Association of adiponectin and resistin with adipose tissue compartments, insulin resistance and dyslipidaemia

Abstract

In this study, we investigated the association of plasma adiponectin and resistin concentrations with adipose tissue compartments in 41 free-living men with a wide range of body mass index (22-35 kg/m(2)).
Using enzyme immunoassays, plasma adiponectin and resistin were measured. Intraperitoneal, retroperitoneal, subcutaneous abdominal and posterior subcutaneous abdominal adipose tissue masses (IPATM, RPATM, SAATM and PSAATM, respectively) were determined using magnetic resonance imaging. Total adipose tissue mass (TATM) was measured using bioelectrical impedance. Insulin resistance was estimated with the help of homeostasis model assessment (HOMA) score.
In univariate regression, plasma adiponectin levels were inversely related to IPATM (r = -0.389, p < 0.05), SAATM (r = -0.500, p < 0.001), PSAATM (r = -0.502, p < 0.001), anterior SAATM (r = -0.422, p < 0.01) and TATM (r = -0.421, p < 0.01). In multiple regression models, adiponectin was chiefly correlated with PSAATM. Plasma adiponectin concentrations were also inversely correlated with HOMA score (r = -0.540, p < 0.001) and triglyceride (r = -0.632, p < 0.001), and positively correlated with high-density lipoprotein cholesterol (r = 0.508, p < 0.001). There were no significant correlations between resistin levels and adipose tissue masses, insulin resistance or dyslipidaemia.
In men, total body fat is significantly correlated with plasma adiponectin, but not with plasma resistin levels. Low plasma adiponectin levels appear to be chiefly determined by the accumulation of posterior subcutaneous abdominal fat mass, as opposed to intra-abdominal fat, and are strongly predictive of insulin resistance and dyslipidaemia.

Full text

Association of adiponectin and resistin with adipose tissue
compartments, insulin resistance and dyslipidaemia
M. S. Farvid,
1
T. W. K. Ng,
2
D. C. Chan,
2
P. H. R. Barrett
2
and G. F. Watts
2
1
School of Public Health, Tehran University of Medical Sciences, Tehran, Iran
2
School of Medicine and Pharmacology, Western Australia Institute of Medical Research, University of Western Australia, Royal
Perth Hospital, Perth, Australia
Aim: In this study, we investigated the association of plasma adiponectin and resistin concentrations with adipose
tissue compartments in 41 free-living men with a wide range of body mass index (22–35 kg/m
2
).
Methods: Using enzyme immunoassays, plasma adiponectin and resistin were measured. Intraperitoneal, retro-
peritoneal, subcutaneous abdominal and posterior subcutaneous abdominal adipose tissue masses (IPATM, RPATM,
SAATM and PSAATM, respectively) were determined using magnetic resonance imaging. Total adipose tissue mass
(TATM) was measured using bioelectrical impedance. Insulin resistance was estimated with the help of homeostasis
model assessment (HOMA) score.
Results: In univariate regression, plasma adiponectin levels were inversely related to IPATM (r ¼0.389, p <0.05),
SAATM (r ¼0.500, p <0.001), PSAATM (r ¼0.502, p <0.001), anterior SAATM (r ¼0.422, p <0.01) and TATM
(r ¼0.421, p <0.01). In multiple regression models, adiponectin was chiefly correlated with PSAATM. Plasma
adiponectin concentrations were also inversely correlated with HOMA score (r ¼0.540, p <0.001) and triglyceride
(r ¼0.632, p <0.001), and positively correlated with high-density lipoprotein cholesterol (r ¼0.508, p <0.001).
There were no significant correlations between resistin levels and adipose tissue masses, insulin resistance or
dyslipidaemia.
Conclusions: In men, total body fat is significantly correlated with plasma adiponectin, but not with plasma resistin
levels. Low plasma adiponectin levels appear to be chiefly determined by the accumulation of posterior subcutaneous
abdominal fat mass, as opposed to intra-abdominal fat, and are strongly predictive of insulin resistance and dys-
lipidaemia.
Keywords: adiponectin, dyslipidaemia, fat compartments, insulin resistance, resistin
Received 4 April 2004; returned for revision 18 May 2004; revised version accepted 18 May 2004
Introduction
Adiponectin is an adipocyte-derived plasma protein with
important insulin-sensitizing properties. Clinical studies
have, accordingly, suggested that adiponectin plays an
important role in the regulation of insulin resistance, glu-
cose and lipid homeostasis [1–3]. Molecular studies have
also shown that in insulin resistance and obesity, the
expression of adiponectin is decreased in adipose tissue
[4]. The association of low adiponectin levels with obes-
ity, insulin resistance, coronary artery disease and dysli-
pidaemia indicates that this protein may be an important
new marker of the metabolic syndrome. Resistin is
another adipocytokine recently identified in mouse and
Correspondence:
Prof Gerald F. Watts, School of Medicine and Pharmacology, The Western Australia Institute for Research, University of Western
Australia, Royal Perth Hospital, GPO Box X2213, Perth, WA 6847, Australia.
