Creatine in Type 2 Diabetes: A Randomized, Double-Blind, Placebo-Controlled Trial

Abstract

Creatine supplementation improves glucose tolerance in healthy subjects.
The aim was to investigate whether creatine supplementation has a beneficial effect on glycemic control of type 2 diabetic patients undergoing exercise training.
A 12-wk randomized, double-blind, placebo-controlled trial was performed. The patients were allocated to receive either creatine (CR) (5 g·d) or placebo (PL) and were enrolled in an exercise training program. The primary outcome was glycosylated hemoglobin (HbA1c). Secondary outcomes included the area under the curve of glucose, insulin, and C-peptide and insulin sensitivity indexes. Physical capacity, lipid profile, and GLUT-4 protein expression and translocation were also assessed.
Twenty-five subjects were analyzed (CR: n=13; PL: n=12). HbA1c was significantly reduced in the creatine group when compared with the placebo group (CR: PRE=7.4 ± 0.7, POST=6.4 ± 0.4; PL: PRE=7.5 ± 0.6, POST=7.6 ± 0.7; P=0.004; difference=-1.1%, 95% confidence interval=-1.9% to -0.4%). The delta area under the curve of glucose concentration was significantly lower in the CR group than in the PL group (CR=-7790 ± 4600, PL=2008 ± 7614; P=0.05). The CR group also presented decreased glycemia at times 0, 30, and 60 min during a meal tolerance test and increased GLUT-4 translocation. Insulin and C-peptide concentrations, surrogates of insulin sensitivity, physical capacity, lipid profile, and adverse effects were comparable between the groups.
Creatine supplementation combined with an exercise program improves glycemic control in type 2 diabetic patients. The underlying mechanism seems to be related to an increase in GLUT-4 recruitment to the sarcolemma.

Full text

Creatine in Type 2 Diabetes: A Randomized,
Double-Blind, Placebo-Controlled Trial
BRUNO GUALANO
1,2
, VITOR DE SALLES PAINNELI
1,2
, HAMILTON ROSCHEL
1,2
,
GUILHERME GIANNINI ARTIOLI
1,2
, MANOEL NEVES JR
2
, ANA LU
´CIA DE SA
´PINTO
2
,
MARIA ELIZABETH ROSSI DA SILVA
3
,MARIAROSA
´RIA CUNHA
3
, MARIA CONCEPCIO
´NGARCI
´AOTADUY
4
,
CLAUDIA DA COSTA LEITE
4
,JU
´LIO CE
´SAR FERREIRA
1
, ROSA MARIA PEREIRA
2
, PATRI
´CIA CHAKUR BRUM
1
,
ELOISA BONFA
´
2
, and ANTONIO HERBERT LANCHA JR
1
1
Laboratory of Applied Nutrition and Metabolism, School of Physical Education and Sports, University of Sa˜o Paulo,
Sa˜o Paulo, BRAZIL;
2
Division of Rheumatology, School of Medicine, University of Sa˜ o Paulo, Sa˜ o Paulo, BRAZIL;
3
Division of Endocrinology Laboratory of Medical Investigation, School of Medicine, University of Sa˜o Paulo, Sa˜o Paulo,
BRAZIL; and
4
Division of Radiology, School of Medicine, University of Sa˜ o Paulo, Sa˜ o Paulo, BRAZIL
ABSTRACT
GUALANO, B., V. DE. SALLES PAINNELI, H. ROSCHEL, G. G. ARTIOLI, M. NEVES JR, A. L. DE SA
´PINTO, M. E. DA SILVA,
M. R. CUNHA, M. C. G. OTADUY, C. DA COSTA LEITE, J. C. FERREIRA, R. M. PEREIRA, P. C. BRUM, E. BO NFA
´,and
A. H. LANCHA JR. Creatine in Type 2 Diabetes: A Randomized, Double-Blind, Placebo-Controlled Trial. Med. Sci. Sports Exerc.,
Vol. 43, No. 5, pp. 770–778, 2011. Creatine supplementation improves glucose tolerance in healthy subjects. Purposes:Theaimwas
to investigate whether creatine supplementation has a beneficial effect on glycemic control of type 2 diabetic patients undergoing
exercise training. Methods: A 12-wk randomized, double-blind, placebo-controlled trial was performed. The patients were allocated
to receive either creatine (CR) (5 gId
j1
) or placebo (PL) and were enrolled in an exercise training program. The primary outcome
was glycosylated hemoglobin (Hb
A1c
). Secondary outcomes included the area under the curve of glucose, insulin, and C-peptide and
insulin sensitivity indexes. Physical capacity, lipid profile, and GLUT-4 protein expression and translocation were also assessed.
Results: Twenty-five subjects were analyzed (CR: n= 13; PL: n=12).Hb
A1c
was significantly reduced in the creatine group when
compared with the placebo group (CR: PRE = 7.4 T0.7, POST = 6.4 T0.4; PL: PRE = 7.5 T0.6, POST = 7.6 T0.7; P= 0.004;
difference = j1.1%, 95% confidence interval = j1.9% to j0.4%). The delta area under the curve of glucose concentration was
significantly lower in the CR group than in the PL group (CR = j7790 T4600, PL = 2008 T7614; P= 0.05). The CR group also
presented decreased glycemia at times 0, 30, and 60 min during a meal tolerance test and increased GLUT-4 translocation. Insulin
and C-peptide concentrations, surrogates of insulin sensitivity, physical capacity, lipid profile, and adverse effects were compa-
rable between the groups. Conclusions: Creatine supplementation combined with an exercise program improves glycemic control
in type 2 diabetic patients. The underlying mechanism seems to be related to an increase in GLUT-4 recruitment to the sarcolemma.
Key Words: CREATINE SUPPLEMENTATION, EXERCISE TRAINING, TYPE 2 DIABETES, THERAPEUTIC EFFECTS
For decades, physical activity has been considered the
major cornerstone of type 2 diabetes management,
along with diet and medication (24). Indeed, strate-
gies capable of mimicking and/or enhancing the effects of
exercise are potentially therapeutic. In this context, creatine
supplementation has emerged as a novel putative candidate
for treating diabetes (8,16).
