Content uploaded by Arnold G Nelson
Author content
All content in this area was uploaded by Arnold G Nelson
Content may be subject to copyright.
ORIGINAL ARTICLE
Arnold G. Nelson áRandy Day
Ellen L. Glickman-Weiss áMaren Hegsted
Joke Kokkonen áB. Sampson
Creatine supplementation alters the response
to a graded cycle ergometer test
Accepted: 20 April 2000
Abstract To determine the eects of creatine supple-
mentation on cardiorespiratory responses during a
graded exercise test (GXT) 36 trained adults (20 male, 16
female; 21±27 years old) performed two maximal GXTs
on a cycle ergometer. The ®rst GXT was done in a non-
supplemented condition, and the second GXT was done
following 7 days of ingesting either 5 g creatine mono-
hydrate, encased in gelatin capsules, four times daily
(CS, 13 male, 6 female), or the same number of glucose
capsules (PL, 7 male, 10 female). CS signi®cantly
(P< 0.05) improved total test time [pre-CS 1217
(240) s, mean (std. dev.) versus post-CS 1289 (215) s],
while PL administration had no eect (P> 0.05) on
total test time [pre-PL 1037 (181) s versus post-
PL 1047 (172) s]. In addition, both oxygen con-
sumption (
_
VO2) and heart rate at the end of each of the
®rst ®ve GXT stages were signi®cantly lower after CS,
but were unchanged after PL. Moreover, the ventilatory
threshold occurred at a signi®cantly greater
_
VO2for CS
[pre-CS 2.2 (0.4) l ámin
)1
or 66% of peak
_
VO2versus
post-CS 2.6 (0.5) l ámin
)1
or 78% of peak
_
VO2; pre-
PL 2.6 (0.9) l ámin
)1
or 70% peak
_
VO2versus post-
PL 2.6 (1.1) l ámin
)1
or 68% of peak
_
VO2]. Neither CS
nor PL had an eect on peak
_
VO2[pre-CS 3.4
(0.7) l ámin
)1
versus post-CS 3.3 (0.7) l ámin
)1
; pre-
PL 3.7 (1.1) l ámin
)1
versus post-PL 3.7 (1.1) l á
min
)1
]. Apparently, CS can alter the contributions of the
dierent metabolic systems during the initial stages of a
GXT. Thus, the body is able to perform the sub-maxi-
mal workloads at a lower oxygen cost with a concomi-
tant reduction in the work performed by the
cardiovascular system.
Key words Heart rate áOxygen consumption á
Ventilatory threshold
Introduction
Recent research has shown that oral creatine supple-
mentation can increase skeletal muscle phosphocreatine
(PC) levels (Balsom et al. 1995; Febbraio et al. 1995;
Greenha et al. 1994; Harris et al. 1992), and that such
supplementation has ergogenic potential (Balsom et al.
1993, 1995; Greenha et al. 1993). For example,
Greenha et al. (1993) showed that following creatine
supplementation, peak torque production was higher in
bouts 2, 3, and 4 of a work protocol consisting of 5
bouts of 30 maximum voluntary isokinetic contractions
interspersed with 60 s rest. Balsom et al. (1993) showed
that following creatine supplementation the power out-
put of a single high-intensity 10-s exercise period pre-
ceded by ®ve 6-s work bouts (each separated by 30 s
rest) was greater than it had been prior to creatine
supplementation. Later, both Balsom et al. (1995) and
Prevost et al. (1997) reported that creatine supplemen-
tation signi®cantly extends one's capacity to maintain a
speci®c level of high-intensity intermittent exercise.