E-mail:
gfwatts@cyllene.uwa.edu.au
ORIGINAL ARTICLE doi: 10.1111/j.1463–1326.2004.00410.x
406 Diabetes, Obesity and Metabolism, 7, 2005, 406–413 #2005 Blackwell Publishing Ltd

in human fat cells [5,6]. Initial studies on mice suggested
that resistin is increased in obesity and may be involved
in the development of insulin resistance [7]. Human stud-
ies also suggested that resistin might link obesity with
insulin resistance and diabetes [8,9].
The importance of obesity and body fat distribution in
the development of insulin resistance and metabolic
syndrome is well recognized [10]. In particular, the vis-
ceral fat compartment has been suggested by most studies
to be a more important contributor to insulin resistance
and the metabolic syndrome than subcutaneous fat
[11,12]. However, studies on body fat distribution and
its related effects on insulin resistance and dyslipidaemia
have been conflicting at best, with evidence for both the
posterior subcutaneous [13] and the intravisceral fat
compartment [10–12]. The study by Misra et al. [13] has
been the only one to specifically address the relation-
ship of posterior subcutaneous abdominal adipose tissue
mass (PSAATM) with insulin resistance.
The association between the distribution of adipose tis-
sue and the levels of plasma adiponectin and resistin has
not been adequately investigated. This question is import-
ant, because these adipocytokines may mediate the effects
of body fat distributions on insulin resistance and cardio-
vascular risk factors. The primary aim of this study was,
therefore, to investigate the strength and independence of
the association of plasma adiponectin and resistin concen-
trations with adipose tissue compartments, measured prin-
cipally using magnetic resonance imaging (MRI), in men
with a wide range of body mass index (BMI).
People and Methods
People
We studied 41 non-smoking Caucasian men selected
from the community with a BMI ranging between 22
and 35 kg/m
2
. People with a history of familial dys-
lipidaemia, intercurrent illness and medical disorders
or taking drugs known to affect lipid metabolism were
excluded. All men were consuming ad libitum, weight
maintenance diets and had been advised by a qualified
dietician to continue an isocaloric intake for 4 weeks.
They were studied at the end of this period if their body
weight, measured serially, varied by <3%. Volunteers
gave written consent and the study was approved by the
ethics committee of the Royal Perth Hospital, Australia.
Protocols
Weight was measured in light clothing without shoes
after emptying bladder. Height was measured as the dis-
tance from the top of the head to the bottom of the feet
(no shoes) by using a fixed stadiometre. BMI was calcu-
lated as the weight (kg) divided by the square of the
height (m). Waist circumference (cm) was taken with a
tape measure as the point midway between the costal
margin and iliac crest in the mid-axillary line, with the
man standing and breathing normally. Hip circumfer-
ence (cm) was measured at the widest point around the
greater trochanter. The waist-to-hip (WHR) ratio was
calculated as the waist measurement divided by the
hip measurement. All measurements in the metabolic
ward were performed after a 14-h fast in a temperature-
controlled room.
Measurement of Total Adipose Tissue Mass and
Fat-Free Mass
The assessment of body composition was determined at
rest in the supine position after emptying bladder by
using a Holtain Body Composition Analyser (Holtain
Ltd, Dyfed, UK). This technique is based on the princi-
ple that the conduction of applied electrical current
results in an impedance to the spread of the current
that is dependent on frequency. Both intracellular and
extracellular fluid compartments and cell membranes
act as electrical conductors and capacitors, respectively.
Low-frequency and high-frequency currents flow
through various pathways in the human body. Low-
frequency currents pass through the extracellular fluid
compartment, whereas high-frequency currents flow
through both fluid compartments. Based on this princi-
ple of bioelectrical impedance, total adipose tissue mass
(TATM) and fat-free mass (FFM) can be determined
using formulas derived from Abate et al. [14] and
Lukaski et al. [15], respectively (FFM ¼(0.85 H
2
/
Z) þ3.04), where H is the height (cm) of the man and Z
is the impedance. For this measure, the men were asked
to fast overnight and to refrain from alcoholic beverages
for 24 h; they were then studied in the morning 15 min
after emptying their bladder and in a temperature-con-
trolled room. The technical error for FFM was <3%,
calculated from three repeated measurements by the
same operator. They were studied in the semirecumbent
position and were allowed to drink only water.
