Embed Size (px)
Citation preview
Clinical Endocrinology (2006)
65
, 722–728 doi: 10.1111/j.1365-2265.2006.02658.x
© 2006 The Authors
722
Journal compilation © 2006 Blackwell Publishing Ltd
O R I G I N A L A R T I C L E
Blackwell Publishing Ltd
Effects of pioglitazone and metformin on plasma adiponectin in newly detected type 2 diabetes mellitus
Pramod Kumar Sharma*, Anil Bhansali†, Ravinder Sialy†, Samir Malhotra* and Promila Pandhi*
Departments of
*
Pharmacology and
†
Endocrinology, PGIMER, Chandigarh, India
Summary
Objective
This prospective study evaluates the effect of insulin
sensitizers, pioglitazone (PGZ) and metformin (MET) on plasma
adiponectin and leptin levels in subjects newly diagnosed with type
2 diabetes mellitus (T2DM).
Design
Double blind, randomized, active control, dose escalation
study of 12 weeks treatment duration.
Patients
Thirty apparently healthy, treatment-naive T2DM patients
diagnosed within the past 6 months.
Measurements
Plasma adiponectin and leptin levels were estimated
by enzyme-linked immunosorbent assay (ELISA), and insulin resist-
ance by the homeostasis model of assessment (HOMA-IR).
Results
Baseline plasma levels of adiponectin were lower in diabetic
(
n
= 30) subjects than matched controls (
n
= 10, 6·6
±
1·1
vs
10·4
±
4·2
µ
g/ml,
P
= 0·021). The 12-week treatment with PGZ
significantly increased adiponectin concentrations (6·6
±
1·1–
17·9
±
7·4
µ
g/ml,
P
< 0·001) with no alteration in the MET treated
group (6·8
±
1·5–6·7
±
2·8
µ
g/ml,
P
= 0·9). A significant decrease in
plasma leptin levels was observed in the MET treated group
(32·0
±
28·9–21·4
±
23·3 ng/ml,
P
= 0·024) but not in the PGZ
treated group (23·9
±
24·1–22·4
±
25·4 ng/ml,
P
= 0·69). The altera-
tions in plasma adiponectin and leptin levels were not associated
with any change in body mass index (BMI). PGZ therapy improved
insulin sensitivity to a greater degree (
P =
0·007 and
P
= 0·001 for
fasting plasma insulin (FPI) and HOMA-IR, respectively) than MET
(
P =
0·75 and
P
= 0·02 for FPI and HOMA-IR, respectively) but this
improvement was not significantly different from that of MET at the
end of 12 weeks (
P =
0·146 and
P
= 0·09 for FPI and HOMA-IR,
respectively). However, improvement in insulin sensitivity with PGZ
was not commensurate with the increase in adiponectin. Better
control of postbreakfast plasma glucose (PBPG) as well as decrease
in serum triglycerides (TGs) were also seen with PGZ (PBPG,
P
< 0·001; TGs,
P
= 0·013). The rest of the parameters were com-
parable. Adverse reactions reported were minor and did not result
in treatment discontinuation.
Conclusions
Pioglitazone therapy appears to be better in achieving
glycaemic control and increasing plasma adiponectin and insulin
sensitivity in newly detected type 2 diabetics.
(Received 28 March 2006; returned for revision 20 April 2006; finally
revised 20 May 2006; accepted 14 July 2006)
Introduction
Adipose tissue is now recognized as an endocrine organ, which
secretes several mediators with diverse functions collectively known
as adipocytokines.
1
One such adipocytokine, adiponectin has been
identified
2,3
and accounts for 0·01–0·03% of total plasma proteins.
4
Studies in animals
5
and humans
6
suggest a role for this cytokine in
the regulation of insulin–glucose homeostasis. Decreased levels of
adiponectin are independently associated with insulin sensitivity
5,6
and are negatively correlated with plasma glucose, fasting plasma
insulin (FPI), triglycerides (TGs)
7
and body mass index (BMI).
4
Patients with diabetes have lower circulating adiponectin concentra-
tions than their age- and sex-matched counterparts
8
and levels in
lean subjects without diabetes are even higher.
9
It has also been sug-
gested that adiponectin might function as an adipostat in regulating
energy homeostasis and that its deficiency might contribute to the
development of obesity and type 2 diabetes mellitus (T2DM).
10
The
putative mechanism by which liver and muscle sensitivity to insulin
is enhanced by adiponectin is less well understood. However, animal
studies
11,12
suggest an enhancement of insulin stimulated tyrosine
phosphorylation of insulin receptors in skeletal muscles and reduced
expression of hepatic glucogenic enzymes may be underlying
mechanisms.
Thiazolidinediones (TZDs) are widely used in the treatment of
T2DM. It has been well established that TZDs enhance insulin
sensitivity by binding with peroxisome proliferator activated
receptors-
γ
(PPAR-
γ
).
13,14
Predominantly PPAR-
γ
is expressed in fat
cells and improvement in insulin sensitivity occurs mainly in skeletal
muscles,
15,16
where PPAR-
γ
is sparsely expressed. This paradox suggests
that TZDs may indeed modulate signalling between fat and muscle
cells and it may involve adiponectin
11
and other adipocytokines.
Studies have shown that TZDs increase mRNA expression and
adiponectin secretion dose dependently.