These post-creatine supplementation performance im-
provements have also been associated with a 2±5%
Eur J Appl Physiol (2000) 83: 89±94 ÓSpringer-Verlag 2000
A. G. Nelson (&)
Department of Kinesiology, Louisiana State University,
112 Long Field House, Louisiana State University,
Baton Rouge, LA 70803, USA
e-mail: anelso@lsu.edu
Tel.: +1-225-388-3114, Fax: +1-225-388-3680
R. Day áB. Sampson
Division of Math and Science,
Brigham Young University-Hawaii,
Laie, HI 96762, USA
E. L. Glickman-Weiss
School of Exercise, Leisure and Sport,
Kent State University,
Kent, OH 44242, USA
M. Hegsted
Department of Human Ecology,
Louisiana State University,
Baton Rouge, LA 70803, USA
J. Kokkonen
Division of Physical Education,
Brigham Young University-Hawaii,
Laie, HI 96762, USA
decrease in lactate accumulation (Balsom et al. 1993,
1995; Febbraio et al. 1995; Prevost et al. 1997), a 2±3%
lower oxygen consumption (
_
VO2) (Balsom et al. 1993;
Prevost et al., unpublished observations), a 2% lower
plasma hypoxanthine accumulation (Balsom et al.
1993), and a 2±12% lower accumulation of ammonia
(Birch et al. 1994; Greenha et al. 1993; Mujika et al.
1996).
Lower levels of blood lactate, ammonia, and
_
VO2
during high-intensity work following a creatine supple-
mentation program suggest that creatine supplementa-
tion might alter physiological and/or metabolic
responses to a graded exercise test (GXT). Presently,
there is little information concerning the eects of cre-
atine supplementation on responses measured during a
GXT. Stroud et al. (1994) found that steady-state
_
VO2,
heart rate, and blood lactate concentration during seven
stages (6 min per stage) of incremental treadmill exercise
did not change following creatine supplementation.
Unfortunately, the experiment was terminated at 90%
of pre-supplementation maximal
_
VO2, so it is uncertain
whether maximal
_
VO2changed following supplementa-
tion. On the other hand, Gordon et al. (1995) discovered
that the performance of chronic heart failure patients on
a progressive incremental (10-W increase per minute)
cycle ergometer test increased by 10% following creatine
supplementation, but no
_
VO2values were reported.
This disagreement between Stroud et al. (1994) and
Gordon et al. (1995) might be due to the dierent
lengths of the exercise stages (6 min versus 1 min). While
6-min stages are recommended for obtaining steady-
state information for each stage of exercise (A
Êstrand and
Rodahl 1986), a GXT consisting of 2- or 3-min stages is
more frequently used. For some time it has been estab-
lished (Cotes et al. 1969) that the
_
VO2measured at each
stage of a GXT consisting of 3-min stages is lower than
that measured at each stage of a GXT comprised of
longer stages (>5 min). This lower
_
VO2is attributed, in
part, to a delay in adjustment by the cardiorespiratory
system to the changes in workload. During this delay,
the metabolic demands of the work are met by the ATP-
PC and glycolytic systems (A
Êstrand and Rodahl 1986;
Karlsson 1971; Sahlin et al. 1988; Scott and Bogdany
1998).
Interestingly, Paganelli et al. (1989) have shown that
a decrease in PC stores results in a higher
_
VO2at the
onset of exercise. Consequently, the capacity of the
ATP-PC system could in¯uence the
_
VO2response to a
short-stage GXT, since lower PC stores would necessi-
tate more reliance on aerobic metabolism. On the other
hand, during a short-stage GXT increasing the muscle
PC stores via a standard creatine supplementation pro-
gram could further delay the reliance upon aerobic
metabolism, and thus decrease the per-stage
_
VO2. Hence,
the purpose of this study was to characterize more
completely the eect of creatine supplementation upon
responses to a 3-min per-stage GXT. We hypothesized
that, following creatine supplementation,
_
VO2, and heart
rate would be lower at each stage of the GXT.
Methods
Participants
Thirty-six physically active healthy (male, n20, female, n16)
university students successfully completed all requirements of the
study, and their mean physical characteristics are presented in
Table 1. A physically active individual was de®ned as a person
whose peak
_
VO2from the initial GXT placed the individual into
either the good or excellent categories (males:
_
VO2max >43 ml á
kg
)1
ámin
)1
; females:
_
VO2max >38 ml ákg
)1
ámin
)1
) established by
the American Heart Association (1972). In addition, each partici-
pant had to have been engaged in lower-extremity rhythmic aerobic
activity for a minimum period of 20 min 3 days per week for at
least 3 months prior to the start of the study. The experimental
protocol was approved by the appropriate institutional review
board, and each volunteer gave written consent prior to performing
any part of the experiment.