MRI
MRI of eight transaxial segments (field of view,
40–48 cm; 10-mm thickness) at intervertebral disc levels
from T11 to the S1 was performed using a 1.0T Picker
MR scanner (Picker International, Cleveland, OH, USA)
and a T1-weighted fast-spin-echo sequence with a high
#2005 Blackwell Publishing Ltd Diabetes, Obesity and Metabolism, 7, 2005, 406–413 407
M. S. Farvid et al. Association of plasma adiponectin and resistin concentrations with adipose tissue compartments OA

fat–water signal ratio [14]. Subcutaneous abdominal
adipose tissue mass (SAATM), intraperitoneal adipose
tissue mass (IPATM) and retroperitoneal adipose tissue
mass (RPATM) areas were calculated by summing the
relevant adipose tissue pixel units with purpose-
designed software. An in-house program written in
Cþþ was used specifically for the estimation of regional
adipose tissue compartments. Because the signal inten-
sity of adipose tissue in the T1-weighted MRI images
was higher than that of non-adipose tissue, we used a
simple threshold method to separate adipose tissue from
the non-adipose tissue. A threshold value was defined
for each image by analysing the intensity histogram and
choosing the value for the lowest point between inten-
sity peaks (i.e. one corresponding to adipose tissue and
the other to non-adipose tissue). The anatomical seg-
mentations were defined manually using a computer
mouse. The landmark used for separating IPATM and
RPATM in the MRI images was the posterior periton-
eum, which overlies the pancreas and the kidneys. In
our experience, this landmark can be identified confi-
dently down to the level of the pelvis. Fat anterior to the
posterior peritoneum and anterior abdominal wall was
defined as IPATM, and fat posterior to the posterior
peritoneum was defined as RPATM. Corresponding adi-
pose tissue volumes were derived by using the method
of Ross et al. [16], from which SAATM, IPATM and
RPATM were calculated by multiplying the density of
adipose tissue (0.9196 kg/l). Anterior and posterior sub-
cutaneous abdominal compartments were separated by
drawing a perpendicular line at the midpoint of the
anterior–posterior line passing through midpoints of
the vertebral bodies in the MRI images [16]. Anterior sub-
cutaneous abdominal adipose tissue mass (ASAATM)
was obtained by subtracting PSAATM from the total
SAATM (TSAATM). The imaging protocol has a techni-
cal error of <10% and is highly correlated (r
2
¼99%) with
measurements obtained from imaging of the abdominal
region using contiguous transaxial slices; this was con-
firmed using four men included in the present study.
Laboratory Measurements
Fasting plasma cholesterol, triglyceride (TG) and high-
density lipoprotein (HDL) cholesterol were determined
by using standard enzymatic methods. Low-density
lipoprotein (LDL) cholesterol was calculated using the
Friedewald’s equation for TG <4.5 mmol/l. Non-HDL
cholesterol was calculated as the total cholesterol
minus HDL cholesterol. Plasma non-esterified fatty
acids (NEFAs) and glucose were measured using an
enzymatic, colourimetric assay and insulin was meas-
ured using immunosorbent assay. These methods have
been described elsewhere [17,18]. Insulin resistance was
estimated using the homeostasis model assessment
(HOMA) score [19]. Plasma adiponectin was determined
using enzyme immunoassay kit (R&D Systems, Missouri,
MO, USA) (coefficient of variation (CV), <7%). Resistin
was measured in plasma by using an enzyme immunoas-
say kit (Phoenix Pharmaceuticals, Belmont, CA, USA)
(CV, <5%).
Statistical Analyses
SPSS 11.5 (SPSS Inc., Chicago, IL) was used in all ana-
lyses. The data were expressed as mean SD. Skewed
data were log-transformed for analysis. Associations
were examined using Pearson’s univariate and using
stepwise and multiple linear regression methods. Partial
correlations were examined after adjusting for TATM.
Multiple and stepwise regression models were
employed to determine the variables that best predicted
plasma adiponectin concentration. Statistical signifi-
cance was defined at the 5% level.
Results
The anthropometric characteristics of the 41 men with a
wide range of BMI and age have been outlined in
(table 1). Thirteen were non-obese and 28 were obese,
defined as BMI 30 kg/m
2
. The mean proportions of
total adipose tissue as IPATM, RPATM and SAATM
were 11, 1.5 and 12.1%, respectively. TSAATM consists
of 4.2% anterior and 7.8% posterior fat mass.
The biochemical characteristics of the men have been
shown in (table 2). The men had a wide range of insulin
resistance. Four had impaired fasting glucose (i.e. fasting
glucose concentrations between 6.1 and 6.9 mmol/l).
Compared to non-obese men, obese men had significant
elevated plasma concentrations of glucose, insulin and
TG and HOMA score (p <0.01), and significantly lower
plasma levels of HDL cholesterol and adiponectin
(p <0.01), but there were no significant group differ-
ences in plasma resistin concentrations.
Plasma adiponectin concentrations were significantly
and inversely correlated with the masses of all adipose
tissue compartments with the exception of RPATM
(table 3). In stepwise regression, PSAATM was the fat
compartment that best predicted adiponectin concen-
trations (figure 1). Plasma resistin levels were not signifi-
cantly associated with any adipose tissue compartments.
The mass of all adipose tissue compartments, with the
exception of RPATM, was significantly and positively
correlated with plasma insulin levels and insulin
408 Diabetes, Obesity and Metabolism, 7, 2005, 406–413 #2005 Blackwell Publishing Ltd
OA Association of plasma adiponectin and resistin concentrations with adipose tissue compartments M. S. Farvid et al.