17
A consistent increase in
serum adiponectin and improvement in insulin sensitivity has been
observed with pioglitazone (PGZ),
8,18
rosiglitazone,
19
and troglitazone
9,7
Correspondence: Dr Anil Bhansali, Department of Endocrinology, PGIMER, Chandigarh, India 160012. Tel.: +91 172 2756583; Fax: +91 172 2744401; E-mail: [email protected]
Insulin sensitizers and plasma adiponectin in new type 2 diabetes
723
© 2006 The AuthorsJournal compilation © 2006 Blackwell Publishing Ltd,
Clinical Endocrinology
,
65
, 722–728
suggesting a class effect of PPAR-
γ
agonists as all of them lead to a
reduction in hepatic fat content, an effect correlated with increase
in adiponectin concentration.
20
Metformin (MET), a biguanide
compound improves peripheral insulin sensitivity and increases
insulin mediated skeletal muscle glucose uptake
21
but has been
shown not to affect circulating adiponectin concentrations.
9,20
To our knowledge, however, there is no randomized controlled
trial comparing the effects of PGZ and MET on circulating adiponec-
tin levels in new onset T2DM. Therefore, our working hypothesis
was that these two treatments are not different as far as effect on
plasma adiponectin concentrations in new onset T2DM patients are
concerned. To test this hypothesis, we, therefore, directly compared
PGZ with MET in a short-term randomized controlled trial of
12 weeks treatment duration, using circulating plasma adiponectin
concentrations as a main outcome measure in newly diagnosed
T2DM patients.
Materials and methods
Thirty consecutive patients with T2DM were recruited from the out-
patient Diabetes Clinic of the Institute. All the subjects recruited were
newly diagnosed (duration
≤
6 months), apparently healthy and
treatment naive. Written informed consent was obtained from all
subjects before participation in the study. The Institutional ethics
committee approved the study protocol. Criteria for patients’ inclu-
sion were as follows:
• diagnosis of T2DM based on American Diabetes Association
(ADA) criteria;
22
• aged between 30 and 60 years;
• BMI = 23 kg/m
2
.
Patients exclusion criteria were:
• diabetes was secondary to another cause;
• presence of ketonuria;
• severe concurrent infectious illness;
• impaired renal function (serum creatinine > 132
µ
mol/l in men
and > 124
µ
mol/l in women);
• pulmonary insufficiency with hypoxaemia;
• impaired hepatic function (> 1·5 times the upper limit of normal);
• congestive heart failure;
• history of alcohol abuse;
• pregnant and lactating women.
Study design
The study was a prospective, double blind, randomized controlled
trial of 12 weeks treatment duration. Four weeks before randomiza-
tion the diagnosis was confirmed by measuring plasma glucose
(fasting and postprandial). The eligible patients were instructed to
consume a calorie restricted diet as per the requirement appropriate
to their BMI and activity level. This protocol was followed for up to
4 weeks. One week before randomization all laboratory investiga-
tions were completed. On the day of randomization, subjects were
asked to report fasting in the morning and plasma glucose was again
measured. Those who despite dietary control had fasting plasma
glucose (FPG)
≥
7·8 mmol/l were enrolled. At this time a blood sample
was taken for biochemistry, HbA1c, leptin, adiponectin, insulin, and
lipid profile. These subjects were assigned, in a double blind manner,
to receive either PGZ 15 mg/d, or MET 1000 mg/d for the initial
3 weeks. Random allocation was achieved by simple randomization
technique using random number table. Medications were dispensed
in identical capsules in labelled envelopes. Both medications were
administered twice daily. Daily dose of PGZ was given in the morn-
ing followed by placebo in the evening whereas daily MET dose was
administered in two divided doses. The doses were increased every
third week until the fasting plasma glucose (FPG) was reduced to
< 6·0 mmol/l or when trial medications reached maximum recom-
mended dose (30 mg for PGZ; 2 g for MET). To achieve euglycaemia,
gliclazide-Mr (30–60 mg/d) was added to both the arms in open
manner from the beginning or anytime during the study, based on
physician’s discretion.
Subjects were evaluated every third week for body weight, blood
pressure and relevant clinical examination and FPG, 2 h postbreak-
fast plasma glucose (PBPG) after a standard breakfast (two bread
slices/oatmeal/two chapatis and a cup of tea). On each visit
adherence to dietary instructions was reinforced and compliance to
therapy checked. At 12 weeks all baseline measurements were
repeated. Patients who did not complete the follow up were excluded
from the study.
Assays
All measurements were carried out with an appropriate kit according
to the manufacturers instructions. Total cholesterol (TC), triglycerides
(TGs) and HDL-cholesterol were measured by (Randox Laboratories
Ltd., Antrim, UK) kits and blood glucose was measured by glucose
oxidase method using AUTOPAK of Bayer Diagnostics India Ltd.
(Baroda, India) LDL-cholesterol was calculated from the Friedwald
equation. Plasma insulin was estimated by immunoradiometric assay
(IRMA) (IMMUNOTECH, Prague, Czech Republic) and adiponectin
by sandwich enzyme immunoassay (Quantikine, R and D System
Inc. Minneapolis, USA). The minimum detection limit of adiponectin
was 0·079 ng/ml. Plasma leptin was measured by sandwich ELISA
assay (DRG, GmbH, Marburg, Germany) with a minimum detection
limit of 1 ng/ml. The intra- and inter-assay coefficients of variation
were 3·5 and 6·5%, respectively. HbA1c was measured by the method
of Postmes
et al
.