Creatine, a natural amine in the human body, is partly
synthesized by kidneys, pancreas, and liver (approximately
1–2 gId
j1
), as well as ingested from food (approximately
1–5 gId
j1
), especially meat and fish, and thereafter mainly
transported to the skeletal muscles, brain, and testes. Cre-
atine has rapidly become one of the most consumed nu-
tritional supplements worldwide owing to its efficacy to
increase muscle phosphorylcreatine content, thereby en-
hancing athletic performance and, consequently, lean mass.
Although the literature supports the role of creatine supple-
mentation in improving acute work capacity during inter-
mittent short-duration high-intensity exercise, the effect of
this supplement on skeletal muscle protein synthesis is less
clear (3,17,27).
A growing body of evidence has now revealed a number
of therapeutic potential applications of this supplement in
a broad range of diseases, notably muscle disorders, neu-
rodegenerative conditions, and metabolic dysfunctions (7).
Interestingly, some studies have also suggested that creatine
supplementation may improve glucose metabolism, partic-
ularly when combined with exercise training (16).
Address for correspondence: Bruno Gualano, Ph.D., Av Mello de Moraes,
65–Butanta
˜, 05508-030, Sa
˜o Paulo, SP, Brazil; E-mail: gualano@usp.br.
Submitted for publication July 2010.
Accepted for publication September 2010.
Supplemental digital content is available for this article. Direct URL citations
appear in the printed text and are provided in the HTML and PDF versions of
this article on the journal’s Web site (www.acsm-msse.org).
0195-9131/11/4305-0770/0
MEDICINE & SCIENCE IN SPORTS & EXERCISE
Ò
Copyright Ó2011 by the American College of Sports Medicine
DOI: 10.1249/MSS.0b013e3181fcee7d
770
CLINICAL SCIENCES
Copyright © 2011 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.

It has been consistently demonstrated that creatine sup-
plementation along with muscle contraction can augment
muscle glycogen accumulation in humans (6,20,26). In ad-
dition, it has been reported (16) that creatine supplementa-
tion offsets the decline in muscle GLUT-4 expression after
2 wk of immobilization and increases GLUT-4 content
during subsequent rehabilitation training in healthy males.
Furthermore, creatine intake has been suggested (5) to
ameliorate hyperglycemia, typical of Huntington transgenic
mice, delaying the onset of diabetes. Supporting these
findings, a study (15) verified that creatine ingestion can
reduce the insulinogenic index in an animal model of in-
herited type 2 diabetes. Accordingly, we demonstrated that
creatine supplementation combined with aerobic training
promoted greater improvement on glucose tolerance than
aerobic training alone in physically inactive males (8). The
observations from the aforementioned studies suggest that
this nutritional intervention merits randomized controlled
trials to fully appreciate the possible therapeutic role of this
supplement in diabetes.
Thus, the aim of this study was to investigate the effects
of creatine supplementation combined with exercise train-
ing on glycemic control in type 2 diabetic patients.
METHODS
Subjects. Men and women (945 yr) prediagnosed with
type 2 diabetes, physically inactive for at least 1 yr, and
with BMI Q30 kgIm
j2
were eligible. The exclusion criteria
included use of exogenous insulin, uncontrolled hyperten-
sion (Q140/90 mm Hg), cardiovascular diseases, and/or
muscle skeletal disturbances that precluded exercise partici-
pation, vegetarian diet, previous use of creatine supplements,
glomerular filtration rate G40 mLIkg
j1
Imin
j1
, glycosylated
hemoglobin (Hb
A1c
)99%, and dyslipidemia. Patients’ charac-
teristics are presented in Table 1.
The study was approved by the local ethical committee,
and all subjects signed the written informed consent. This
trial was registered at ClinicalTrials.gov as NCT00992043.
Experimental protocol. A 12-wk, double-blind, ran-
domized, parallel-group, placebo-controlled trial was con-
ducted between October 2009 and January 2010 in Sao
Paulo (Brazil), according to the guidelines of the CONSORT
Statement.
Thepatientswererandomlyassigned (1:1) to receive
either creatine (CR) or placebo (PL) in a double-blind fash-
ion. We assigned patients to treatment sequence by using a
computer-generated randomization code with a block of eight
and stratified by gender. All the patients undertook a pro-
gram of moderate intensity aerobic training combined with
strengthening exercises for 3 months. The patients were
assessed at baseline (PRE) and after 12 wk (POST). The pri-
mary outcome was the glycemic control, as assessed through
Hb
A1c
concentrations. Secondary outcomes included the area
under the curve (AUC) of glucose, insulin, and C-peptide,
obtained from Meal Tolerance Tests (MTT), and insulin
sensitivity indexes. Muscle function and strength, aerobic
conditioning, body composition, and lipid profile were also
measured. Adverse events were recorded throughout the
trial. Possible differences in dietary intake were assessed by
means of three 24-h dietary recalls. In a subgroup of patients
randomly selected (n= 6 per group), muscle phosphoryl-
creatine content was measured through phosphorus magnetic
resonance spectroscopy (
31
P MRS), and muscle biopsies
were performed to assess GLUT-4 protein expression and
translocation.
Creatine supplementation protocol and blinding
procedure. The CR group received 5 gId
j1
of creatine
monohydrate throughout the trial. The PL group was given
the same dose of dextrose. The individuals consumed the
supplement as a single dose during their lunch. The sup-
plement packages were coded so that neither the inves-
tigators nor the participants were aware of the contents
until completion of the analyses. The compliance with the
supplementation was monitored weekly by asking the
patients personally. To verify the purity of the creatine used,
a sample was analyzed by high-performance liquid chro-
matography, and purity was established as 99.9%. The
supplementation was interrupted 72 h before the posttest
evaluation.
