You are on page 1of 8

Eur J Appl Physiol (2008) 103:3340

DOI 10.1007/s00421-007-0669-3

O R I G I N A L A R T I CL E

EVects of creatine supplementation on renal function:


a randomized, double-blind, placebo-controlled clinical trial
Bruno Gualano Carlos Ugrinowitsch Rafael Batista Novaes
Guilherme Gianini Artioli Maria Heloisa Shimizu Antonio Carlos Seguro
Roger Charles Harris Antonio Herbert Lancha Jr

Accepted: 28 December 2007 / Published online: 11 January 2008


Springer-Verlag 2008

Abstract Creatine (CR) supplementation is commonly


used by athletes. However, its eVects on renal function
remain controversial. The aim of this study was to evaluate
the eVects of creatine supplementation on renal function in
healthy sedentary males (1835 years old) submitted to
exercise training. A randomized, double-blind, placebo-controlled trial was performed. Subjects (n = 18) were randomly
allocated to receive treatment with either creatine (CR)
(10 g day1 over 3 months) or placebo (PL) (dextrose).
All subjects undertook moderate intensity aerobic training,
in three 40-min sessions per week, during 3 months. Serum
creatinine, serum and urinary sodium and potassium were
determined at baseline and at the end of the study. Cystatin
C was assessed prior to training (PRE), after 4 (POST 4) and
12 weeks (POST 12). Cystatin C levels (mg L1) (PRE CR:
0.82 0.09; PL: 0.88 0.07 vs. POST 12 CR: 0.71 0.06;
PL: 0.75 0.09, P = 0.0001) were decreased over time,
suggesting an increase in glomerular Wltration rate. Serum
creatinine decreased with training in PL but was unchanged
B. Gualano C. Ugrinowitsch R. B. Novaes G. G. Artioli
A. H. Lancha Jr
Laboratory of Applied Nutrition and Metabolism,
School of Physical Education and Sport,
University of Sao Paulo, Sao Paulo, Brazil
M. H. Shimizu A. C. Seguro
School of Medicine, Nephrology Department,
University of Sao Paulo, Sao Paulo, Brazil
R. C. Harris
School of Sport, Exercise and Health Sciences,
University of Chichester, Chichester, UK
B. Gualano (&)
Av. Professor Mello Moraes, 65, Butant,
Sao Paulo, SP 05508-900, Brazil
e-mail: gualano@usp.br

with training in CR. No signiWcant diVerences were


observed within or between groups in other parameters
investigated. The decrease in cystatin C indicates that highdose creatine supplementation over 3 months does not provoke any renal dysfunction in healthy males undergoing aerobic training. In addition, the results suggest that moderate
aerobic training per se may improve renal function.
Keywords
Exercise

Adverse eVects Kidneys Cystatin C

Introduction
Despite well documented beneWcial eVects of creatine supplementation on exercise performance, neurodegenerative
disorders, musculoskeletal abnormalities and mitochondrial
cytopathies (Terjung et al. 2000), several case reports have
suggested that supplementation may impair renal function
(Kuehl et al. 1998; Pritchard and Kalra 1998; Koshy et al.
1999; Thorsteinsdottir et al. 2006). In addition, recent Wndings have demonstrated creatine-induced deleterious eVects
in rats with cystic kidney disease (Edmunds et al. 2001) and
sedentary animals (Ferreira et al. 2005). In contrast, other
studies have indicated no deleterious eVect in sedentary
(Taes et al. 2003) or exercised (Ferreira et al. 2005) rats,
and in pre-existing renal failure rats (Taes et al. 2003).
Studies in humans have indicated that creatine supplementation does not aVect renal function (Poortmans et al. 1997;
Poortmans and Francaux 1999; Poortmans et al. 2005; Robinson et al. 2000). However, these studies have been criticized by the absence of prospective randomization, the lack
of any uniform creatine source and dosage, the limited
study power of the studies, and particularly the low sensitivity of the methods used to evaluate glomerular Wltration