23
Insulin resistance was measured by homeostasis model (HOMA-
IR)
24
from fasting plasma glucose (FG) and insulin (FPI) concen-
trations as:
FG (mmol/l)
×
FPI (mIU/l)/22·5
Statistical analysis
All analysis was carried out on SPSS statistical package (SPSS,
Version 10·0 for Windows, Chicago, IL, USA). Data are presented as
the mean
±
SD unless otherwise specified. Intra- and inter-group
differences were calculated using appropriate parametric paired and
unpaired
t
-test and nonparametric Wilcoxon sign rank and Mann–
Whitney
U
-tests after checking for data distribution. For correlation
coefficients between selected variables Spearman’s or Pearsons’s
correlation statistics was used. Adiponectin levels were adjusted for
724
P. K. Sharma
et al.
© 2006 The AuthorsJournal compilation © 2006 Blackwell Publishing Ltd,
Clinical Endocrinology
,
65
, 722–728
baseline variables, e.g. BMI, sex, and blood urea nitrogen (BUN)
using multiple regression analysis. To reveal actual effect of treat-
ments on PBPG (both treatment group and daily gliclazide included
as covariate in regression analysis followed by controlling for
gliclazide effect in both the treatment group). Test significance was
declared at
P
< 0·05.
Results
The patient recruitment and screening is summarized in Fig. 1.
Demography and baseline parameters are summarized in Table 1.
The treated groups were comparable at baseline. Mean daily dose of
PGZ and MET consumed in the study was 21·9 mg and 1291 mg,
respectively.
Glycaemic control
The fasting and PBPG levels were significantly lower at the end of
12 weeks of treatment with PGZ as well as MET (Table 2). PBPG
reduction was significantly greater at 12 weeks in the PGZ than the
MET group despite correcting for the effect of concomitant gliclazide
in both groups using multiple regression analysis (Table 2). By
12 weeks there was a nonsignificant fall in HbA1c levels in both the
groups (Table 2).
Body mass index
There was no alteration in BMI at the end of 12 weeks of therapy
and the two groups were not different from each other (
P =
0·96).
Plasma insulin and lipids
Circulating insulin levels decreased with both treatments. Whereas
PGZ therapy decreased it significantly (
P =
0·007), the decline with
MET was not significant (
P =
0·75). Despite respective falls of 47%
and 9% in FPI, the two treatments were not different at the end of
12 weeks (Table 2). Fasting TGs concentrations decreased significantly
with PGZ treatment compared to baseline (Tables 2,
P
= 0·006) but such
an effect was not observed in the MET arm (
P =
0·47). Additionally,
at the end of 12 weeks TGs levels were significantly lower in the PGZ
group than in the MET group (Tables 2,
P
= 0·013). Circulating TC,
HDL, and LDL levels in both groups were unremarkable.
HOMA-IR
Twelve weeks treatment with PGZ led to a greater decrease (
P =
0·001)
in HOMA-IR than MET treatment (
P =
0·02). However, the decrease
in HOMA-IR was not statistically different between groups at the
end of 12 weeks (
P =
0·09).
Adiponectin
At baseline, plasma adiponectin concentrations were lower in
diabetic subjects compared to values in 10 healthy age-, sex- and
BMI-matched controls (10·42
±
4·24
µ
g/ml, data not shown,
P
= 0·021).
At the end of 12 weeks, adiponectin levels significantly increased
Fig. 1 Trial summary.
Table 1. Comparative baseline clinical characteristics in patients of type 2 diabetes randomized to pioglitazone and metformin
Parameters
Pioglitazone
(n = 15)
Metformin
(n = 15) P-value
No. (M/F) 8/7 10/5 –
Age (years) 50·8 ± 6·9 47·7 ± 9·5 0·32
Weight (kg) 71·9 ± 8·3 70·7 ± 9·9 0·7
BMI (kg/m2) 27·9 ± 3·4 28·6 ± 3·9 0·49
FPG (mmol/l) 10·4 ± 2·7 9·2 ± 4·0 0·14
PBPG (mmol/l) 12·9 ± 3·7 13·5 ± 3·6 0·52
HbA1c (%) 7·7 ± 1·1 8·03 ± 0·9 0·42
FPI (pmol/l) 88·2 ± 60·2 76·0 ± 54·5 0·6
HOMA-IR 5·4 ± 3·3 4·2 ± 3·5 0·2
Leptin (ng/ml) 23·9 ± 24·1 32·0 ± 28·9 0·41
(range) (2·0–76·0) (4·1–86·0)
Adiponectin (µg/ml) 6·6 ± 1·1 6·8 ± 1·5 0·33
SBP (mmHg) 128·5 ± 13·2 119·6 ± 15·3 0·09
DBP (mmHg) 81·1 ± 6·6 78·7 ± 10·2 0·45
TC (mmol/l) 5·7 ± 1·5 5·54 ± 1·2 0·91
TGs (mmol/l) 2·0 ± 1·0 2·5 ± 2·3 0·5
HDL (mmol/l) 1·2 ± 0·2 1·17 ± 0·2 0·8
LDL (mmol/l) 3·6 ± 1·4 3·3 ± 0·8 0·9
Values are expressed as mean ± SD.BMI, body mass index; FPG, fasting plasma glucose; PBPG, post-breakfast plasma glucose; FPI, fasting plasma insulin; HOMA-IR, homeostasis model assessment insulin resistance; SBP, systolic blood pressure; DBP, diastolic blood pressure; TC, total cholesterol; TG, triglycerides; HDL, LDL, high density and low density lipoproteins.