resistance, as measured using the HOMA score. Total
intra-abdominal ATM and IPATM were also positively
correlated with plasma glucose concentrations. In step-
wise regression analyses, insulin and HOMA were
strongly associated with IPATM.
The association of adiponectin and resistin with lipid
and biochemical variables has been summarized in table 4.
There was a significant inverse correlation between
adiponectin concentration and plasma cholesterol, TG,
non-HDL cholesterol, insulin and HOMA score, as well
as a significant positive correlation with plasma HDL
cholesterol. No significant association was obtained for
LDL cholesterol or NEFAs. In multiple regression analy-
sis, adiponectin was significantly correlated with HDL
cholesterol after adjusting for SAATM (r ¼0.359,
p¼0.02) or HOMA score (r ¼0.360, p ¼0.03); adiponec-
tin and IPATM were the significant predictors of HOMA
score (r ¼–0.387, p ¼0.004 and r ¼0.441, p ¼0.002) after
adjusting for age and NEFAs; HOMA score and adi-
ponectin were the significant predictors of plasma TG con-
centrations (r ¼0.399, p ¼0.01 and r ¼0.376, p ¼0.01,
respectively) after adjusting for TATM, NEFAs and age.
In multiple regression analysis, elevated HOMA score
was independently associated with increased IPATM
(p ¼0.001) and low adiponectin levels (p ¼0.05);
adjusted r
2
for model ¼0.46. Plasma adiponectin levels
were not significantly correlated with fasting plasma
glucose, free fatty acid, LDL cholesterol, age or resistin
levels. Plasma resistin levels were not significantly asso-
ciated with any lipid and biochemical parameters.
After adjusting for TATM, adiponectin was signifi-
cantly associated with subcutaneous and posterior sub-
cutaneous fat masses only. HOMA score and glucose
were significantly correlated with total intra-abdominal
ATM and IPATM only. Resistin levels were again not
significantly correlated with any of the fat compartments
(table 5).
Table 6 gives multiple regression models, showing
that PSAATM (model 1), as opposed to ASAATM
(model 2), was an independent predictor of adiponectin
levels after including age, total body fat and NEFAs in
the models. Model 3 also shows that TSAATM was a
powerful predictor of adiponectin levels as equal as
PSAATM (model 1). We did not include TSAATM and
PSAATM within the same model, because both these
adipose tissue masses were highly correlated. On the
contrary, this effect was attributed to PSAATM. In step-
wise regression, including adiponectin, age, total body
fat and TSAATM, it was shown that PSAATM was still
the best predictor of adiponectin (p <0.001; adjusted
r
2
¼0.24).
Discussion
A novel finding in the present study was that in men
with a wide range of BMI, PSAATM was the fat compart-
ment that best predicted plasma adiponectin concen-
trations. We also found that low plasma adiponectin was
closely associated with insulin resistance, as measured
by HOMA score, and with dyslipidaemia, as reflected by
hypertriglyceridaemia and low HDL cholesterol.
Another important new finding was that in these men,
plasma resistin levels were not significantly correlated
with total adiposity, type of adiposity, measures of insu-
lin resistance or dyslipidaemia.
Plasma adiponectin levels have been previously
reported to be correlated negatively with BMI and
WHR in human studies [1,2]. We extend these studies
by demonstrating that low adiponectin levels are most
Table 1 Anthropometric and adipose tissue mass character-
istics of the 41 men
Characteristics Mean –SD Range Median
Age (years) 47.0 8.6 25–61 49
Weight (kg) 96.912.4 66.6–117 98.6
Body mass index (kg/m
2
) 30.4 3.3 22.1–34.9 31.6
Waist-to-hip ratio 1.00 0.05 0.87–1.09 1.01
Fat-free mass (kg) 63.2 7.6 40.1–82.7 63.0
IPATM (kg) 3.68 1.50 1.21–8.25 3.26
RPATM (kg) 0.50 0.056 0.11–3.73 0.40
SAATM (kg) 4.03 1.40 1.40–6.85 4.36
PSAATM (kg) 2.62 0.87 0.84–4.36 2.72
ASAATM (kg) 1.410.63 0.19–2.91 1.56
TATM (kg) 33.4 9.9 13.1–56.1 34.3
ASAATM, anterior subcutaneous abdominal adipose tissue mass;
IPATM, intraperitoneal adipose tissue mass; PSAATM, posterior
subcutaneous abdominal adipose tissue mass; RPATM,
retroperitoneal adipose tissue mass; SAATM, subcutaneous
abdominal adipose tissue mass; TATM, total adipose tissue mass.