123

34

Eur J Appl Physiol (2008) 103:3340

rate (GFR) (Kuehl et al. 2000). The relevance of the rat or


other animal models to the human is also questionable
(Kreider 2003; Tarnopolsky et al. 2003) given that creatine
is not a normal nutrient in such species and may not be
absorbed to any great extent from the gut, as for example in
the horse (Sewell and Harris 1995). This contrasts with
humans where creatine is rapidly absorbed and intake
(expressed as g monohydrate) may routinely be 12 g daily,
and in some diets as high as 57 g daily. Still higher daily
intakes, which on a g per kg body weight basis may exceed
on even those taken by athletes, are found in carnivorous
animals with no apparent eVect on renal function.
The glomerular Wltration rate is widely accepted as the
best overall measure of kidney function (Dharnidharka
et al. 2002). While inulin clearance and 51Chromium-ethylenediamine tetraacetic acid (Cr-EDTA) are traditionally
considered as the gold standard for estimation of GFR,
these methods are highly specialized and take several hours
to perform. In addition, a number of practical considerations (i.e. cumbersome methods of determination, use of
radioactive isotopes, and high costs) limit their use in clinical practice. In the last 40 years, serum creatinine has
become the most commonly used marker of renal function.
However, since creatine is converted spontaneously to creatinine (Wyss and Kaddurah-Daouk 2000), the use of creatine supplements to increase tissue creatine levels will
invariably increase creatinine production, falsely suggesting renal function impairment. Therefore, the use of creatinine related parameters alone to estimate GFR is largely
inadequate in creatine supplementation trials (Pline and
Smith 2005). As an alternative, several studies (Kazama
et al. 2002; Stevens et al. 2006) including some meta-analyzes (Dharnidharka et al. 2002; Roos et al. 2007) have
suggested that cystatin C (13.3 kDa), a novel and speciWc
marker of GFR, has a role in the assessment of renal function in certain groups of patient where creatinine cannot be
used. These studies have demonstrated in several populations that serum cystatin C is more sensitive than creatinine
Table 1 Subject characteristics
measured at baseline and week
12

No signiWcant diVerences were


observed between creatine
(n = 9) and placebo (n = 9)
groups at baseline and at 12th
week (unpaired t test). * versus
baseline (P 0.05). VT ventilatory gas exchange threshold; HR
heart rate; NA not assessed at
this time

123

in the detection of early and less-severe changes in kidney


function. The validity and diagnostic accuracy of cystatin C
were investigated by Kazama et al. (2002). In addition,
Taes et al. (2003) reported cystatin C to be signiWcantly
correlated to inulin clearance in pre-existing renal failure
and control rats (r = 0.82; P = 0.001), despite 4 weeks creatine supplementation. Thus, cystatin C appear to be a potentially useful marker for the estimation of GFR particularly
in creatine supplementation trials.
We performed a double blind, randomized, placebo-controlled trial that aimed to investigate the eVects of highdose creatine supplementation on renal function in sedentary healthy males submitted to 3 months of moderate
intensity aerobic training.

Methods and materials


Subjects
Participants were recruited between August and December
2005 from Sao Paulo, Brazil, through advertisement in
local newspapers and through the distribution of leaXets to
private homes. Respondents were invited to take part if
they were male (1835 years old) and had lead a sedentary
life for at least the past 7 years, did not have any pre-existing renal or cardiovascular diseases, were eutrophic, nonsmokers, and non-vegetarians. Subject characteristics are
presented in Table 1.
The study was approved by the local Ethics and
Research Committee. Subjects were informed of the experimental procedures and of any possible risks before giving
their written informed consent to participate.
Experimental protocol
A 12-week, double-blind, randomized, placebo-controlled
trial was conducted between 1 March and 1 June 2006.
Creatine group
Baseline

Placebo group
Week 12

Baseline

Week 12

Age (years)

24.2 5.0

NA

24.6 4.2

NA

VO2peak (ml kg1 min1)

37.6 4.6

NA

37.1 4.1

NA

VT (ml kg1 min1)

30.7 6.1

NA

28.9 5.9

NA

BMI (kg m )

23.1 4.5

23.2 2.3

24.0 3.2

23.9 2.2

Body fat content (%)

17.9 3.1

13.1 5.1

17.2 2.8

12.75 4.9*

Waist-to-hip ratio (cm)

0.83 0.06

0.82 0.09

0.83 0.05

0.82 0.07

Waist circumference (cm)

80.1 8.2

79.9 7.2

79.9 9.6

79.2 4.2

64.1 6.1

58.2 2.3

65.1 6.5

59.4 3.1*

3.8 0.8

5.4 0.2

3.8 0.4

5.2 0.8*

Resting HR (beats min )


Running distance per
training session (km)