Insulin sensitizers and plasma adiponectin in new type 2 diabetes 725
© 2006 The AuthorsJournal compilation © 2006 Blackwell Publishing Ltd, Clinical Endocrinology, 65, 722–728
from baseline in the PGZ group (P < 0·001, Table 2) but no such
change was observed in the MET treated group. The difference in
adiponectin levels between two treatment groups at the end of
12 weeks was significant (P < 0·001). The coefficients of correlation
between baseline adiponectin levels either with BMI or with HOMA-
IR were −0·28 and −0·3, respectively, and were nonsignificant
(P = 0·14 and 0·1, respectively).
Leptin
At baseline, leptin levels were higher in patients (n = 30) than healthy
matched controls (n = 10; 28·0 ± 26·5 vs 4·1 ± 3·4 ng/ml, P < 0·001).
At 12 weeks, leptin concentrations decreased significantly in the
MET group (P = 0·024) and no change was detected in the PGZ
treatment arm (P = 0·41).
Leptin, adiponectin and correlation with parameters of insulin resistance
BMI, FPI and HOMA-IR showed nonsignificant but negative cor-
relation with baseline plasma adiponectin concentrations (r = –0·28,
–0·22 and −0·3; P = 0·14, 0·24 and 0·1, respectively) but significant
and positive correlation with baseline leptin levels (r = 0·75, 0·50 and
0·58, P < 0·001, 0·001 and 0·001, respectively). However, this correla-
tion could not be sustained neither with MET nor with PGZ treatment
after 12 weeks of therapy.
Compliance and adverse drug reactions
The compliance was good and comparable in both the treatment
groups. Adverse effects were recorded using combination of open and
closed methods. Gastrointestinal side-effects were the commonest
adverse effects in metformin group. Eight patients in PGZ group
and three patients in MET group reported weight gain (2–7 kg,
median 4·25 kg for PGZ; 1–6 kg, median 1·5 kg for MET) but there
was no difference in mean weight in two treatment arms at 12 weeks
(Table 2, P = 0·38). Three patients developed mild lower limb
oedema in pioglitazone group. No apparent changes were observed
in the haematological or biochemical parameters but one patient in
the PGZ group showed mild increase in liver enzymes (< three-fold).
Symptoms suggestive of hypoglycaemia were verbally communi-
cated by three patients on follow up visit in the PGZ group. However
none of these could be documented.
Discussion
In this short-term study of 12 weeks treatment duration, we have
confirmed that PGZ therapy increased plasma adiponectin levels
with no effect on plasma leptin whereas MET treatment decreased
leptin levels significantly from baseline. No alteration in anthro-
pometry was demonstrated in either treatment groups. However,
the improvement in insulin sensitivity as demonstrated by decrease
in FPI and HOMA-IR was significant in the PGZ-treated group but
not the MET-treated group.
The mean daily doses of PGZ and MET in the study were 21·9 mg
and 1291 mg, respectively. Twelve subjects in the PGZ and nine in
the MET group used gliclazide (mean daily doses were 26·3 and
17·8 mg in the PGZ and MET groups, respectively, P = 0·17). At
12 weeks, only one patient each in both the treatment groups was
taking 60 mg of gliclazide and the remaining patients were using
30 mg/day. Both the treatments resulted in good glycaemic control.
The mean reduction in FPG was 30% and 42·5% (P = 0·22), whereas
PBPG was reduced by 41·2% and 47·2% (P ≤ 0·001) with MET and
PGZ, respectively. Despite the comparable fasting glucose control,
PGZ treatment was more effective than MET in decreasing PBPG
levels even after discounting the effect of gliclazide in both groups
Table 2. Changes in various parameters compared to baseline in response to treatment with pioglitazone and metformin
Parameters
Pioglitazone (n = 15) Metformin (n = 15) P-values b/w
groups at
12 weeksBaseline 12 week P-value Baseline 12 week P-value
Weight (kg) 71·9 ± 8·3 73·2 ± 10·3 0·28 70·7 ± 9·9 69·9 ± 9·7 0·43 0·38
BMI (kg/m2) 27·9 ± 3·4 28·3 ± 4·0 0·33 28·6 ± 3·9 28·3 ± 4·0 0·41 0·96
FPG (mmol/l) 10·4 ± 2·7 6·0 ± 1·4 0·001 9·2 ± 4·0 6·4 ± 1·2 0·008 0·221
PBPG (mmol/l) 12·9 ± 3·7 6·8 ± 0·6 0·000 13·5 ± 3·6 7·9 ± 0·7 0·008 0·000
HbA1c (%) 7·72 ± 1·1 7·30 ± 0·8 0·34 8·03 ± 0·9 7·56 ± 0·8 0·14 0·43
HOMA-IR 5·4 ± 3·3 1·7 ± 1·0 0·001 4·2 ± 3·5 2·6 ± 1·6 0·02 0·09
FPI (pmol/l) 88·2 ± 60·0 47·0 ± 28·0 0·007 76·0 ± 54·5 69·0 ± 45·0 0·75 0·146
Leptin (ng/ml) 23·9 ± 24·1 22·4 ± 25·4 0·69 32·0 ± 28·9 21·4 ± 23·3 0·024 0·9
(range) (2·0–76·0) (2·0–72·0) (4·1–86·0) (2·2–73·0)
Adiponectin (ug/ml) 6·6 ± 1·1 17·9 ± 7·4 0·001 6·8 ± 1·5 6·7 ± 2·8 0·82 0·000
TC (mmol/l) 5·7 ± 1·5 5·0 ± 1·0 0·125 5·5 ± 1·2 5·4 ± 0·9 0·71 0·3
TGs (mmol/l) 2·0 ± 1·0 1·3 ± 0·3 0·006 2·5 ± 2·3 2·0 ± 0·9 0·47 0·013
HDL (mmol/l) 1·19 ± 0·2 1·23 ± 0·2 0·4 1·17 ± 0·2 1·2 ± 0·2 0·29 0·69
LDL (mmol/l) 3·6 ± 1·4 3·2 ± 1·1 0·349 3·3 ± 0·8 3·3 ± 0·9 0·85 0·75
Gliclazide MR (mg/d) – 26·3 ± 16·6 – – 17·8 ± 16·8 – 0·17
Values are expressed as mean ± SD.