Exercise training program
The exercise program consisted of 12 wk of supervised
training. Exercise sessions occurred three times a week.
Training sessions consisted of a 5-min treadmill warm-up
followed by 25 min of resistance training, 30 min of treadmill
aerobic training, and 5 min of stretching exercises. All ses-
sions were monitored by at least one fitness professional. The
exercise program was performed in an intrahospital gymna-
sium (Laboratory of Assessment and Conditioning in Rheu-
matology, School of Medicine, University of Sao Paulo).
Resistance training included five exercises for the main
muscle groups: bench press, leg press, lat pulldown, leg ex-
tension, and seated row. Patients were required to perform four
sets of 8–12 repetitions maximum (RM), except during the first
TABLE 1. Patients’ characteristics.
Creatine
(n= 13)
Placebo
(n= 12)
P
(CR vs PL)
Gender (F/M) 8/5 8/4 0.56
Disease duration (years since diagnosis),
mean TSD
7T37T3 0.91
Age (yr), mean TSD 57.5 T5 56.4 T8.23 0.68
Systolic blood pressure
(mm Hg), mean TSD
125.0 T5.0 125.0 T5.0 0.92
Diastolic blood pressure
(mm Hg), mean TSD
85 T2.0 85 T1.0 0.88
Drugs, n(%)
Metformin 13 (100) 12 (100) 0.61
Sulfonylurea 7 (53.8) 6 (50) 0.58
A-blocker 2 (15.4) 2 (16.7) 0.67
ACE inhibitor 3 (23.1) 3 (25.0) 0.63
Angiotensin receptor antagonist 13 (100) 12 (100) 0.61
Thiazide 4 (30.8) 4 (33.3) 0.61
Statin 11 (84.6) 10 (83.3) 0.50
Fibrate 2 (15.4) 2 (16.7) 0.67
No significant differences were found (nonpaired t-test or Fisher exact test).
CREATINE AND DIABETES Medicine & Science in Sports & Exercise
d
771
CLINICAL SCIENCES
Copyright © 2011 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.

week, when a reduced volume of two sets of 15–20 RM for
each exercise was performed (as an adaptation period to re-
sistance training). Overload progression was implemented
when the subject could perform Q12 repetitions on the last
training set for two consecutive workouts. Aerobic training
intensity was set at the corresponding heart rate of approxi-
mately 70% of the V
˙O
2peak
.
Food intake assessment. Food intake was assessed
by means of three 24-h dietary recalls undertaken on sepa-
rate days (two weekdays and one weekend day) using a vi-
sual aid photo album of real foods. The 24-h dietary recall
consists of listing the foods and the beverages consumed
during 24 h before the recall. Energy and macronutrient in-
takes were analyzed by the Brazilian software Virtual Nutri
Ò
.
Glycosylated hemoglobin (Hb
A1c
). Hb
A1c
was mea-
sured using the BioRad Variant II automated analyzer
(BioRad, Irvine, CA).
MTT. The subjects were requested to refrain from inten-
sive physical activity 72 h before the MTT. After an over-
night fasting, the patients were given a 4-h meal challenge.
The mixed meal (500 kcal, 60% CHO, 20% fat, and 20%
protein) contained approximately 72 g of CHO. Blood
samples were collected at 0, 30, 60, 120, 180, and 240 min
for plasma glucose, insulin, and C-peptide for plasma glu-
cose, insulin, and C-peptide measurements. The former was
assessed through a colorimetric enzymatic assay (Bioclin,
Brazil), and the two latter were assessed using human-
specific radioimmunoassay techniques (Diagnostic Products
Corporation, CA, EUA). Fasting plasma insulin and glucose
concentrations were used to perform the G/I and the Homeo-
stasis Model Assessment (HOMAIR and HOMAB) indexes.
Strength and functional muscle assessments.
The patients underwent three familiarization sessions, sepa-
rated for at least 72 h, for all strength and functional tests.
Before the 1-RM test, two light warm-up sets interspaced
for 2 min were performed. Then, the patients had up to five
attempts to achieve the 1-RM load (e.g., maximum weight
that could be lifted once with proper technique), with a
3-min interval between attempts. One-RM tests were con-
ducted for bench press and leg press. Moreover, upper- and
lower-limb isometric strength was determined with a hand-
grip (dominant arm) and lower-back extension dynamometer,
respectively. Finally, we assessed balance, mobility, and
muscle function (timed-up-and-go and timed-stands tests),
according to previous descriptions (13). Training volume
(defined as total amount of weight lifted in each training
session) was calculated by multiplying the total number of
sets and repetitions by the amount of weight lifted in each set.
Body composition. Body composition and bone min-
eral density (BMD) were measured by dual x-ray absorp-
tiometry (DXA), using Hologic densitometry equipment
(Discovery model; Hologic, Inc., Bedford, MA) at the fol-
lowing regions: lumbar spine, femoral neck, total femur,
and whole body. All measurements were carried out by
the same trained technologist. Precision error for BMD
measurements was determined according to standard Inter-
national Society for Clinical Densitometry protocols. Least
significant changes with 95% confidence were 0.033 gIcm
j2
at the lumbar spine, 0.047 gIcm
j2
at the femoral neck,
0.039 gIcm
j2
at the total femur, and 0.020 gIcm
j2
at the
whole body.
Maximal oxygen consumption (V
˙O
2max
) tests.
All the subjects underwent a treadmill cardiopulmonary
test, before and after the intervention, according to the
conventional Bruce protocol. Attainment of V
˙O
2max
was
accepted when two of three criteria were met: a plateau
in V
˙O
2
, a respiratory exchange ratio 91.1, and/or volitional
exhaustion. The ventilatory anaerobic threshold (VAT) was
determined to occur at the break point between the increase
of carbon dioxide output (V
˙CO
2
) and V
˙O
2
. The respiratory
compensation point was determined to occur where the
ventilatory equivalent for carbon dioxide (V
˙
E
/V
˙CO
2
ratio)
was the lowest before a systematic increase.