Eur J Appl Physiol (2008) 103:3340

Subjects were randomly assigned to receive either creatine (CR; n = 9) or placebo (PL; n = 9) in a double-blind
fashion. As the majority of creatine consumers are physically active, both groups undertook a program of moderate
intensity aerobic training for 3 months. At baseline (PRE)
and after 4 (POST 4) and 12 weeks (POST 12) subjects
underwent renal function tests. Blood samples were
obtained 48 h after the last training session from an antecubital vein following 12 h overnight fast, at all time points
(PRE, POST 4 and POST 12). Twenty-four-hour urine collection was also obtained 48 h after the last training session
at baseline and at the end of creatine supplementation. Possible diVerences in dietary intake were assessed by a 24-h
dietary recall at PRE and POST 12. The experimental
design is illustrated in Fig. 1.
Creatine supplementation protocol and blinding
The CR group received 0.3 g day1 kg1 of body weight for
the Wrst week, and 0.15 g day1 kg1 of body weight for the
next 11 weeks; the PL group was given dextrose in place of
creatine, at the same dose. During the loading phase, individuals received the supplement in four packages, with
equal content, and were instructed to ingest one supplement
package at breakfast, lunch and dinner and the fourth at
10 pm. During the maintenance phase, subjects consumed
the supplement as two equally divided doses, one during
lunch and the second during dinner. These packages were
coded so that neither the investigators nor the participants
were aware of the contents until completion of the analyses.
All subjects were instructed to dissolve the supplements,
preferably in juice, in order to mask both the low solubility
of creatine and the taste of dextrose. The compliance to creatine supplementation was monitored weekly by personal
communication. The subjects completed a weekly questionnaire, adapted from Volek et al. 2000, to ascertain any possible adverse eVects of creatine supplementation.

Fig. 1 Experimental design. The subjects were randomly allocated to


receive creatine supplementation (0.3 g day1 kg1 of body weight for
the Wrst week, and 0.15 g day1 kg1 of body weight for the next
11 weeks, n = 9) or placebo (dextrose, n = 9) in a double-blind fashion.
Blood samples were obtained at baseline and after 4, 8 and 12 weeks.
The 24-h urine collections were performed at baseline and at the end of
the study. Renal parameters analyzed include serum cystatin C, creatinine, sodium and potassium and urinary sodium and potassium. See
Methods and materials section for more details. suppl supplementation; HPLC high-performance liquid chromatography

35

In order to verify the purity of the creatine used, a sample was analyzed by HPLC. This established 99.9% of
purity, with no other peaks detected (creatinine, dicyandiamide and cyclocreatine < 0.01%).
VO2peak test
Peak oxygen uptake capacity was determined at the start
of the study using a treadmill exercise test (Bruce et al.
1973). This test is conducted in 3 min stages. The Bruce
protocol starts at 4.6 METS of work, at a speed of 2.7 km/
h and a grade of 10 degrees. Every 3 min, the workload is
increased by a combination of increasing speed and the
gradient of the treadmill. VO2peak was obtained when two
of three criteria were met, namely a plateau in VO2, a
respiratory exchange ratio (RER) > 1.1, and volitional
exhaustion.
Exercise training protocol
The exercise training protocol has been described previously (Gualano et al. 2008). BrieXy, both groups were submitted to aerobic training, three times a week, for 40 min
per session, for 3 months. Running intensity was set at the
heart rate corresponding to 70% of VO2peak. Subjects were
requested to run at the pre-set heart rate during the whole of
the training session. Thus, the total amount of exercise
undertaken during training increased in both running distance and intensity in response to the training. All sessions
were monitored by a Wtness-professional. From the outset it
was ruled that subjects would be dropped from the study if
they missed three non-consecutives training sessions, or
two consecutives sessions. One subject from the PL group
was excluded for this reason.
Food intake assessment
Food intake was assessed by means of three 24-h dietary
recalls undertaken on separate days (2 weekdays and
1 weekend day) using a visual aid photo album of real
foods. The 24-h dietary recall consists of the listing of
foods and beverages consumed during 24 h prior to the
recall. The interviews were conducted face to face by a
trained dietitian, using the multiple-pass method developed
by USDA (Conway et al. 2004). Energy and macronutrient
intakes were analyzed by the Brazilian software, Virtual
Nutri, based on Handbook number 8 of the USDA Human
Nutrition Information Service (Composition of food: raw;
processed; prepared, series 116, revised 19761986).
Additional information was obtained from the Brazilian
Table of Food Composition (version 1.0, 1997, University
of So Paulo, Sao Paulo, SP). The nutrient database was
complete for all the nutrients studied.

123

36

Eur J Appl Physiol (2008) 103:3340

Blood and urinary analyses


All samples were analyzed in duplicate and the average
value was used for data analysis. The intra-class coeYcient (ICC) were 0.869, 0.953, 0.967 and 0.805 (P < 0.001)
and the coeYcient of variation (CV) were 2, 0.4, 1.1, 2.8%
for creatinine, sodium, potassium and cystatin C, respectively.
Creatinine was determined in plasma and urine using an
enzymatic colorimetric test PAP (Labtest, So Paulo,
Brazil).
Serum and urinary sodium, and potassium were assessed
using a Xame photometer (model 143: Instrumentation Laboratory, Lexington, MA, USA).
Cystatin C was measured using a BN II Nephelometer
(Dade Behring) and a particle-enhanced turbidimetric
immunoassay (N Latex Cystatin C).