Table 2 Biochemical characteristics of the men
Characteristics Mean –SD Range Median
Cholesterol (mmol/l) 5.73 0.93 3.8–8.3 5.70
Triglyceride (mmol/l) 2.70 1.93 0.5–8.8 2.10
HDL cholesterol (mmo l/l) 1.01 0.26 0.6–1.8 0.90
LDL cholesterol (mmol/l) 3.59 0.85 1.5–5.8 3.60
Non-HDL cholesterol (mmol/l) 4.73 0.95 3.1–7.4 4.60
Glucose (mmol/l) 5.41 0.60 4.1–6.9 5.40
Insulin (mU/l) 11.68 8.51 2.6–41.8 8.80
HOMA score 2.89 2.26 0.55–10.77 2.06
Adiponectin (ng/ml) 4.43 2.38 1.39–11.41 3.86
Resistin (ng/ml) 20.41 6.92 5.32–36.63 20.5
HOMA, homeostasis model assessment; HDL, high-density
lipoprotein; LDL, low-density lipoprotein.
#2005 Blackwell Publishing Ltd Diabetes, Obesity and Metabolism, 7, 2005, 406–413 409
M. S. Farvid et al. Association of plasma adiponectin and resistin concentrations with adipose tissue compartments OA

closely correlated with accumulation of PSAATM. How-
ever, clinical studies have suggested that plasma adipo-
nectin concentrations and insulin sensitivity are more
closely correlated with intra-abdominal than with sub-
cutaneous fat in men and in women [3,20]. However,
using single-slice computed tomography, they only
assessed these fat compartments and did not measure
PSAATM or IPATM. We also showed in multiple regres-
sion model that IPATM and adiponectin were independ-
ent predictors of insulin resistance, as measured with
the help of HOMA score. In vitro data have shown that
cultured visceral adipocytes express and secrete adipo-
nectin more actively than subcutaneous adipocytes [21],
but the latter were not apparently derived from the pos-
terior subcutaneous abdominal adipose region.
Adiponectin gene expression in subcutaneous abdom-
inal adipose tissue is significantly decreased in obese
non-diabetic men [22]. Subcutaneous abdominal adipo-
cytes are the net exporter of tumour necrosis factor-a
(TNF-a) and interleukin-6 (IL-6) [23,24]. Adiponectin
expression and secretion are inhibited by both TNF-a
and IL-6 [25]. In vivo studies have accordingly shown
that plasma levels of adiponectin and IL-6 are inversely
correlated [25]. Some in vitro studies suggest that
subcutaneous abdominal adipocytes have a higher
basal lipolytic rate and are less susceptible to insulin-
mediating inhibition of lipolysis than visceral adipocytes
[26–28]. Deep adipose tissue layer is metabolically more
active than superficial adipose tissue and displays
greater activities of the metabolic lipogenic enzymes.
Approximately, three-quarters of deep subcutaneous adipose
tissue is found in the posterior compartment [13,29]. This
may explain the strong association of plasma adiponectin
with PSAATM than other fat compartments and
Table 3 Associations (Pearson’s correlation coefficients) between adipose tissue compartments and plasma adiponectin,
resistin and insulin resistance levels
TIAATM IPATM RPATM TSAATM ASAATM PSAATM TATM
Adiponectin 0.307 0.389z0.092 0.500* 0.422y0.502* 0.421y
Resistin 0.022 0.016 0.024 0.218 0.144 0.248 0.176
Insulin 0.553* 0.594* 0.121 0.481* 0.371z0.509* 0.510*
Glucose 0.439y0.447y0.161 0.265 0.217 0.271 0.166
HOMA score 0.579* 0.617* 0.138 0.486* 0.377z0.513* 0.498*
p-value: *<0.001, y<0.01, z<0.05.
ASAATM, anterior subcutaneous abdominal adipose tissue mass; HOMA, homeostasis model assessment; IPATM, intraperitoneal adipose
tissue mass; PSAATM, posterior subcutaneous abdominal adipose tissue mass; RPATM, retroperitoneal adipose tissue mass TATM, total
adipose tissue mass; TIAATM, total intra-abdominal adipose tissue mass; TSAATM, total subcutaneous abdominal adipose tissue mass
PSAATM (k
g
)
4.54.03.53.02.52.01.51.00.5
Adiponectin (ng/ml)
12
10
8
6
4
2
0
r = –0.502
p < 0.001
Fig. 1 Association between plasma
adiponectin concentrations and
posterior subcutaneous abdominal
adipose tissue mass (PSAATM).
410 Diabetes, Obesity and Metabolism, 7, 2005, 406–413 #2005 Blackwell Publishing Ltd
OA Association of plasma adiponectin and resistin concentrations with adipose tissue compartments M. S. Farvid et al.

specifically anterior SAATM. We consider that the asso-
ciations of PSAATM with insulin resistance and adipo-
nectin reflect chiefly both the mass of this fat
compartment and the tightly packed adipocyte architec-
ture [13]. Whether other specific features of adipocytes
in the PSAAT region contribute to insulin resistance
requires further investigation. We should concede, how-
ever, that, in the present study, the greater ability of the
PSAATM compartment, compared to that of TSAATM
in predicting plasma adiponectin levels was only
observed in stepwise regression analysis.
Our study confirms strong relationships between
plasma adiponectin levels, insulin resistance and dys-
lipidaemia reported previously [1–3]. Tschritter et al. [2]
showed that relationship between plasma adiponectin
concentrations and insulin sensitivity was independent
of measures of adiposity, such as BMI, percentage of
body fat or WHR. We demonstrated specifically that
these relationships are even independent of IPATM,
the fat compartment that best predicted insulin resist-
ance, as measured using HOMA score, in the present
study.