Lipid profile. Serum concentrations of blood cholesterol,
HDL-cholesterol, and triglycerides were assessed by means
of colorimetric enzymatic methods (CELM, Brazil). From
these, VLDL-cholesterol (VLDL-cholesterol = triglycerides/5)
and LDL-cholesterol (LDL-cholesterol = total cholesterol j
[HDL-cholesterol + VLDL-cholesterol]) concentrations were
calculated. Serum apolipoproteins A1 and B and lipoprotein(a)
concentrations were determined by using immunoturbidimetric
assays (Roche Diagnostics, Germany). Serum apolipoproteins
A2 and E were measured through nephelometric assays
(Behring Diagnostics, Germany).
Muscle phosphorylcreatine content. In a subsam-
ple of patients (n= 6 per group), muscle phosphorylcre-
atine content was assessed in vivo by
31
P MRS using a
whole-body 3.0-T magnetic resonance imaging scanner
(Achieva Intera, Philips, Best, the Netherlands) and a 14-cm-
diameter
31
P surface coil. In brief, the surface coil was placed
centered under the calf muscle of the left leg. The scanner body
coil was used to obtain conventional anatomical T1-weighted
magnetic resonance images in the three orthogonal planes.
31
P MRS was acquired using the image selected in vivo spec-
troscopy sequence with an echo time and repetition time of
0.62 and 4500 ms, respectively. Spectrum bandwidth was
3000 Hz with 2048 data points and 64 repetitions. Spectrum
raw data were analyzed with Java Magnetic Resonance User
Interface software, and processing steps included apodization
to 5 Hz, Fourier transform, and phase correction. For spectrum
quantification, the AMARES algorithm was used taking into
account the previous knowledge of inorganic phosphate, pho-
sphodiester and phosphorylcreatine singlets, >-ATP and F-
ATP doublets, and A-ATP triplets. The phosphorylcreatine
signal was quantified relative to the A-ATP signal, assuming a
constant A-ATP concentration of 5.5 mmolIkg
j1
.
Muscle biopsies. In a subsample of patients (n= 6 per
group), muscle samples were obtained from the midportion
of the vastus lateralis using the percutaneous needle biopsy
technique with suction. Thereafter, an aliquot of each mus-
cle sample was immediately freed from blood and visible
connective tissue, rapidly frozen in liquid nitrogen, and
http://www.acsm-msse.org772 Official Journal of the American College of Sports Medicine
CLINICAL SCIENCES
Copyright © 2011 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.

stored at j80-C for subsequent analysis. The postintervention
biopsies were done through an adjacent incision to the baseline
site, 72 h after the last training session. All biopsies were car-
ried out after an 8-h overnight fast, and the last meal was a
standard dinner. Muscle biopsies were also obtained from
nondiabetic volunteers (n=6;age=60T5 yr, body mass
index = 31.6 T1.3 kgIm
j2
,Hb
A1c
=5.1%T0.3%, V
˙O
2max
=
25 T3mLIkg
j1
Imin
j1
) who were matched for age, gender,
body mass index, and V
˙O
2max
to the diabetic patients.
Cellular fractionation. Muscle samples were minced
and homogenized in ice-cold lysis buffer (2 mM EDTA,
10 mM EGTA, 0.25 M sucrose, 1:300 Sigma protease inhib-
itor cocktail, and 20 mM Tris–HCl at pH 7.5). The homoge-
nate was centrifuged at 100,000gfor 30 min (4-C). The
resulting pellet was dissolved in 1% Triton X-100 lysis buffer
and centrifuged again at 100,000gfor 30 min (4-C) to obtain
the nuclear pellet and the membrane fraction (supernatant).
Western blot analysis. GLUT-4 expression and translo-
cation were measured through Western blot analysis. Briefly,
samples were subjected to SDS-PAGE in polyacrylamide gel
(10%). After electrophoresis, proteins were electrotransferred
to the nitrocellulose membrane (BioRad Biosciences,
Piscataway, NJ). Equal loading of samples and transfer effi-
ciency were monitored with the use of 0.5% Ponceau S
staining of the blot membrane. The blotted membrane was
then blocked (5% nonfat dry milk, 10 mM Tris-HCl (pH 7.6),
150 mM NaCl, and 0.1% Tween 20) for 2 h at room temper-
ature and then incubated overnight at 4-C with specific anti-
bodies against GLUT-4 (Millipore, Bedford, MA). Binding of
the primary antibody was detected with the use of peroxidase-
conjugated secondary goat antirabbit antibody for 2 h at room
temperature and developed using enhanced chemilumines-
cence (Amersham Biosciences, Piscataway, NJ) detected by
autoradiography. Quantification analysis of blots was per-
formed with the use of Scion Image software (Scion based
on National Institutes of Health image). Total and membrane
fractions were normalized against glyceraldehyde-3-phosphate
dehydrogenase (GAPDH) and G
>o
.
Sample size. Before the commencement of the trial, we
determined that 24 patients would be needed to provide 80%
power (5% significance) to detect a difference of 0.6% be-
tween groups (similar effect expected with exercise training
alone (25)) in Hb
A1c
concentrations, which is the primary
outcome of this clinical trial. To account for midtrial with-
drawals, we enlarged our study population by approximately
15% to 28 participants.
Statistical analysis. Each comparison was by intention
to treat, irrespective of compliance with supplement intake.
Data were tested by a mixed model with repeated measures
using the software SAS version 9.1. A post hoc test adjusted
by Tukey was used for multicomparison purposes. Non-
repeated measures were compared using Student’s t-test
or Fisher exact test. Significance level was previously set at
PG0.05. Data are presented as mean, SD, estimated dif-
ference of means after intervention, and 95% confidence
interval (CI), except when otherwise stated.
RESULTS
Patients. The number of subjects recruited to the study
is shown in Supplemental Digital Content 1 (see Figure;
Fluxogram of participants; http://links.lww.com/MSS/A55).