Statistical analysis
Renal parameters were tested by a two-way ANOVA
(mixed model) with repeated measures. Group (CR and PL)
and time (pre, 4 and 12 weeks) were considered Wxed factors and subjects were deWned as a random factor. A post
hoc test adjusted by Tukey was used for multicomparison
purposes. Baseline characteristics between groups were
compared using Students t test for unpaired data. SigniWcance level was set at P < 0.05. Data are presented as
mean standard deviation. SAS proc IML was used to
perform a simulation to determine the power of the statistical tests (Ugrinowitsch et al. 2004) as suggested by Littell
et al. (2006). Expected means and variance were used as
input to simulate cystatin C data using an autoregressive
correlation structure within subjects measurements. Nine
subjects produced a power of 0.95 in the comparison
between groups for cystatin C, the most important dependent variable.

Results
The number of subjects recruited to the study is shown in
Fig. 2. Of the 103 people who responded to the initial
request for volunteers, 81 were excluded because they
performed regular physical activity or expressed selfreported chronic disease. The remaining 22 subjects were
randomly assigned to the CR or PL groups. Four subjects
were subsequently lost: three withdrew for personal reasons (two in CR and one in PL) and one was excluded
because they missed two consecutive training sessions
(PL). The compliance to creatine or placebo supplementation was 100%.

123

Fig. 2 Flow diagram of participants

As mentioned earlier, creatine purity was assessed by


HPLC as 99.9%. Moreover, there was no diVerence
between or within groups in nutrient or energy intake
(Table 2). Hence, it is probable that neither the presence of
trace contaminants in the creatine or diVerences in dietary
intake were a factor in the study. Despite the high dose of
creatine ingested (compared with ACSM recommendations, Terjung et al. 2000), there was no report of any deleterious eVects assessed through subjective questionnaire.
The compliance to creatine supplementation was 100%.
The blinding was considered successful because the ability
of participants to accurately guess their group assignment
was no better than chance in both groups and was not diVerent between groups (CR: 33% and PL: 44%; P = 0.78).
Twelve weeks of aerobic training signiWcantly increased
the running distance per session. There was also a time
main eVect for body fat content measured by skinfold thickness and resting heart rate; other measured variables
remained unaltered (Table 1).
Cystatin C (Fig. 3) decreased at the same rate in both
groups from the beginning to the end of the training protocol (main time eVect) (PRE CR: 0.82 0.09; PL:
0.88 0.07 vs. POST 12 CR: 0.71 0.06; PL:
0.75 0.09 mg L1; P = 0.0001). With the exception of
one subject from PL group, all subjects showed a decrease
in serum cystatin C after 12 weeks.
PL showed a decrease in serum creatinine at weeks 4 and
12 (main time eVect, P = 0.004 and 0.003, respectively)
while CR did not (Fig. 4). The diVerence in response to the
treatment between groups was signiWcant at weeks 4 and 12
(P = 0.0001 and 0.0004, respectively). Unfortunately, it
was not possible to calculate the creatinine clearance
because most of the subjects failed to adequately collect the
24-h urine samples.

Eur J Appl Physiol (2008) 103:3340


Table 2 Food intake assessed
by three 24-h dietary recalls at
baseline and after 12 weeks in
creatine (n = 9) and placebo
(n = 9) groups

No signiWcant diVerences were


found

37

CR group

PL group

PRE

POST 12

PRE

POST 12

Energy (kcal)

2728 238

2833 201

2844 201

2899 287

Protein (%)

28.0 4.3

30.0 3.1

29.8 4.3

31.1 3.9

Lipid (%)

25.5 4.9

24.5 4.5

24.9 4.1

26 3.1

Carbohydrate (%)

46.3 3.9

43.3 5.3

45.3 4.9

42.9 5.1

1.1 0.3

1.2 0.3

1.1 0.2

1.2 0.3

Protein/body weight (g/kg)

Fig. 3 Serum cystatin C concentration at baseline and after 4 and


12 weeks in creatine (CR; n = 9) and placebo (PL; n = 9) groups. Main
time eVect: * versus week 4 (P = 0.001); # versus baseline
(P = 0.0001). a Individual data; b Mean standard deviation

Fig. 4 Serum creatinine concentration at baseline, after 4 and


12 weeks in creatine (CR; n = 9) and placebo (PL; n = 9) groups.
*Interaction eVect: PL versus CR (P < 0.005) and #time eVect: PL
versus baseline (P < 0.0001). a Individual data; b Mean standard
deviation

No signiWcant diVerences within or between groups were


found in serum and urinary sodium and potassium concentration at the start and end of the study.