726 P. K. Sharma et al.
© 2006 The AuthorsJournal compilation © 2006 Blackwell Publishing Ltd, Clinical Endocrinology, 65, 722–728
(Table 2). The greater efficacy of TZDs to increase glucose disposal
may be due, at least in part, to enhanced insulin action not only in
muscle but in adipose tissue as well and improvement in both hepatic
and peripheral insulin sensitivity.8 Treatment with PGZ increased
adiponectin levels by 2·7-fold without any alterations in BMI. No
such effect was observed with MET therapy. These results are in
agreement with previous findings, that TZDs treatment is associated
with an increase in circulating adiponectin levels with concomitant
improvement in insulin sensitivity8,9,17,18 as also observed by a
decrease in FPI and HOMA-IR in the present study. Decrease in
HOMA-IR and FPI was observed in the MET arm as well, but this
was nonsignificant compared to the PGZ treated group. The increase
in circulating adiponectin by PGZ is related to enhanced production
by smaller adipocytes (as preadipocytes are differentiated into
smaller and metabolically more active ones) and decreased lipo-
toxicity.25 Adiponectin in turn exerts insulin sensitizing effect
through AMP-activated protein kinase (AMPK) in liver, muscles,
and adipocytes26,27 improving tissue lipid oxidation resulting in
reduced lipotoxicity.28
Consistent with previous observations up-regulation of adiponec-
tin was not seen with MET.9,21 The failure of metformin treatment
to affect adiponectin concentrations, suggests that perhaps it is not
the alteration in the plasma glucose and insulin but may be a direct
action on the adipose tissues that regulates adiponectin production.9
Moreover, metformin does not affect glucose transport or insulin
signalling in adipocytes.9,25 In fact, MET may suppress adiponectin
production and release from differentiated adipocytes through stimu-
lation of AMPK.29 MET induces fatty acid oxidation and decreases
hepatic gluconeogenesis secondary to activation of AMPK,30 an
effect possibly similar to physiological effect of adiponectin on lipid
and glucose metabolism; one could argue that MET does not need
to stimulate adiponectin secretion because MET directly stimulates
AMPK.30 This argument might help to explain the results obtained
with MET therapy in our patients. Moreover, MET administration
might lead to suppression of adiponectin synthesis independent of
improvement in fasting hyperglycaemia.31
It seems unlikely that the mechanism behind increased adiponec-
tin with PGZ is secondary to improved glycaemic control as com-
parable glycaemic control was also observed with MET. In our study,
increase in adiponectin concentration with PGZ treatment did not
correlate with change in FPG and HbA1c; these findings indicate that
the improvement in glucose control may not be the underlying
mechanism as far as the increase in plasma adiponectin is concerned.
Disease states, like diabetes are characterized by elevated concentra-
tions of TNF-α, known to be a contributing factor for insulin resistance
in these patients.32 An alternative possibility is that the direct
stimulation of adipocytes by PGZ through PPAR-γ to increase
mRNA expression of adiponectin17 and reduction in plasma and
tissue TNF-α concentration,8,33,34 which is reported to reduce the
expression and secretion of adiponectin in human preadipocytes.32
Reduction in TNF-α levels, may at least in part be secondary to
increase in plasma adiponectin.35 Kern et al.36 found a good inverse
correlation between plasma adiponectin and TNF-α mRNA expres-
sion in adipose tissue and also reported that lean individuals have
higher adipose tissue adiponectin expression with lower TNF-αexpression. In addition, improvement in insulin sensitivity by TZDs
may not be proportional to increase in plasma adiponectin suggest-
ing that increase in adiponectin may not be the sole mechanism in
improving insulin sensitivity,37 and results of our study support this
notion. Moreover, insulin might have an independent effect on
adiponectin modulation from adipocytes.
After treatment with MET, there was a significant decrease in
leptin levels even without alterations in anthropometry. This has
been demonstrated earlier that with improvement in glycaemic con-
trol and insulin sensitivity, leptin levels decrease. Despite of decrease
in leptin levels by MET, improvement in insulin sensitivity was not
commensurate with its parallel decrease, as demonstrated by
HOMA-IR and FPI, thereby suggesting that decrease in leptin levels
does not influence the insulin sensitivity in a short period of time.
In contrast, PGZ therapy did not affect circulating leptin levels as
reported earlier8,35 but troglitazone has been shown to reduce leptin
levels significantly in T2DM.38 The reasons for such contradictory
observations need to be explored.