We concur with previous reports of an association
between low adiponectin and elevated TGs and low
HDL cholesterol. Experimentally, adiponectin increases
the expression of CD36, acyl-CoA oxidase and uncoup-
ling protein in skeletal muscle, thereby potentially
exerting a TG-regulatory effect on fatty acid oxidation
and TG content. Low plasma adiponectin levels would
be expected to increase fatty acid supply to the liver, as
well as skeletal muscle TG contents [4]. These metabolic
changes, together with hepatic insulin resistance and
increased hepatic lipase activity, could explain the asso-
ciation of low adiponectin with elevated TG, small dense
LDL particle size and low HDL cholesterol [3]. That we
found no significant correlation between plasma adipo-
nectin and insulin resistance with plasma NEFA levels
possibly reflects the high within-person variation of
plasma NEFAs, as well as measurements taken from the
peripheral as opposed to the portal circulation.
The role of resistin in the pathogenesis of insulin
resistance remains questionable, with conflicting data
in animal models and negative findings in clinical obser-
vations [30]. Moon et al. demonstrated that resistin
impaired insulin-stimulated glucose uptake in K6 skel-
etal muscle cells [31], but that this did not indicate an
effect on insulin signalling. However, in an animal
model, Rajala et al. showed that resistin induced hepatic
but not peripheral insulin resistance [32].
McTernan et al. found that resistin was expressed
selectively in abdominal subcutaneous and omental adi-
pose tissue depots [6]. By contrast, others have reported
that resistin is chiefly expressed and secreted by non-
adipose tissue cells, such us mononuclear blood cells
[33,34]. Two cross-sectional studies in human beings
have failed to show an association between resistin and
indices of obesity or insulin resistance [35,36]. Our
results are compatible with these results.
Table 4 Association (Pearson’s correlation coefficients) of
plasma lipids and insulin resistance levels and plasma
adiponectin and resistin concentrations
Adiponectin Resistin
Cholesterol (mmol/l) 0.386* 0.007
Triglyceride (mmol/l) 0.632y0.034
HDL cholesterol (mmol/l) 0.508y0.075
LDL cholesterol (mmol/l) 0.157 0.009
Non-HDL cholesterol (mmol/l) 0.505y0.014
NEFAs (mmol/l) 0.109 0.198
Glucose (mmol/l) 0.192 0.127
Insulin (mU/l) 0.552y0.191
HOMA score 0.540y0.193
p-value: *<0.05, y<0.001.
HOMA, homeostasis model assessment; HDL, high-density
lipoprotein; LDL, low-density lipoprotein; NEFAs, non-esterified
fatty acids.
Table 5 Partial correlation coefficients adjusted for total adipose tissue mass (TATM) for associations between adipose tissue
compartments and adiponectin, resistin and insulin resistance levels
TIAATM IPATM RPATM TSAATM ASAATM PSAATM
Adiponectin 0.016 0.115 0.256 0.340y0.241 0.345y
Resistin 0.146 0.153 0.031 0.131 0.037 0.189
Glucose 0.460* 0.471* 0.118 0.229 0.145 0.249
Insulin 0.318y0.385y0.039 0.128 0.052 0.170
HOMA score 0.370y0.433* 0.014 0.157 0.073 0.198
p-value: y<0.05, *<0.01.
ASAATM, anterior subcutaneous abdominal adipose tissue mass; HOMA, homeostasis model assessment; IPATM, intraperitoneal adipose
tissue mass; PSAATM, posterior subcutaneous abdominal adipose tissue mass; RPATM, retroperitoneal adipose tissue mass TATM, total
adipose tissue mass; TIAATM, total intra-abdominal adipose tissue mass; TSAATM, total subcutaneous abdominal adipose tissue mass.
#2005 Blackwell Publishing Ltd Diabetes, Obesity and Metabolism, 7, 2005, 406–413 411
M. S. Farvid et al. Association of plasma adiponectin and resistin concentrations with adipose tissue compartments OA

Although our study was based on correlational
analyses, it, nonetheless, highlights the importance of
body fat distribution on adipocytokine concentration.
Our results also indicate the importance of adiponectin
as a determinant of insulin resistance and dyslipidaemia.
This finding is consistent with previous results,
demonstrating the role of adiponectin as a risk factor
for cardiovascular disease [1–3,37]. Our present findings
require confirmation in women and other racial groups.
If adiponectin turns out to be a critical regulator of
energy metabolism, further studies should also show
the benefit of recombinant DNA therapy in controlling
body weight and certain aspects of the metabolic
syndrome in overweight/obese persons [37]. Our find-
ings also question whether resistin plays a role in the
pathogenesis of insulin resistance and the metabolic
syndrome.