Discussion
The aim of the study was to investigate the eVects of creatine supplementation on renal function in sedentary healthy
males submitted to aerobic training. To our knowledge, this
is the Wrst randomized, double-blind, placebo-controlled
trial to demonstrate that high-dose long-term creatine supplementation (10 g over 3 months) does not impair renal
function in healthy subjects, initially sedentary but undergoing a period of moderate training. In addition, we believe
that this is the Wrst study to use cystatin C as a marker of
GFR in humans supplemented with creatine.
The eVects of creatine supplementation on renal function
have been intensely discussed in the literature. Some case

reports suggested that creatine supplementation could cause


deterioration in renal function in healthy individuals (Thorsteinsdottir et al. 2006), and in subjects with pre-existing
renal failure (Pritchard and Kalra 1998). On the other hand,
longitudinal data have indicated otherwise for healthy individuals (Poortmans et al. 1997, 2005; Poortmans and
Francaux 1999; Kreider et al. 2003). Long-term studies
with creatine supplementation in humans have presented
severe limitations such as a lack of prospective randomization, the absence of any uniform creatine source and dosage, limited statistical power and, most frequently, the
absence of a marker of renal function other than creatinine
(Kuehl et al. 2000).
The classic way to estimate renal function in humans is
through creatinine clearance. However, creatine is converted spontaneously to creatinine (Ferreira et al. 2005;
Pline and Smith 2005) and elevation of tissue creatine levels in supplemented subjects will inevitably increase serum
creatinine levels (Wyss and Kaddurah-Daouk 2000), falsely
suggesting renal dysfunction. In order to avoid this, noncreatinine related markers of GFR are needed of which the
use of inulin clearance and Cr-EDTA are considered to be
the most reliable (Dharnidharka et al. 2002). These techniques, however, are time-consuming, labour-intensive,
expensive, and require administration of substances that
make them incompatible with the routine monitoring of
subjects. Human cystatin C, a basic low molecular mass
protein, with 120 amino acid residues, is freely Wltered by
the glomerulus and almost completely reabsorbed and
catabolized by the proximal tubular cells. The use of serum
cystatin C to estimate GFR is based on the same logic as the
use of creatinine, but because it does not return to bloodstream and is not secreted by renal tubules, it has been suggested to be closer to an ideal endogenous marker.
Several studies have pointed out the advantages of cystatin
C in estimating GFR compared to creatinine (Dharnidharka
et al. 2002; Kazama et al. 2002). Perhaps the main advantage of cystatin C over creatinine is that the former protein
is unaVected by food ingestion (Preiss et al. 2007). As a
result cystatin C is an attractive marker of renal function in
creatine trials and clinical practice. However, caution needs
to be exercised as there is currently only one study that
evaluates the correlation between cystatin C and inulin

123

38

clearance in creatine supplemented rats. Thus, further studies need to be conducted in order to determine the diagnostic eYcacy of cystatin C when compared with other
established markers.
Serum creatinine was signiWcantly lower in PL than in
CR subjects after creatine supplementation, which might
suggest renal function impairment. However, in contrast
cystatin C was reduced in both groups, which suggests otherwise. The higher concentrations of serum creatinine in the
CR group are most probably a reXection of the increase in
tissue creatine levels brought about by creatine supplementation, and the non-enzymic Wrst-order conversion of this to
creatinine at the rate of 1.7% per day (Wyss and Kaddurah-Daouk 2000; Pline and Smith 2005). A similar conclusion was reached by Taes et al. (2003).
However, the eVect of creatine supplementation on renal
function remains controversial even in controlled studies in
rats, which used established methods to evaluate GFR. Taes
et al. (2003) failed to demonstrate any deleterious eVects in
rats with either renal failure or normal renal function. In
contrast, Ferreira et al. (2005) suggested that creatine supplementation induced renal dysfunction in sedentary
healthy rats. In the latter study, the authors believed that
renal impairment was provoked by creatine-induced vasoconstriction and/or cytotoxic compounds. Unfortunately,
only Taes et al. 2003 reported the purity of the creatine
administered. Hence, trace contaminants associated with
the creatine could also have been partially responsible for
the opposing Wndings in the literature. In the present study,
creatine purity was assessed by HPLC and no trace contaminants were found.
Notwithstanding the observations made in rats, the relevance of these results to humans which routinely encounter
creatine in the diet is also open to question. As noted earlier
the typical non-vegetarian diet may contain 12 g creatine
(expressed as the monohydrate) daily, reaching a maximum
of 57 g. Still higher levels are found in the diets of carnivorous animals of equivalent body weight. It would be surprising if, given the span of human evolution, some
adaptation to creatine intake had not occurred. Indeed the
rapid and complete absorption of creatine supplied in the
diet (Harris et al. 1992, 2002), in contrast to the lack of
absorption in a herbivorous animal such as the horse
(Sewell and Harris 1995), suggests that adaptation has
occurred at least at one level. The elegant study conducted
by Tarnopolsky et al. (2003) highlighted the species- and
tissue-speciWc response to creatine intake. These authors
demonstrated that creatine administration can induce
chronic hepatitis in mice, but not in rats. Interestingly, neither the mice nor rats showed renal dysfunction after longterm supplementation, supporting our Wndings.
Another relevant Wnding of this study refers to the eVects
of exercise training per se on renal function. We observed