PGZ was found to be significantly better than MET in decreasing
TGs from baseline by 35% compared to MET at 12 weeks. However,
percentage reduction is appreciably larger than reported earlier
with PGZ as monotherapy39 or in combination with sulphonylureas.40
The effects of TZDs on TGs have been somewhat more variable
owing to partial PPAR-α agonistic activity of PGZ, whereas rosigli-
tazone seems to be a pure PPAR-γ agonist. 41 Despite reducing serum
free fatty acids concentrations to a similar extent,42 a decrease in TGs
was observed more often with PGZ. The decrease in TGs with PGZ
is attributed to the reduced availability of free fatty acids to the liver
for very low density lipoprotein (VLDL) synthesis, consequently
resulting in a decrease in circulating TGs. No improvement in HDL
concentrations may in part be related to the lower doses of PGZ used
by our patients (≤ 30 mg/d).
Long-term outcome benefits of MET on all cause mortality and
cardiovascular disease have been substantiated in United Kingdom
Prospective Diabetes Study (UKPDS).43 However the effects of PGZ
on cardiovascular disease outcome has also been highlighted in the
PROactive study (PROspective pioglitAzone Clinical Trial In Macro
Vascular Events),44 but a small sample size, short-term intervention
trial like present study does not establish the superiority of PGZ over
MET.
In conclusion, this short-term study shows PGZ to be superior to
MET as far as increase in adiponectin and insulin sensitivity is con-
cerned. However, until the increased adiponectin levels with PGZ are
sustained and translated into reduction in relevant clinical events,
further long-term studies are warranted.
Acknowledgements
Token of thanks to Novo Nordisk India Ltd. for making the adi-
ponectin kit available for the study.
References
1 Pittas, A.G., Joseph, N.A. & Greenberg, A.S. (2004) Adipocytokinesand insulin resistance. Journal of Clinical Endocrinology and Metab-olism, 89, 447–452.
2 Scherer, P.E., Williams, S., Fogliano, M., Baldini, G. & Lodish, H.F.
Insulin sensitizers and plasma adiponectin in new type 2 diabetes 727
© 2006 The AuthorsJournal compilation © 2006 Blackwell Publishing Ltd, Clinical Endocrinology, 65, 722–728
(1995) A novel protein similar to C1q, produced exclusively in adi-pocytes. Journal of Biological Chemistry, 270, 26746–26749.
3 Hu, E., Liang, P. & Spiegelman, B.M. (1996) AdipoQ is a noveladipose-specific gene dysregulated in obesity. Journal of BiologicalChemistry, 271, 10697–10703.
4 Arita, Y., Kihara, S., Ouchi, N., Takahashi, M., Maeda, K., Miyagawa, J.,Hotta, K., Shimomura, I., Nakamura, T. & Miyaoka, K. (1999)Parodoxical decrease of an adipose specific protein, adiponectinin obesity. Biochemistry and Biophysics Research Communications,257, 79–83.
5 Hotta, K., Funahashi, T., Bodkin, N.L., Ortmeyer, H.K., Arita, Y.,Hansen, B.C. & Matsuzawa, Y. (2001) Circulating concentrations ofthe adipocyte protein adiponectin are decreased in parallel withreduced insulin sensitivity during the progression to type 2 diabetesin rhesus monkeys. Diabetes, 50, 1126–1133.
6 Weyer, C., Funahashi, T., Tanaka, S., Hotta, K., Matsuzawa, Y.,Pratley, R.E. & Tataranni, P.A. (2001) Hypoadiponectinemia in obes-ity and type 2 diabetes: close association with insulin resistance andhyperinsulinemia. Journal of Clinical Endocrinology and Metabolism,86, 1930–1935.
7 Hotta, K., Funahashi, T., Arita, Y., Takahashi, M., Mastuda, M.,Okamoto, Y., Iwasashi, H., Kuriyama, H., Ouchi, N. & Maeda, K.(2000) Plasma concentrations of a novel, adipose-specific protein,adiponectin, in type 2 diabetic patients. Arteriosclerosis Thrombosisand Vascular Biology, 20, 1595–1599.
8 Miyazaki, Y., Mahankali, A., Wajcberg, E., Bajaj, M., Mandarino, L.J.& DeFronzo, R.A. (2004) Effect of pioglitazone on circulating adi-pocytokine levels and insulin sensitivity in type 2 diabetic patients.Journal of Clinical Endocrinology and Metabolism, 89, 4312–4319.
9 Phillips, S.A., Ciaraldi, T.P., Kong, A.P.S., Bandukwala, R., Aroda, V.,Carter, L., Baxi, S., Mudaliar, S.R. & Henry, R.R. (2003) Modulationof circulating and adipose tissue adiponectin level by antidiabetictherapy. Diabetes, 52, 667–674.
10 Statnick, M.A., Beavers, L.S., Conner, L.J., Corominola, H., Johnson, D.,Hammond, C.D., Rafaeloff-Phail, R., Seng, T., Suter, T.M., Sluka, J.P.,Ravussin, E., Gadski, R.A. & Caro, J.F. (2000) Decrease expressionof apM1 in omental and subcutaneous tissue of human with type 2diabetes. International Journal of Reproduction Diabetes Research, 1,51–58.
11 Yamauchi, T., Kamon, J., Waki, H., Terauchi, Y., Kubota, N., Hara, K.,Mori, Y., Ide, T., Murakami, K., Tsuboyama-Kasaoka, N., Ezaki, O.,Akanuma, Y., Gavrilova, O., Vinson, C., Reitman, M.L., Kagechika, H.,Shudo, K., Yoda, M., Nakano, Y., Tobe, K., Nagai, R., Kimura, S.,Tomita, M., Froguel, P. & Kadowaki, T. (2001) The fat derived hor-mone adiponectin reverses insulin resistance associated with bothlipoatrophy and obesity. Nature Medicine, 7, 941–946.