Of the 121 people who responded to the initial request for
volunteers, 56 were screened and 28 met the inclusion cri-
teria. These patients were randomly assigned to either the
CR (n= 14) or PL (n= 14) groups. Three patients were
subsequently lost: two withdrew for personal reasons (one
from each group) and one was excluded because of an is-
chemic stroke episode in the first week of intervention (PL).
Then, 25 patients were analyzed (CR = 13, PL = 12).
Assessment of blinding, adherence to the exer-
cise program, and food intake. Five (38.4%) of the
patients correctly identified the supplement in the CR group,
whereas six (46.1%) patients were able to identify the cor-
rect supplement in the PL group (P= 0.57, Fisher exact test).
The adherence to the exercise program was 64.4% T
19.9% and 73.3% T19.8% for CR and PL groups, respec-
tively (P= 0.55, Student’s t-test).
Food intake did not significantly differ within or be-
tween groups (see Table Supplemental Digital Content 2;
Food intake at baseline and after the intervention;
http://links.lww.com/MSS/A56).
Muscle phosphorylcreatine content. There was
no significant difference between groups at baseline
(P= 0.77). After the intervention, the CR group presented
higher muscle phosphorylcreatine content when com-
pared with the PL group (CR: PRE = 44 T10 mmolIkg
j1
wet weight, POST = 70 T18; PL: PRE = 52 T13
POST = 46 T13 mmolIkg
j1
wet weight; P= 0.03; esti-
mated difference of means = 23.6 mmolIkg
j1
wet weight;
95% CI = 1.42–45.8 mmolIkg
j1
wet weight).
Primary outcome measure. Figure 1 demonstrates
Hb
A1c
concentrations. No significant difference between
groups was observed at baseline (P= 0.92). After the inter-
vention, Hb
A1c
concentrations were significantly reduced
only in the CR group compared with baseline (P=0.0001).
FIGURE 1—Effects of creatine supplementation combined with exer-
cise training in type 2 diabetic patients on Hb
A1c
concentrations. *Inter-
action effect (P= 0.004; estimated difference of means = j1.1%, 95%
CI = j1.9% to j0.4%). Data are means TSD. Mixed model for repeated
measures was used to compare placebo (n= 12) versus creatine (n=13).
CREATINE AND DIABETES Medicine & Science in Sports & Exercise
d
773
CLINICAL SCIENCES
Copyright © 2011 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.

A significant difference between groups was observed
(P= 0.004; estimated difference of means = j1.1%, 95%
CI = j1.9% to j0.4%).
Secondary outcome measures. Figure 2 shows the
effects of creatine supplementation on serum glucose, insu-
lin, and C-peptide concentrations.
The CR group presented a significant decrease in glycemia
at 0 (P=0.001,difference=j48 mgIdL
j1
,95%CI=j75 to
j21 mgIdL
j1
), 30 (P=0.004,difference=j48 mgIdL
j1
,
95% CI = j103 to j15 mgIdL
j1
), and 60 min (P=0.003,
difference = j68 mgIdL
j1
,95%CI=j117 to j19 mgIdL
j1
)
compared with the PL group. The remaining points were
unchanged. No significant differences were observed for
insulin and C-peptide concentrations during MTT.
The delta AUC for glucose concentration was significantly
lower in the CR group compared with the PL group (Fig. 3).
No significant differences were observed in delta AUC
for insulin and C-peptide concentrations (data not shown).
FIGURE 2—Effects of creatine supplementation combined with exercise training in type 2 diabetic patients on glucose (A and B), insulin (C and D),
and C-peptide (E and F) concentrations. Left panels represent creatine data, whereas right panels represent placebo data. *Interaction effect (group 
time) at time 0 (P= 0.001; difference = j48 mgIdL
j1
, 95% CI = j75 to j21 mgIdL
j1
), 30 (P= 0.004; difference = j48 mgIdL
j1
, 95% CI = j103 to
j15 mgIdL
j1
), and 60 (P= 0.003; difference = j68 mgIdL
j1
, 95% CI = j117 to j19 mgIdL
j1
). Data are means TSD. Mixed model for repeated
measures was used to compare placebo (n= 12) versus creatine (n= 13).
http://www.acsm-msse.org774 Official Journal of the American College of Sports Medicine
CLINICAL SCIENCES
Copyright © 2011 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.

There were no significant differences in HOMAIR, HOMAB,
and G/I indexes (see Table; Supplemental Digital Content
3; Effects of creatine supplementation combined with
exercise training on the surrogates of insulin sensitivity;
http://links.lww.com/MSS/A57).
Physical capacity, lipid profile, and body com-
position. Table 2 shows the effects of creatine supple-
mentation on physical capacity. Both groups presented
increased strength (main time effects) in the 1-RM leg press
and 1-RM bench press, and low back strength. Handgrip
strength tended to increase in both groups but did not reach
statistical significance (P= 0.06). Muscle function, as
assessed through the timed-stands test, was also significantly
improved in both groups, whereas the performance in the
timed-up-and-go remained unchanged.
V
˙O
2
correspondent to the ventilatory anaerobic threshold
(V
˙O
2
–VAT) and to the respiratory compensation point
(V
˙O
2
–RCP) were significantly increased in both groups
(mean time effects). V
˙O
2max
did not significantly change.
No significant differences were observed between the
groups for any physical capacity variable.
Total training volume did not significantly differ be-
tween the groups (CR = 3,128,294 T1,175,009 kg, PL =
2,171,652 T1,396,930 kg, P=0.11).
No significant differences were observed either in blood
lipoproteins or in blood apolipoproteins (see Table; Sup-
plemental Digital Content 4; Effects of creatine supple-
mentation combined with exercise training on lipid profile;
http://links.lww.com/MSS/A58). Also, there were no signif-
icant differences for body composition variables (see Table;
Supplemental Digital Content 5; Effects of creatine supple-
mentation combined with exercise training on body composi-
tion; http://links.lww.com/MSS/A59).