123

Eur J Appl Physiol (2008) 103:3340

that 3 months of moderate aerobic training reduced serum


cystatin C and serum creatinine (only in the placebo group),
suggesting improvement in renal function. Despite the relevance of our Wnding, we cannot provide any comparative
discussion with other studies because, to the best of our
knowledge, there are none that have investigated the eVects
of exercise training on GFR. In fact, Finkelstein et al.
(2006) showed a strong positive association between physical activity and GFR in a cross-sectional review of The
Third National Health and Nutrition Examination Survey
(NHANES III). However, the authors concluded that a longitudinal study would be necessary to clearly deWne this
relationship. Although the mechanism(s) responsible for
the renal function improvement observed in our study
remains unknown, it should be noted that there is strong
evidence in literature to indicate that intensive glycemic
control, blood pressure control (Klahr et al. 1994), lipid
lowering strategies (Fried et al. 2001), oxidative stress
decrease (Gaede et al. 2001) and weight management
(Stengel et al. 2003) are important factors in arresting or
even to preventing chronic kidney disease progression.
According to Jaber and Madias (2005), it is possible to
achieve regression in the age-related decline in renal function through the control of these factors, all of which are
beneWcially improved by moderate intensity exercise training. We have previously shown (Gualano et al. 2006, 2008)
that the exercise protocol used in this study was capable of
improving glucose tolerance and the lipid proWle, as well as
reducing markers of oxidative stress. In the absence of a
suitable control group, however, we cannot conclude a
cause-and-eVect relationship in this case.
Shao and Hathcock (2006) reviewed the available data
and suggested a safe limit for creatine monohydrate intake
of 5 g day1 although creatine users frequently consume
higher doses than this (Harris et al. 1992). In their opinion
this was a dose which avoided any side eVects. However,
their conclusion was inXuenced by the fact that there were
only three studies in existence that used 10 g day1 in the
maintenance phase. Two of these did not report side eVects
but neither was not controlled or randomized. The third
study investigated patients with amyotrophic lateral scleroses for 310 days did not Wnd signiWcant changes in serum
urea or urinary albumin levels with creatine supplementation (Groeneveld et al. 2005). However, because of the disease nature of the population studied, this did not allow
further extrapolation to healthy subjects. Our data contribute to the hypothesis that 10 g day1 of creatine is safe and
has no eVect on renal function in healthy subjects undergoing moderate intensity exercise training.
Although our study presents some important positive
points, such as well-controlled design, satisfactory statistical power and a GFR estimate marker not related to creatinine, we also recognize that there were some limitations.

Eur J Appl Physiol (2008) 103:3340

Firstly, we did not include a non-exercised training group.


Thus, although the improvement in renal function appears
to be related to exercise training, we cannot exclude the
possibility that other factors within the experimental protocol may have contributed to this. Further randomized controlled trials should be performed to investigate the eVects
of aerobic training on renal function. Secondly, the study
comprised of initially sedentary healthy males who were
undergoing moderate aerobic exercise training. Therefore,
these Wndings cannot be generalized to people with renal
disease, athletes in intensive training or resistance trained
subjects, non-exercising sedentary subjects, the elderly and
other populations not examined in this study. Finally, our
study did not aim to explore mechanisms. Therefore,
although we speculated that renal function improvement
might be the result of exercise training, further work is necessary to explain the mechanisms behind these Wndings.
In conclusion, high-dose creatine supplementation above
that likely to be encountered in the diet did not result in any
adverse changes in renal function in sedentary healthy
males submitted to moderate aerobic training during
3 months. Moreover, our results suggest that moderate aerobic training per se may even improve renal function.
Acknowledgments We are grateful to Fundao de Amparo Pesquisa do Estado de So Paulo (FAPESP) and Conselho Nacional de
Desenvolvimento CientWco e Tecnolgico (CNPq) for Wnancial support. We would also like to express our thanks to the subjects who took
part in the study.