12 Persseghin, G., Lattuada, G., Danna, M., Sereni, L.P., Maffi, P.,De Cobelli, F., Battezzati, A., Secchi, A., Del Maschio, A. & Luzi, L.(2003) Insulin resistance, intramyocellular lipid content, and plasmaadiponectin in patients with diabetes. American Journal of Physiology,285, E1174–E1181.
13 Day, C. (1999) Thiazolidinediones: a new class of antidiabetic drugs.Diabetes Medicine, 16, 179–192.
14 Tontonoz, P., Hu, E. & Speigelman, B.M. (1995) Regulation of adi-pocytes gene expression and differentiation by peroxisome prolifer-ator activated receptor gamma. Current Opinions in Genetics andDevelopment, 5, 571–576.
15 Inzucchi, S.E., Maggs, D.G., Spollett, G.R., Page, S.L., Rife, F.S.,Walton, V. & Shulman, G.I. (1998) Efficacy and metabolic effects ofmetformin and troglitazone in type II diabetes mellitus. New EnglandJournal of Medicine, 338, 867–872.
16 Peterson, K.F., Krssak, M., Inzucchi, S., Cline, G.W., Dufour, S. &Shulman, G.I. (2000) Mechanism of troglitazone action in type 2diabetes. Diabetes, 49, 827–831.
17 Maeda, N., Takahashi, M., Funahashi, T., Kihara, S., Nishizawa, H.,Kishida, K., Nagaretani, H., Matsuda, M., Komuro, R., Ouchi, N.,Kuriyama, H., Hotta, K., Nakamura, T., Shimomura, I. & Matsuzawa, Y.(2001) PPAR-γ ligands increase expression and plasma concentrationsof adiponectin, an adipose-derived protein. Diabetes, 50, 2094–2099.
18 Hirose, H., Kawai, T., Yamamoto, Y., Taniyama, M., Tomita, M.,Matsubara, K., Okazaki, Y., Ishii, T., Oguma, Y., Takei, I. & Saruta, T.(2002) Effects of pioglitazone on metabolic parameters, body fat dis-tribution and serum adiponectin levels in Japanese male patientswith type 2 diabetes. Metabolism, 51, 314–317.
19 Combs, T.P., Wagner, J.A., Berger, J., Doebber, T., Wang, W.-J.,Zhang, B.B., Tanen, M., Berg, A.H., O’Rahilly, S. & Savage, D.B.(2002) Induction of adipocyte complement–related protein of 20kilo Daltons by PPAR gamma agonists: a potential mechanism ofinsulin sensitization. Endocrinology, 143, 998–1007.
20 Tiikkainen, M., Hakkinen, A.M., Korsheninnikova, E., Nayman, T.,Makimattila, S. & Yki-Jarvinen, H. (2004) Effects of rosiglitazoneand metformin on liver fat content, hepatic insulin resistance, insulinclearance, and gene expression in adipose tissue in patients with type2 diabetes. Diabetes, 53, 2169–2176.
21 Cusi, K. & DeFronzo, R.A. (1998) Metformin: a review of its metaboliceffects. Diabetes Review, 6, 89–131.
22 American diabetes Association: Diagnosis and Classification of Dia-betes Mellitus. (2004) Diabetes Care, 27, S5–S10.
23 Postmes, T.J., Halders, S., Herpers, H., Sels, J.P. & Coenegracht, J.M.(1981) Colorimetry vs chromatography in HbA1 assay. ClinicalChemistry, 27, 635–636.
24 Matthews, D.R., Hosker, J.P., Rudenski, A.S., Naylor, B.A., Treacher, D.F.& Turner, R.C. (1985) Homeostasis model assessment: insulin resist-ance and β-cell function from fasting plasma glucose and insulinconcentration in man. Diabetologia, 28, 412–419.
25 Ciaraldi, T.P., Kong, A.P.S., Chu, N.V., Kim, D.D., Baxi, S., Loviscach, M.,Plodkowski, R., Reitz, R., Caulfield, M., Mudaliar, S. & Henry, R.R.(2002) Regulation of glucose transport and insulin signaling by tro-glitazone or metformin in adipose tissue of type 2 diabetic subjects.Diabetes, 51, 30–36.
26 Berg, A.H., Du Combs, T.P.X., Brownlee, M. & Scherer, P.E. (2001)The adipocyte secreted protein Acrp30 enhances hepatic insulinaction. Nature Medicine, 7, 947–953.
27 Wu, X.D., Motoshima, H., Mahadev, K., Stalker, T.J., Scalia, R. &Goldstein, B.J. (2003) Involvement of AMP-activated protein kinasein glucose uptake stimulated by the globular domain of adiponectinin primary rat adipocytes. Diabetes, 52, 1355–1363.
28 Kadowaki, T. & Yamauchi, T. (2005) Adiponectin and adiponectinreceptors. Endocrine Reviews, 26, 439–451.
29 Huypens, P., Quartier, E., Pipeleers, D. & Van de Casteele, M. (2005)Metformin reduces adiponectin protein expression and release in3T3-L1 adipocytes involving activation of AMP activated proteinkinase. European Journal of Pharmacology, 518, 90–95.