GLUT-4 expression and translocation. Figure 4
expresses muscle GLUT-4 content and translocation data.
There was no significant difference in muscle GLUT-4
content between the diabetic patients and the healthy
individuals. In addition, Muscle GLUT-4 content was sig-
nificantly unchanged after the intervention (P= 0.91).
At baseline, membrane GLUT-4 content was significantly
lower in the diabetic patients compared with the healthy sub-
jects. After the intervention, membrane GLUT-4 content was
significantly raised in the CR and PL groups (P= 0.0008) so
that differences between the diabetic patients and healthy
subjects were no longer observed (P= 0.92). Following the
same trend, the membrane–total GLUT-4 content ratio was
increased in both experimental groups (P= 0.0001). However,
the CR group presented a greater increase than the PL in
FIGURE 3—Effects of creatine supplementation combined with exercise training in type 2 diabetic patients on the area under the curve of glucose
concentration. A, Individual data for AUC. B, Means TSD for the delta area under the curve of glucose concentration. *Significant treatment effect
compared with placebo (P= 0.05). Student’s t-test was used to compare placebo (n= 12) versus creatine (n= 13).
TABLE 2. Effects of creatine supplementation combined with exercise training on physical capacity in type 2 diabetic patients.
Creatine (n= 13) Placebo (n= 12)
Difference (CI 95%) P(CR vs PL)
Variable Pre Post Pre Post
1-RM bench press (kg)
a
56 T24 64 T27 53 T12 67 T93(j19 to 25) 0.52
1-RM leg press (kg)
b
54 T17 68 T26 60 T11 73 T13 j2(j25 to 21) 0.99
Handgrip (kg)
c
34 T10 36 T11 36 T937T10 5 (j3 to 13) 0.82
Low-back strength (kg)
d
95 T40 108 T26 102 T24 112 T23 10 (j21 to 43) 0.49
Timed-stands test (rep)
e
14 T218T215T217T20(j2 to 2) 0.48
Timed-up-and-go test (s) 6.8 T0.8 6.6 T0.6 6.4 T1.0 5.6 T0.8 0.8 (0.1 to 1.6) 0.21
V
˙O
2
–VAT (mLIkg
j1
Imin
j1
)
f
16 T317T217T319T3j1(j4 to 1) 0.68
V
˙O
2
–RCP (mLIkg
j1
Imin
j1
)
g
19 T322T426T630T6j8(j16 to 0) 0.69
V
˙O
2max
(mLIkg
j1
Imin
j1
)24T425T530T531T7j5(j13 to 3) 0.87
Data expressed as mean TSD, estimated mean of differences (95% CI) and level of significance (P) between CR and PL (mixed model for repeated measures).
Symbols represent main time effects (
a
P= 0.007,
b
P= 0.0018,
c
P= 0.06,
d
P= 0.03,
e
P= 0.0001,
f
P= 0.0004,
g
P= 0.08). No significant differences were found between groups at
baseline.
RM, repetition maximum; rep, repetitions; V
˙O
2
–VAT, oxygen consumption correspondent to ventilator anaerobic threshold; V
˙O
2
–RCP, oxygen consumption correspondent to respiratory
compensation point; V
˙O
2
, maximal oxygen consumption.
CREATINE AND DIABETES Medicine & Science in Sports & Exercise
d
775
CLINICAL SCIENCES
Copyright © 2011 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.

membrane GLUT-4 content as well as in the membrane–total
GLUT-4 content ratio (P=0.05andP= 0.03, respectively).
Adverse events. No severe adverse events in the CR
group were observed. Symptoms such as nausea, diar-
rhea, and cramps were reported by some patients. How-
ever, these events did not significantly differ between
the groups (see Table; Supplemental Digital Content 6;
Adverse events in the creatine and the placebo groups;
http://links.lww.com/MSS/A60).
DISCUSSION
This is the first randomized controlled trial to describe the
beneficial effects of creatine supplementation on glycemic
control in type 2 diabetic patients who underwent exercise
training. As such, this supplement emerges as a valuable
nonpharmacological approach for treating diabetes.
There is evidence showing that creatine supplementation
improves insulin sensitivity in a rodent model of inherited
type 2 diabetes (15) and delays the onset of diabetes in
Huntington transgenic mice (5). In healthy young individu-
als, creatine supplementation has been proven to increase
muscle glycogen accretion and glucose tolerance, particularly
when combined with exercise training. Altogether, these
findings provided the rationale for investigating the thera-
peutic role of creatine supplementation in diabetic patients.
As a consequence of creatine supplementation, we ob-
served a significant increase in muscle phosphorylcreatine
content (n= 6) and subsequent improvement in glycemic
control. Importantly, the mean reduction in Hb
A1c
(j1.1%)
was superior to that commonly seen with exercise training
(25) or metformin (10) treatment alone, pointing out the
therapeutic potential of this novel nutritional intervention.
The glycemia-lowering effect of this supplement is most
likely an explanation for this metabolic response. On the
other hand, no change in insulinemia was verified, which is
apparently in contrast to observations from in vitro studies,
which indicated creatine-induced insulin secretion in incu-
bated mouse islets (1,12). Furthermore, it has been previ-
ously observed (21) that creatine supplementation induces
hyperinsulinemia in rats, which could ultimately lead to
disruption in glucose homeostasis. However, studies in
humans have recurrently refuted such creatine-induced
hyperinsulinemia (8,14,26). The current findings extrapolate
this idea to diabetic patients.
Regarding mechanisms, we showed suboptimal GLUT-4
translocation in the diabetic patients compared with their
healthy peers, corroborating early evidence that insulin re-
sistance in type 2 diabetes is not generally associated with
decreased muscle GLUT-4 content (9). In opposite, it is
speculated that insulin stimulation fails to induce normal
GLUT-4 protein translocation in skeletal muscle from these
patients (22,28). In this respect, it is worth noting that ex-
ercise training was able to resolve impaired GLUT-4 trans-
location in the diabetic patients. Interestingly, this response
was further enhanced by creatine supplementation, sug-
gesting that this supplement acts directly on type 2 diabetes
pathogenesis (i.e., suboptimal GLUT-4 translocation),
thereby ameliorating hyperglycemia and consequently gly-
cemic control.