Acknowledgements
Our work on lipoprotein metabolism is supported by
research grant (G03P1171) from the National Heart
Foundation of Australia, the National Health and Medical
Research Council, the Raine Foundation, Pfizer Inc and
Glaxo-Wellcome. PHRB is a Career Development Fellow
of the National Heart Foundation and was also sup-
ported by the NIH (NIH/NIBIB P41 EB-001975). DCC
was supported by a post-doctoral fellowship from the
Raine/National Heart Foundation of Australia. MSF was
supported by Iranian Ministry of Health.
References
1 Shand BI, Scott RS, Elder PA, George PM. Plasma adipo-
nectin in overweight, nondiabetic individuals with or
without insulin resistance. Diabetes Obes Metab 2003;
5: 349–353.
2 Tschritter O, Fritsche A, Thamer C et al. Plasma adipo-
nectin concentrations predict insulin sensitivity of both
glucose and lipid metabolism. Diabetes 2003; 52: 239–243.
3 Cnop M, Havel PJ, Utzschneider KM et al. Relationship
of adiponectin to body fat distribution, insulin sensitiv-
ity and plasma lipoproteins: evidence for independent
roles of age and sex. Diabetologia 2003; 46: 459–469.
4 Kadowaki T, Hara K, Yamauchi T, Terauchi Y, Tobe K,
Nagai R. Molecular mechanism of insulin resistance and
obesity. Exp Biol Med 2003; 228: 1111–1117.
5 Steppan CM, Brown EJ, Wright CM et al. A family of
tissue-specific resistin-like molecules. Proc Natl Acad
Sci USA 2001; 98: 502–506.
6 McTernan PG, McTernan CL, Chetty R et al. Increased
resistin gene and protein expression in human abdom-
inal adipose tissue. J Clin Endocrinol Metab 2002; 87:
2407.
7 Steppan CM, Bailey ST, Bhat S et al. The hormone resistin
links obesity to diabetes. Nature 2001; 409: 307–312.
8 Silha JV, Krsek M, Skrha JV, Sucharda P, Nyomba BL,
Murphy LJ. Plasma resistin, adiponectin and leptin
levels in lean and obese subjects: correlations with
insulin resistance. Eur J Endocrinol 2003; 149: 331–335.
9 Azuma K, Katsukawa F, Oguchi S et al. Correlation
between serum resistin level and adiposity in obese
individuals. Obes Res 2003; 11: 997–1001.
10 Bjorntorp P. Abdominal fat distribution and disease: an
overview of epidemiological data. Ann Med 1992; 24:
15–18.
Table 6 Multiple linear regression models showing association between adiponectin concentrations and posterior subcuta-
neous fat mass, including total body fat, non-esterified fatty acids (NEFAs), subcutaneous fat mass and age
Predictor variables Regression coefficient SEM p-value
Model 1
Posterior subcutaneous ATM (kg) 0.587 0.07 0.04
Total body fat (kg) 0.094 0.006 0.74
NEFAs (mmol/l) 0.007 0.253 0.96
Age (years) 0.043 0.004 0.77
Model 2
Anterior subcutaneous ATM (kg) 0.289 0.07 0.151
Total body fat (kg) 0.226 0.005 0.270
NEFAs (mmol/l) 0.051 0.257 0.741
Age (years) 0.059 0.004 0.69
Model 3
Subcutaneous ATM (kg) 0.532 0.04 0.042
Total body fat (kg) 0.027 0.006 0.918
NEFAs (mmol/l) 0.018 0.251 0.907
Age (years) 0.045 0.004 0.756
ATM, adipose tissue mass; NEFAs, non-esterified fatty acids.
412 Diabetes, Obesity and Metabolism, 7, 2005, 406–413 #2005 Blackwell Publishing Ltd
OA Association of plasma adiponectin and resistin concentrations with adipose tissue compartments M. S. Farvid et al.

11 Chan DC, Watts GF, Sussekov AV et al. Adipose tissue
compartments and insulin resistance in overweight-obese
Caucasian men. Diabetes Res Clin Pract 2004; 63: 77–85.
12 Ross R, Aru J, Freeman J, Hudson R, Janssen I. Abdom-
inal adiposity and insulin resistance in obese men. Am J
Physiol Endocrinol Metab 2002; 282: E657–E663.
13 Misra A, Garg A, Abate N, Peshock RM, Stray-Gundersen J,
Grundy SM. Relationship of anterior and posterior sub-
cutaneous abdominal fat to insulin sensitivity in nondia-
betic men. Obes Res 1997; 5: 93–99.
14 Abate N, Burns D, Peshock RM, Garg A, Grundy SM.
Estimation of adipose tissue mass by magnetic resonance
imaging: validation against dissection in human cada-
vers. J Lipid Res 1994; 35: 1490–1496.
15 Lukaski HC, Johnson PE, Bolonchuk WW, Lykken GI.
Assessment of fat-free mass using bioelectrical impe-
dance measurements of the human body. Am J Clin Nutr
1985; 41: 810–817.
16 Ross R, Leger L, Morris D, de Guise J, Guardo R. Quantifi-
cation of adipose tissue by MRI: relationship with anthro-
pometric variables. J Appl Physiol 1992; 72: 787–795.