References
Bruce RA, Kusumi F, Hosmer D (1973) Maximal oxygen intake and
nomographic assessment of functional aerobic impairment in cardiovascular disease. Am Heart J 85:546562
Conway JM, Ingwersen LA, Moshfegh AJ (2004) Accuracy of dietary
recall using the USDA Wve-step multiple-pass method in men: an
observational validation study. J Am Diet Assoc 104:595603
Dharnidharka VR, Kwon C, Stevens G (2002) Serum cystatin C is
superior to serum creatinine as a marker of kidney function: a
meta-analysis. Am J Kidney Dis 40:221226
Edmunds JW, Jayapalan S, DiMarco NM, Saboorian MH, Aukema
HM (2001) Creatine supplementation increases renal disease progression in Han:SPRD-cy rats. Am J Kidney Dis 37:7378
Ferreira LG, De Toledo Bergamaschi C, Lazaretti-Castro M, Heilberg
IP (2005) EVects of creatine supplementation on body composition and renal function in rats. Med Sci Sports Exerc 37:1525
1529
Finkelstein J, Joshi A, Hise MK (2006) Association of physical activity
and renal function in subjects with and without metabolic syndrome: a review of the Third National Health and Nutrition Examination Survey (NHANES III). Am J Kidney Dis 48:372382
Fried LF, Orchard TJ, Kasiske BL (2001) EVect of lipid reduction on
the progression of renal disease: a meta-analysis. Kidney Int
59:260269
Gaede P, Poulsen HE, Parving HH, Pedersen O (2001) Double-blind,
randomised study of the eVect of combined treatment with vita-

39
min C and E on albuminuria in Type 2 diabetic patients. Diabet
Med 18:756760
Groeneveld GJ, Beijer C, Veldink JH, Kalmijn S, Wokke JH, van den
Berg LH (2005) Few adverse eVects of long-term creatine supplementation in a placebo-controlled trial. Int J Sports Med 26:307
313
Gualano B, Novaes RB, Shimuzu HM, Artioli GG, Rogeri PS, Ugrinowitsch C, Seguro AC, Lancha AH Jr (2006) Aerobic training
improves renal function and decreases oxidative stress in healthy
sedentary males, despite high-dose creatine supplementation. J
Am Soc Nephrol 203A
Gualano B, Novaes RB, Artioli GG, Freire TO, Coelho DF, Scagliusi
FB, Rogeri PS, Roschel H, Ugrinowitsch C, Lancha AH Jr (2008)
EVects of creatine supplementation on glucose tolerance and insulin sensitivity in sedentary healthy males undergoing aerobic
training. Amino Acids (in press)
Harris RC, Nevill M, Harris DB, FallowWeld JL, Bogdanis GC, Wise
JA (2002) Absorption of creatine supplied as a drink, in meat or
in solid form. J Sports Sci 20:147151
Harris RC, Soderlund K, Hultman E (1992) Elevation of creatine in
resting and exercised muscle of normal subjects by creatine supplementation. Clin Sci (Lond) 83:367374
Jaber BL, Madias NE (2005) Progression of chronic kidney disease:
can it be prevented or arrested?. Am J Med 118:13231330
Kazama JJ, Kutsuwada K, Ataka K, Maruyama H, Gejyo F (2002) Serum cystatin C reliably detects renal dysfunction in patients with
various renal diseases. Nephron 91:1320
Klahr S, Levey AS, Beck GJ, Caggiula AW, Hunsicker L, Kusek JW,
Striker G (1994) The eVects of dietary protein restriction and
blood-pressure control on the progression of chronic renal disease. ModiWcation of diet in Renal Disease Study Group. N Engl
J Med 330:877884
Koshy KM, Griswold E, Schneeberger EE (1999) Interstitial nephritis
in a patient taking creatine. N Engl J Med 340:814815
Kreider RB (2003) Species-speciWc responses to creatine supplementation. Am J Physiol Regul Integr Comp Physiol 285:R725R726
Kreider RB, Melton C, Rasmussen CJ, Greenwood M, Lancaster S,
Cantler EC, Milnor P, Almada AL (2003) Long-term creatine
supplementation does not signiWcantly aVect clinical markers of
health in athletes. Mol Cell Biochem 244:95104
Kuehl K, Goldberg L, Elliot D (1998) Renal insuYciency after creatine
supplementation in a college football athlete (abstract). Med Sci
Sports Exerc 30:S235
Kuehl K, Goldberg L, Elliot D (2000) Re: long-term oral creatine supplementation does not impair renal function in healthy athletes.
Med Sci Sports Exerc 32:248249
Littell RC, Milliken GA, Stroup WW, WolWnger RD, Schabenberger
O (2006) SAS system for mixed models. SAS Institute, Cary
Pline KA, Smith CL (2005) The eVect of creatine intake on renal function. Ann Pharmacother 39:10931096
Poortmans JR, Auquier H, Renaut V, Durussel A, Saugy M, Brisson
GR (1997) EVect of short-term creatine supplementation on renal
responses in men. Eur J Appl Physiol Occup Physiol 76:566567
Poortmans JR, Francaux M (1999) Long-term oral creatine supplementation does not impair renal function in healthy athletes. Med
Sci Sports Exerc 31:11081110
Poortmans JR, Kumps A, Duez P, Fofonka A, Carpentier A, Francaux
M (2005) EVect of oral creatine supplementation on urinary
methylamine, formaldehyde, and formate. Med Sci Sports Exerc
37:17171720
Preiss DJ, Godber IM, Lamb EJ, Dalton RN, Gunn IR (2007) The
inXuence of a cooked-meat meal on estimated glomerular Wltration rate. Ann Clin Biochem 44:3542
Pritchard NR, Kalra PA (1998) Renal dysfunction accompanying oral
creatine supplements. Lancet 351:12521253