On the basis of the current knowledge, it is difficult to
reveal the molecular basis for the increase in muscle GLUT-
4 translocation that occurred as a consequence of creatine
FIGURE 4—Effects of creatine supplementation combined with exer-
cise training in type 2 diabetic patients on muscle GLUT-4 content (B),
membrane GLUT-4 content (C), and membrane–total GLUT-4 ratio
(D). Representative immunoblots are shown in panel A, Dotted lines
represent mean data from healthy subjects (n= 6). #Mean time effects
for membrane GLUT-4 content (P= 0.0008) and membrane–total
GLUT-4 ratio (P= 0.0001). *Interaction effects for membrane GLUT-4
content (P= 0.05) and membrane–total GLUT-4 ratio (P= 0.03). Data
are means TSD and are expressed relative to the baseline value that
was set to be equal to 1. Mixed model for repeated measures was used to
compare placebo (n= 12) versus creatine (n= 13).
http://www.acsm-msse.org776 Official Journal of the American College of Sports Medicine
CLINICAL SCIENCES
Copyright © 2011 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.

supplementation. In this regard, previous observation from
nondiabetic models indicates that creatine supplementation
may upregulate (i) protein kinase B mRNA expression,
which is suggested to promote GLUT-4 translocation to the
sarcolemma (23); (ii) insulin-like growth factor-1 mRNA
and protein content, which is believed to enhance insulin
action (4); and (iii) nuclear content and DNA binding ac-
tivity of myocyte enhancer factor-2 isoforms, which are
transcription factors that regulate GLUT-4 expression in
muscle (11). The influence of creatine supplementation on
the insulin-signaling pathway and transcription factors of
GLUT-4 in type 2 diabetes is currently under investigation
in our laboratory.
It is also interesting to note that only the CR group ex-
perienced improvements in glycemic control, although both
groups had undertaken an exercise training program. The
American Diabetic Association has recommended that ex-
ercise training should be performed at least 3 dIwk
j1
and
with no more than two consecutive days without physical
activity to improve glycemic control (24). Although our
exercise program was previously designed to achieve this
recommendation, the patients failed to properly accomplish
it (see the adherence data). Moreover, the short duration of
the training protocol can be also responsible for the lack of
improvements in glycemic control. Thus, it is possible to
assume that our exercise training program alone did not
reach sufficient frequency and/or volume levels to promote
improvements in glycemic control, although significant
changes in physical capacity (i.e., muscle function, aerobic
condition, and strength) were observed in both groups. In
this scenario, it is tempting to speculate that the addition of
creatine supplementation might have maximized the effects
of exercise on insulin sensitivity and glycemic control.
Nonetheless, it is impossible to distinguish whether the
current findings result from the creatine treatment per se or
the interaction between creatine and exercise training. On
one hand, some authors have speculated that dietary crea-
tine-associated changes in CHO metabolism are a result of
an interaction between creatine supplementation and other
mediators of muscle glucose transport, such as muscle con-
traction. On the other hand, recent evidence suggests that
creatine per se can modulate the expression of key proteins
and genes related to insulin sensitivity and glycemic control
(e.g., GLUT-4, protein kinase B, myocyte enhancer factor-2,
insulin-like growth factor-1) (23). Indeed, further inves-
tigations are needed to address this question.
Despite some evidence indicating that creatine supple-
mentation may be capable of improving lipid profile, lean
mass, and strength (2,7), we did not observe such adapta-
tions. In fact, it was expected from the outset that a lower
adaptive response for lipid profile and BMD would be
shown by our patients because their baseline values were
within the reference range. Compelling evidence has indi-
cated no direct benefit of creatine supplementation on mus-
cle protein synthesis (17), suggesting that major mechanism
by which chronic creatine supplementation exerts its effects
on body composition and athletic performance is by en-
hancing training volume at every training session. Because
we did not observe higher training volume, the lack of
changes in strength and lean mass was also expected.
Moreover, creatine supplementation seems to be highly
safe based on data from short-term and long-term human
studies and results from several (18,19) therapeutic trials. In
this trial, the adverse effects were comparable between
groups, further supporting this notion and extending it to
diabetic patients.
This study presented some limitation. First, considering
the sample’s characteristics, these findings cannot be ex-
trapolated to patients with poorer glycemic control, older
age, or at a different pharmacological treatment (e.g., ex-
ogenous insulin therapy). Second, the patients underwent
exercise training along with creatine intervention, thus we
cannot also generalize these results to physically inactive
individuals. Third, it currently is unknown how long the
effects of creatine may persist. Studies evaluating the effi-
cacy and safety of long-term creatine supplementation are
needed. Likewise, the optimal creatine protocol (i.e., high vs
low dose, short vs long term, continuous vs cycled regimen)
remains to be determined.
In summary, we reported a novel therapeutic role of cre-
atine supplementation on metabolic control in type 2 dia-
betic patients. Moreover, we provided convincing evidence
that creatine supplementation might modulate glucose up-
take in these patients mainly via an increase in GLUT-4
recruitment to the sarcolemma.
Bruno Gualano, Guilherme G. Artioli, and Eloisa Bonfa receive
support from Conselho Nacional de Desenvolvimento Cientı´ fico e
Tecnolo´ gico (CNPq).
The authors declare that they do not have conflict of interests.
The authors inform that the results of the present study do
not constitute endorsement by the American College of Sports
Medicine.
REFERENCES
1. Alsever RN, Georg RH, Sussman KE. Stimulation of insulin se-
cretion by guanidinoacetic acid and other guanidine derivatives.
Endocrinology. 1970;86(2):332–6.
2. Andres RH, Ducray AD, Schlattner U, Wallimann T, Widmer HR.
Functions and effects of creatine in the central nervous system.