Experimental protocol
All participants performed two GXTs to exhaustion on a Monark
cycle ergometer (model 818 E). The ®rst GXT was done while the
subject was in a non-supplemented condition. Following the ®rst
GXT, the subjects were randomly placed into either a creatine
supplementation group (CS, 13 male, 6 female) or a placebo group
(PL, 7 male, 10 female). Initially each group consisted of 15 males
and 10 females; however, non-compliance with study requirements
necessitated the removal of 14 individuals. Hence, at the end of the
study, CS had 13 males and 6 females, while PL had 7 males and 10
females.
Each GXT began at a work rate of 30 W (90 rev ámin
)1
against
a resistance of 3.24 N). The work rate was incremented by 30 W
every 3 min until the subject could no longer maintain the required
pedal cadence. Expired gases and minute ventilations (V
e
) were
monitored continuously with an automated system (SensorMedics
series 2500). Oxygen consumption was measured breath-by-breath,
and was averaged at 20-s time periods. Peak
_
VO2was
de®ned as the highest 20-s average
_
VO2obtained during the last 4 min
of the test. In addition, heart rate was recorded during the ®nal 15 s
of the 3rd min of each stage. To ensure unbiased results, each GXT
was administered by three to four technicians who were naive both
to the supplementation protocol, and to the results of the ®rst GXT.
The test administers were asked to give each participant maximal
Table 1 Subject characteristics.
Values are mean (SD). (CS
Creatine supplemented, PL
placebo)
Group Height (cm) Mass (kg) Age (years)
_
VO2peak (ml ákg
)1
ámin
)1
)
Mean SD Mean SD Mean SD Mean SD
CS male 187 9 90 17 24 1 47.7 7.6
CS female 162 4 61 10 23 2 45.0 5.7
PL male 178 7 79 10 25 2 51.5 9.0
PL female 167 5 63 8 24 2 45.2 8.1
90
encouragement to perform to exhaustion. At least one of the co-
authors was present at each test to assure that the test protocol was
strictly followed; but, to prevent investigator bias, the second GXT
was overseen by a co-author who was not present at the ®rst GXT,
and was unaware of the participant's prior performance.
After each GXT was completed, the 20-s outputs were used to
calculate T
vent
.T
vent
was de®ned as the
_
VO2(l ámin
)1
) at the in-
tersection point of the high and low slope portions of the
_
VE:
_
VO2
relationship and calculated according to the procedure outlined by
Orr et al. (1982). In brief, the intersection point of the high and low
slope portions of the
_
VE:
_
VO2relationship was calculated by deter-
mining the least-squares regression lines for each portion of the
_
VE:
_
VO2relationship.
Supplementation
Supplementation began on the day following the ®rst GXT, and
consisted of either placebo or creatine. Placebo administration
consisted of sugar encased in gelatin capsules. Creatine supple-
mentation consisted of 5 g creatine monohydrate encased in gelatin
capsules administered four times daily for 7 days. The placebo and
creatine capsules were distributed following a double-blind proto-
col, and were indistinguishable in appearance. The number of
placebo and creatine capsules ingested daily was the same.
Statistical analysis
Statistical comparisons between the creatine and placebo condi-
tions for peak
_
VO2, ventilatory breakpoint, and total work time
were made via a two-way (treatment ´pre-post) repeated-measures
ANOVA. A three-way (treatment ´pre-post ´GXT stage) re-
peated-measures ANOVA was used to analyze stage-by-stage
changes in
_
VO2and heart rate. The major term of interest in the
statistical models was either the two-way (treatment ´pre-post) or
the three-way (treatment ´pre-post ´GXT stage) interaction.
Post-ANOVA analyses of the signi®cant interactions involved the
use of Tukey's Protected t-test. The experimental error rate was set
at 0.05, and was maintained throughout all statistical tests.
