolic effects of combined ephedrine and caf- berry or blueberry. J. Nutr. 132:1865–1871. 2002.
feine. Clin. Pharmacol. Ther. 75:259–273. 2004.
18. HANCKE, J.L., R. BURGOS, G. WIKMAN, E. EWERTZ, AND F. AHU- Acknowledgments
MADA. Schizandra chinensis, a potential phytodrug for recovery This study was funded by a research grant from General Nutri-
of sports horses. Fitoterapia. 65:113–118. 1994. tion Corp.
19. KHOKHAR, S., AND S.G. MAGNUSDOTTIR. Total phenol, catechin,
and caffeine contents of teas commonly consumed in the United Address correspondence to Moh H. Malek, mmalek@
Kingdom. J. Agric. Food Chem. 50:565–570. 2002. unlserve.unl.edu.