30 Zhou, G., Myers, R., Li, Y., Chen, Y., Shen, X., Fenyk-Melody, J., Wu, M.,Ventre, J., Doebber, T., Fujii, N., Musi, N., Hirshman, M.F.,Goodyear, L.J. & Moller, D.E. (2001) Role of AMP-activated proteinkinase in mechanism of metformin action. Journal of Clinical Inves-tigations, 108, 1167–1174.
31 Fujita, H., Fujishima, H., Koshimura, J., Hosoba, M., Yoshioka, N.,Shimotomai, T., Morii, T., Narita, T., Kakei, M. & Ito, S. (2005) Effectsof antidiabetic treatment with metformin and insulin on serum and
728 P. K. Sharma et al.
© 2006 The AuthorsJournal compilation © 2006 Blackwell Publishing Ltd, Clinical Endocrinology, 65, 722–728
adipose tissue adiponectin levels in db/db mice. Endocrine Journal,52, 427–433.
32 Kappes, A. & Loffler, G. (2000) Infuences of ionomycin, dibutyryl-cyclo-AMP and tumour necrosis factor-alpha on intracellular amountand secretion of apM1 in differentiating primary human preadipocytes.Hormone Metabolism Research, 32, 548–554.
33 Hofmann, C., Lorenz, K., Braithwaite, S.S., Colca, J.R., Palazuk, B.J.,Hotamisligil, G.S. & Spiegelman, B.M. (1994) Altered gene expres-sion for tumour necrosis factor-alpha and its receptors during drugand dietary modulation of insulin resistance. Endocrinology, 134,264–270.
34 Murase, K., Odaka, H., Suzuki, M., Tayuki, N. & Ikeda, H. (1998)Pioglitazone time-dependently reduces tumour necrosis factor-alphalevel in muscles and improves metabolic abnormalities in Wistarfatty rats. Diabetologia, 41, 257–264.
35 Shimizu, H., Oh, I.S., Tsuchiya, T., Othani, K.-I., Okada, S. & Mori, M.(2006) Pioglitazone increases circulating adiponectin levels andsubsequently reduces TNF-á levels in type 2 diabetic patients:a randomized study. Diabetes Medicine, 23, 253–257.
36 Kern, P.A., Di Gregorio, G.B., Lu, T., Rassouli, N. & Ranganathan, G.(2003) Adiponectin expression from human adipose tissue: relationto obesity, insulin resistance, and tumor necrosis factor-α expression.Diabetes, 52, 1779–1785.
37 Rasouli, N., Yao-Borengasser, A., Miles, L.M., Elbein, S.C. & Kern, P.A.(2006) Increased plasma adiponectin in response to pioglitazonedoes not result from increased gene expression. American Journal ofPhysiological Endocrinology and Metabolism, 290, E42–E46.
38 Shimizu, H., Tsuchiya, T., Sato, N., Shimomura, Y., Kobayashi, I. &Mori, M. (1998) Troglitazone reduces plasma leptin concentration,but increase hunger in NIDDM patients. Diabetes Care, 21, 1470–1474.
39 Aronoff, S., Rosenblatt, S., Braithwaite, S., Egan, J.W., Mathisen, A.L.& Schneider, R.L. (2000) Pioglitazone hydrochloride monotherapy
improves glycaemic control in the treatment of patients with type 2diabetes: a 6 months randomized placebo-controlled dose–responsestudy. The Pioglitazone 001 Study Group. Diabetes Care, 23, 1605–1611.
40 Kipnes, M.S., Krosnick, A., Rendell, M.S., Egan, J.W., Mathisen, A.L.& Schneider, R.L. (2001) Pioglitazone hydrochloride in combinationwith sulphonylurea therapy improves glycaemic control in patientswith type 2 diabetes mellitus: a randomized placebo-controlledstudy. American Journal of Medicine, 111, 10–17.
41 Sakamoto, J., Kimura, H., Moriyama, S., Odaka, H., Momose, Y.,Sugiyama. Y. & Sawada, H. (2000) Activation of human peroxisomeproliferator-activated receptor (PPAR) subtypes by pioglitazone.Biochemistry and Biophysics Research Communications, 278, 704–711.
42 Miyazaki, Y., Mahankali, A., Mastuda, M., Glass, L., Mahankali, S.,Ferrannini, E., Cusi, K., Mandarino, L.J. & DeFronzo, R.A. (2001)Improved glycaemic control and enhanced insulin sensitivity in type2 diabetic subjects treated with pioglitazone. Diabetes Care, 24, 710–719.
43 United Kingdom Prospective Diabetes Study Group. (1998) Effectof intensive blood-glucose control with metformin on complicationsin overweight patients with type 2 diabetes (UKPDS 34). Lancet, 352,854–865.
44 Dormandy, J.A., Charbonnel, B., Eckland, D.J., Erdmann, E., Massi-Benedetti, M., Moules, I.K., Skene, A.M., Tan, M.H., Lefebvre, P.J.,Murray, G.D., Standl, E., Wilcox, R.G., Wilhelmsen, L., Betteridge,J., Birkeland, K., Golay, A., Heine, R.J., Koranyi, L., Laakso, M.,Mokan, M., Norkus, A., Pirags, V., Podar, T., Scheen, A., Scherbaum,W., Schernthaner, G., Schmitz, O., Skrha, J., Smith, U., Taton, J. &PROactive investigators (2005) Secondary prevention of macro-vascular events in patients with type 2 diabetes in the PROactiveStudy (PROspective pioglitAzone Clinical Trial In macroVascularEvents): a randomised controlled trial. Lancet, 366, 1279–1289.