123

40
Robinson TM, Sewell DA, Casey A, Steenge G, GreenhaV PL (2000)
Dietary creatine supplementation does not aVect some haematological indices, or indices of muscle damage and hepatic and renal
function. Br J Sports Med 34:284288
Roos JF, Doust J, Tett SE, Kirkpatrick CM (2007) Diagnostic accuracy
of cystatin C compared to serum creatinine for the estimation of
renal dysfunction in adults and children-a meta-analysis. Clin
Biochem 40:383391
Sewell DA, Harris RC (1995) EVect of creatine supplementation in the
Thoroughbred horse. Equine Vet J 18:239242
Shao A, Hathcock JN (2006) Risk assessment for creatine monohydrate. Regul Toxicol Pharmacol 45:242251
Stengel B, Tarver-Carr ME, Powe NR, Eberhardt MS, Brancati FL
(2003) Lifestyle factors, obesity and the risk of chronic kidney
disease. Epidemiology 14:479487
Stevens LA, Coresh J, Greene T, Levey AS (2006) Assessing kidney
function-measured and estimated glomerular Wltration rate. N
Engl J Med 354:24732483
Taes YE, Delanghe JR, Wuyts B, van de Voorde J, Lameire NH (2003)
Creatine supplementation does not aVect kidney function in an
animal model with pre-existing renal failure. Nephrol Dial Transplant 18:258264

123

Eur J Appl Physiol (2008) 103:3340


Tarnopolsky MA, Bourgeois JM, Snow R, Keys S, Roy BD, Kwiecien
JM, Turnbull J (2003) Histological assessment of intermediate
and long-term creatine monohydrate supplementation in mice and
rats. Am J Physiol Regul Integr Comp Physiol 285:R762R769
Terjung RL, Clarkson P, Eichner ER, GreenhaV PL, Hespel PJ, Israel
RG, Kraemer WJ, Meyer RA, Spriet LL, Tarnopolsky MA, Wagenmakers AJ, Williams MH (2000) American College of Sports
Medicine roundtable. The physiological and health eVects of oral
creatine supplementation. Med Sci Sports Exerc 32:706717
Thorsteinsdottir B, Grande JP, Garovic VD (2006) Acute renal failure
in a young weight lifter taking multiple food supplements, including creatine monohydrate. J Ren Nutr 16:341345
Ugrinowitsch C, Fellingham GW, Ricard MD (2004) Limitations of
ordinary least squares models in analyzing repeated measures data. Med Sci Sports Exerc 36:21442148
Volek JS, Duncan ND, Mazzetti SA, Putukian M, Gomez AL, Kraemer
WJ (2000) No eVect of heavy resistance training and creatine supplementation on blood lipids. Int J Sport Nutr Exerc Metab
10:144156
Wyss M, Kaddurah-Daouk R (2000) Creatine and creatinine metabolism. Physiol Rev 80:11071213