Brain Res Bull. 2008;76(4):329–43.
3. Branch JD. Effect of creatine supplementation on body composi-
tion and performance: a meta-analysis. Int J Sport Nutr Exerc
Metab. 2003;13(2):198–226.
4. Deldicque L, Louis M, Theisen D, et al. Increased IGF mRNA in
human skeletal muscle after creatine supplementation. Med Sci
Sports Exerc. 2005;37(5):731–6.
CREATINE AND DIABETES Medicine & Science in Sports & Exercise
d
777
CLINICAL SCIENCES
Copyright © 2011 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.

5. Ferrante RJ, Andreassen OA, Jenkins BG, et al. Neuroprotective
effects of creatine in a transgenic mouse model of Huntington’s
disease. J Neurosci. 2000;20(12):4389–97.
6. Green AL, Hultman E, Macdonald IA, Sewell DA, Greenhaff PL.
Carbohydrate ingestion augments skeletal muscle creatine accu-
mulation during creatine supplementation in humans. Am J Phys-
iol. 1996;271(5 Pt 1):E821–6.
7. Gualano B, Artioli GG, Poortmans JR, Lancha AH Jr. Exploring
the therapeutic role of creatine supplementation. Amino Acids.
38(1):31–44.
8. Gualano B, Novaes RB, Artioli GG, et al. Effects of creatine
supplementation on glucose tolerance and insulin sensitivity in
sedentary healthy males undergoing aerobic training. Amino Acids.
2008;34(2):245–50.
9. Henriksen EJ. Invited review: effects of acute exercise and exercise
training on insulin resistance. J Appl Physiol. 2002;93(2):788–96.
10. Johansen K. Efficacy of metformin in the treatment of NIDDM.
Meta-analysis. Diabetes Care. 1999;22(1):33–7.
11. Ju JS, Smith JL, Oppelt PJ, Fisher JS. Creatine feeding increases
GLUT4 expression in rat skeletal muscle. Am J Physiol Endocrinol
Metab. 2005;288(2):E347–52.
12. Marco J, Calle C, Hedo JA, Villanueva ML. Glucagon-releasing
activity of guanidine compounds in mouse pancreatic islets. FEBS
Lett. 1976;64(1):52–4.
13. Newcomer KL, Krug HE, Mahowald ML. Validity and reliability
of the timed-stands test for patients with rheumatoid arthritis and
other chronic diseases. J Rheumatol. 1993;20(1):21–7.
14. Newman JE, Hargreaves M, Garnham A, Snow RJ. Effect of cre-
atine ingestion on glucose tolerance and insulin sensitivity in men.
Med Sci Sports Exerc. 2003;35(1):69–74.
15. Op’t Eijnde B, Jijakli H, Hespel P, Malaisse WJ. Creatine sup-
plementation increases soleus muscle creatine content and lowers
the insulinogenic index in an animal model of inherited type 2
diabetes. Int J Mol Med. 2006;17(6):1077–84.
16. Op’t Eijnde B, Urso B, Richter EA, Greenhaff PL, Hespel P. Effect
of oral creatine supplementation on human muscle GLUT4 protein
content after immobilization. Diabetes. 2001;50(1):18–23.
17. Paddon-Jones D, Borsheim E, Wolfe RR. Potential ergogenic
effects of arginine and creatine supplementation. J Nutr. 2004;
134(10 suppl):2888–94S; discussion 95S.
18. Persky AM, Rawson ES. Safety of creatine supplementation.
Subcell Biochem. 2007;46:275–89.
19. Poortmans JR, Francaux M. Adverse effects of creatine supple-
mentation: fact or fiction? Sports Med. 2000;30(3):155–70.
20. Robinson TM, Sewell DA, Casey A, Steenge G, Greenhaff PL.
Dietary creatine supplementation does not affect some haemato-
logical indices, or indices of muscle damage and hepatic and renal
function. Br J Sports Med. 2000;34(4):284–8.
21. Rooney K, Bryson J, Phuyal J, Denyer G, Caterson I, Thompson
C. Creatine supplementation alters insulin secretion and glucose
homeostasis in vivo.Metabolism. 2002;51(4):518–22.
22. Ryder JW, Yang J, Galuska D, et al. Use of a novel impermeable
biotinylated photolabeling reagent to assess insulin- and hypoxia-
stimulated cell surface GLUT4 content in skeletal muscle from
type 2 diabetic patients. Diabetes. 2000;49(4):647–54.
23. Safdar A, Yardley NJ, Snow R, Melov S, Tarnopolsky MA. Global
and targeted gene expression and protein content in skeletal muscle
of young men following short-term creatine monohydrate supple-
mentation. Physiol Genomics. 2008;32(2):219–28.
24. Sigal RJ, Kenny GP, Wasserman DH, Castaneda-Sceppa C. Physi-
cal activity/exercise and type 2 diabetes. Diabetes Care.2004;27
(10):2518–39.
25. Thomas DE, Elliott EJ, Naughton GA. Exercise for type 2 diabetes
mellitus. Cochrane Database Syst Rev. 2006;3:CD002968.
26. van Loon LJ, Murphy R, Oosterlaar AM, et al. Creatine sup-
plementation increases glycogen storage but not GLUT-4 ex-
pression in human skeletal muscle. Clin Sci (Lond).2004;106(1):
99–106.
27. Wyss M, Kaddurah-Daouk R. Creatine and creatinine metabolism.
Physiol Rev. 2000;80(3):1107–213.
28. Zierath JR, He L, Guma A, Odegoard Wahlstrom E, Klip A,
Wallberg-Henriksson H. Insulin action on glucose transport and
plasma membrane GLUT4 content in skeletal muscle from patients
with NIDDM. Diabetologia. 1996;39(10):1180–9.
http://www.acsm-msse.org778 Official Journal of the American College of Sports Medicine
CLINICAL SCIENCES
Copyright © 2011 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.