Results
Maximal values
The initial peak
_
VO2was 3.41 (0.68) l ámin
)1
[mean
(SD)] for CS and 3.69 (1.06) l ámin
)1
for PL. Following
CS and PL, peak
_
VO2[CS 3.33 (0.72) l ámin
)1
;
PL 3.74(1.07) l ámin
)1
] remained unchanged (P
0.17 for the two-way interaction). Likewise, the maxi-
mum heart rate for both CS [pre 189 (8) bpm,
post 186 (4) bpm] and PL [pre 189 (10) bpm, 189
(12) bpm] was not signi®cantly changed (P0.25 for
the two-way interaction). However, a signi®cant
(P0.01) two-way interaction was found for total test
time. Post-hoc analysis revealed that CS signi®cantly
improved (P< 0.05) total test time by more than 70 s
[pre 1217 (240) s, post 1289 (215) s]. In contrast,
PL had no eect (P> 0.05) on total test time
[pre 1038 (181) s, post 1047 (172) s].
Sub-maximal values
All participants completed at least four stages during
both the ®rst and second GXT. Therefore, stage-by-stage
comparisons of
_
VO2and heart rate were made for stag-
es 1±4. For each stage, the last three 20-s readings for
_
VO2
were averaged, and the average used in the analyses. The
_
VO2responses are depicted in Figures 1 and 2. Fig. 1
shows the attenuating eect of the creatine supplemen-
tation on
_
VO2across stages 1±4 of the GXTs. The three-
way interaction for
_
VO2was signi®cant (P0.03). For
each stage, the post-CS
_
VO2was signi®cantly lower
(P< 0.05) than the pre-CS
_
VO2. In contrast, Fig. 2
shows that the pre-PL and post-PL
_
VO2did not dier
(P> 0.05) across stages. Likewise, the three-way inter-
action for heart rate was signi®cant (P0.03). Within
the CS group, post-supplementation heart rates were
signi®cantly (P< 0.05) lower at each stage. This dier-
ence was approximately 6 bpm (see Fig. 3). Heart rates
within PL were not signi®cantly dierent (P> 0.05)
between pre and post, at any stage (see Fig. 4).
The two-way interaction for T
vent
was signi®cant for
both absolute
_
VO2(P0.02) (see Fig. 5) and relative
_
VO2
(P0.02) (see Fig. 6). Again, post-hoc analysis revealed
that both measures of T
vent
were signi®cantly improved
(P< 0.05) following CS, but were unaected by PL.
Discussion
It is apparent that CS altered the contributions of the
dierent metabolic systems during these short-stage cy-
cle ergometer GXTs. This alteration resulted in the body
being able to perform a given sub-maximal workload at
a lower oxygen cost coupled with reduced work by the
cardiovascular system. Presumably, the reductions in the
sub-maximal
_
VO2and heart rate are due to increased PC
concentration within the muscle cells. Tissue oxygen
utilization is tied to mitochondrial respiration, and a
Fig. 1 The eect of creatine supplementation on oxygen consump-
tion
_
VO2during the graded exercise test (GXT). Values represented
are mean (SD). *The pre-supplemented mean
_
VO2at each GXT
stage is signi®cantly higher (P< 0.05) than the post-supplemented
mean
_
VO2at that stage
91
decrease in ATP coupled with an increase in ADP serves
as a stimulator of mitochondrial respiration (Balaban
1990; Scott 1995). Hence, an increase in muscle PC levels
due to creatine supplementation would have delayed the
decrease in the ATP/ADP ratio needed to stimulate
mitochondrial respiration and
_
VO2. Support for this idea
comes from the work of Connett and Honig (1989).
These authors showed that during in vivo stimulation of
muscle,
_
VO2was linearly related to PC concentration
signi®cantly. When PC concentrations were high,
_
VO2
was low, and vice versa.
While an increase in PC can yield a decrease in
_
VO2,
it should be noted that the lower
_
VO2of approxi-
mately 0.25 l ámin
)1
corresponds to approximately
60±70 mmol ATP. Since the usual increase in the total
creatine pool ranges from 15 to 35 mmol ákg
)1
dry
matter (Greenha et al. 1994; Harris et al. 1992), it is
unlikely that the increase in PC stores was the only
factor responsible for attenuating the
_
VO2response. In-
creased mechanical eciency could also have contrib-
uted to the decrease in
_
VO2. Pahud et al. (1980) reported
that the eciency during the oxygen-de®cit period of
exercise was 33%. This eciency was a combination of
both the aerobic eciency of 27% and the anaerobic
eciency of 41%. Since the eciency of anaerobic ex-
ercise is about 50% greater than that of aerobic exercise,
an increase in the relative contribution of the anaerobic
Fig. 2 The eect of placebo administration on
_
VO2during the
GXT. Values represented are mean (SD)
Fig. 3 The eect of creatine supplementation on heart rate during
the GXT. Values represented are mean (SD). *The pre-supple-
mented mean heart rate at each GXT stage is signi®cantly higher
(P< 0.05) than the post-supplemented mean heart rate at that
stage
Fig. 4 The eect of placebo administration on heart rate during
the GXT. Values represented are mean (SD)
Fig. 5 Eect of creatine and placebo supplementation on the
absolute
_
VO2at ventilatory breakpoint. Values represented are
mean (SD). *The post-creatine supplemented mean absolute
_
VO2is
signi®cantly greater (P< 0.05) than the pre-supplemented abso-
lute
_
VO2
92
energy systems results in an increase in the ratio between
work output and ATP utilization.
The ®ndings of this study suggest that CS can modify
the relationship between pulmonary gas exchange and
heart rate responses during a GXT. This modi®cation
could have important practical rami®cations. For ex-
ample, one of the methods recommended by the Amer-
ican College of Sports Medicine (ACSM) for
determining an appropriate exercise intensity is to plot
heart rate obtained during a GXT against either
_
VO2or
workload to determine a training-sensitive zone with
respect to exercise intensity (American College of Sports
Medicine 1995). Since CS apparently elicits a lower heart
rate at sub-maximal workloads of a short-stage GXT,
the work intensity range determined would be overesti-
mated. In addition, many sub-maximal tests for pre-
dicting maximal
_
VO2base extrapolations on the heart
rate: workload relationship at one or more workloads
(American College of Sports Medicine 1995). Again, the
lower than expected heart rate following CS could result
in the maximal
_
VO2being overestimated. The use of this
overestimated maximal
_
VO2could result in the prescrip-
tion of an exercise intensity that is too high. Finally,
ASCM also recommends that the exercise intensity for
pulmonary patients should exceed the anaerobic
threshold (American College of Sports Medicine 1995).
Since almost all of the automated gas analysis systems
currently available predict an individual's anaerobic
threshold from the T
vent
obtained during the GXT, it is
not unlikely that some clinicians might use this measure
to determine exercise intensity. Again, this study
demonstrates that the use of a GXT to obtain a T
vent
-
based prediction of anaerobic threshold would be
contraindicated in a CS individual.
In summary, creatine loading alters the initial meta-
bolic responses seen during a short-stage GXT. These
alterations are most signi®cant at the early stages of the
GXT and are manifested by a lower sub-maximal
_
VO2
and heart rate at the end of each GXT stage.
Acknowledgement This experiment complied with all current state
and federal laws in force at the time it was conducted.
References
American College of Sports Medicine (1995) ACSM's guidelines
for exercise testing and prescription. Williams and Wilkins,
Baltimore
American Heart Association (1972) Exercise testing and training of
apparently healthy individuals: a handbook for physicians.
American Heart Association, New York
A
Êstrand PO, Rodahl K (1986) Textbook of work physiology:
physiological bases of exercise. McGraw-Hill, New York
Balaban RS (1990) Regulation of oxidative phosphorylation in the
mammalian cell. Am J Physiol 258: C377±C389
Balsom PD, Ekblom B, So
Èderlund K, Sjo
Èdin B, Hultman E (1993)
Creatine supplementation and dynamic high-intensity inter-
mittent exercise. Scand J Med Sci Sports 3: 143±149
Balsom PD, So
Èderlund K, Sjo
Èdin B, Ekblom B (1995) Skeletal
muscle metabolism during short duration high-intensity exer-
cise: in¯uence of creatine supplementation. Acta Physiol Scand
154: 303±310
Birch R, Noble D, Greenha PL (1994) The in¯uence of dietary
creatine supplementation on performance during repeated
bouts of maximal isokinetic cycling in man. Eur J Appl Physiol
69: 268±270
Connett RJ, Honig CR (1989) Regulation of
_
VO2max in red muscle:
do current biochemical hypotheses ®t in vivo data? Am J
Physiol 256: R898±R908
Cotes JE, Allsopp D, Sardi F (1969) Human cardiopulmonary
response to exercise: comparisons between progressive and
steady state exercise, between arm and leg exercise, and between
subjects diering in body weight. Q J Exp Physiol Cogn Med
Sci 54: 211±222
Febbraio MA, Flanagan TR, Snow RJ, Zhao S, Carey MF
(1995) Eect of creatine supplementation on intramuscular
TCr, metabolism and performance during intermittent, su-
pramaximal exercise in humans. Acta Physiol Scand 155: 387±
395
Gordon A, Hultman E, Kaijser L, Kristjansson S, Rolf CJ, Nyquist
O, Sylven C (1995) Creatine supplementation in chronic heart
failure increases skeletal muscle creatine phosphate and muscle
performance. Cardiovasc Res 30: 413±418
Greenha PL, Casey A, Short AH, Harris R, So
Èderlund K, Hult-
man E (1993) In¯uence of oral creatine supplementation on
muscle torque during repeated bouts of maximal voluntary
exercise in man. Clin Sci 84: 565±571
Greenha PL, Bodin K, So
Èderlund K, Hultman E (1994) Eect of
oral creatine supplementation on skeletal muscle phosphocre-
atine resynthesis. Am J Physiol 266: E725±E730
Harris RC, So
Èderlund K, Hultman E (1992) Elevation of creatine
in resting and exercised muscle of normal subjects by creatine
supplementation. Clin Sci 83: 367±374
Karlsson J (1971) Lactate and phosphagen concentrations in
working muscle of man with special reference to oxygen de®cit
at the onset of work. Acta Physiol Scand Suppl 358: 1±72
Mujika I, Chatard JC, Lacoste L, Barale F, Geyssant A (1996)
Creatine supplementation does not improve sprint perfor-
mance in competitive swimmers. Med Sci Sports Exerc
28: 1435±1441
Orr GW, Green HJ, Hughson RL, Bennett GW (1982) A computer
linear regression model to determine ventilatory anaerobic
threshold. J Appl Physiol 52: 1349±1352
Fig. 6 Eect of creatine and placebo supplementation on the
relative
_
VO2at ventilatory breakpoint. Values represented are mean
(SD). *The post-creatine supplemented mean relative
_
VO2is
signi®cantly greater (P< 0.05) than the pre-supplemented relative
_
VO2
93
Paganelli W, Pendergast DR, Koness J, Cerretelli P (1989) The
eect of decreased muscle energy stores on
_
VO2kinetics at the
onset of exercise. Eur J Appl Physiol 59: 321±326
Pahud P, Ravussin E, Jequier E (1980) Energy expended during
oxygen de®cit period of submaximal exercise in man. J Appl
Physiol 48: 770±775
Prevost MC, Nelson AG, Morris GS (1997) Creatine supplemen-
tation enhances intermittent work performance. Res Q Exerc
Sport 68: 233±240
Sahlin K, Ren JM, Broberg S (1988) Oxygen de®cit at the onset of
submaximal exercise is not due to a delayed oxygen transport.
Acta Physiol Scand 134: 175±180
Scott CB (1995) Anaerobic metabolic in¯uences on oxygen uptake
behavior. J Strength Condit Res 9: 59±62
Scott CB, Bogdany GM (1998) Aerobic and anaerobic energy
expenditure during exhaustive ramp exercise. Int J Sports Med
19: 277±280
Stroud MA, Holliman D, Bell D, Green AL, MacDonald IA,
Greenha PL (1994) Eect of oral creatine supplementation on
respiratory gas exchange and blood lactate accumulation dur-
ing steady-state incremental treadmill exercise and recovery in
man. Clin Sci 87: 707±710
94