17 Riches FM, Watts GF, Naoumova RP, Kelly JM, Croft KD,
Thompson GR. Hepatic secretion of very-low-density
lipoprotein apolipoprotein B-100 studied with a stable
isotope technique in men with visceral obesity. Int J
Obes Relat Metab Disord 1998; 22: 414–423.
18 Riches FM, Watts GF, Hua J, Stewart GR, Naoumova RP,
Barrett PH. Reduction in visceral adipose tissue is asso-
ciated with improvement in apolipoprotein B-100 meta-
bolism in obese men. J Clin Endocrinol Metab 1999; 84:
2854–2861.
19 Matthews DR, Hosker JP, Rudenski AS, Naylor BA,
Treacher DF, Turner RC. Homeostasis model assessment:
insulin resistance and beta-cell function from fasting
plasma glucose and insulin concentrations in man. Dia-
betologia 1985; 28: 412–419.
20 Park KG, Park KS, Kim MJ et al. Relationship between
serum adiponectin and leptin concentrations and body
fat distribution. Diabetes Res Clin Pract 2004; 63: 135–142.
21 Motoshima H, Wu X, Sinha MK et al. Differential regul-
ation of adiponectin secretion from cultured human
omental and subcutaneous adipocytes: effects of insulin
and rosiglitazone. J Clin Endocrinol Metab 2002; 87:
5662–5667.
22 Engeli S, Feldpausch M, Gorzelniak K et al. Association
between adiponectin and mediators of inflammation in
obese women. Diabetes 2003; 52: 942–947.
23 Kern PA, Saghizadeh M, Ong JM, Bosch RJ, Deem R,
Simsolo RB. The expression of tumor necrosis factor in
human adipose tissue. Regulation by obesity, weight
loss, and relationship to lipoprotein lipase. J Clin Invest
1995; 95: 2111–2119.
24 Mohamed-Ali V, Flower L, Sethi J et al. Beta-adrenergic
regulation of IL-6 release from adipose tissue: in vivo
and in vitro studies. J Clin Endocrinol Metab 2001; 86:
5864–5869.
25 Bruun JM, Lihn AS, Verdich C et al. Regulation of adi-
ponectin by adipose tissue-derived cytokines: in vivo
and in vitro investigations in humans. Am J Physiol
Endocrinol Metab 2003; 285: E527–E533.
26 Bolinder J, Kager L, Ostman J, Arner P. Differences at the
receptor and postreceptor levels between human omen-
tal and subcutaneous adipose tissue in the action of
insulin on lipolysis. Diabetes 1983; 32: 117–123.
27 Smith U, Hammersten J, Bjorntorp P, Kral JG. Regional
differences and effect of weight reduction on human fat
cell metabolism. Eur J Clin Invest 1979; 9: 327–332.
28 Ostman J, Arner P, Engfeldt P, Kager L. Regional differ-
ences in the control of lipolysis in human adipose tissue.
Metabolism 1979; 28: 1198–1205.
29 Kelley DE, Thaete FL, Troost F, Huwe T, Goodpaster BH.
Subdivisions of subcutaneous abdominal adipose tissue
and insulin resistance. Am J Physiol Endocrinol Metab
2000; 278: E941–E948.
30 Rea R, Donnelly R. Resistin: an adipocyte-derived hor-
mone. Has it a role in diabetes and obesity? Diabetes
Obes Metab 2004; 6: 163–170.
31 Moon B, Kwan JJ, Duddy N, Sweeney G, Begum N.
Resistin inhibits glucose uptake in L6 cells indepen-
dently of changes in insulin signaling and GLUT4 trans-
location. Am J Physiol Endocrinol Metab 2003; 285:
E106–E115.
32 Rajala MW, Obici S, Scherer PE, Rossetti L. Adipose-
derived resistin and gut-derived resistin-like molecule-beta
selectively impair insulin action on glucose production.
J Clin Invest 2003; 111: 225–230.
33 Fain JN, Cheema PS, Bahouth SW, Lloyd Hiler M. Resis-
tin release by human adipose tissue explants in primary
culture. Biochem Biophys Res Commun 2003; 300:
674–678.
34 Nagaev I, Smith U. Insulin resistance and type 2 diabetes
are not related to resistin expression in human fat cells
or skeletal muscle. Biochem Biophys Res Commun 2001;
285: 561–564.
35 Lee JH, Chan JL, Yiannakouris N et al. Circulating resis-
tin levels are not associated with obesity or insulin
resistance in humans and are not regulated by fasting
or leptin administration: cross-sectional and interven-
tional studies in normal, insulin-resistant, and diabetic
subjects. J Clin Endocrinol Metab 2003; 88: 4848–4856.
36 Youn BS, Yu KY, Park HJ et al. Plasma resistin concen-
trations measured by enzyme-linked immunosorbent
assay using a newly developed monoclonal antibody
are elevated in individuals with type 2 diabetes mellitus.
J Clin Endocrinol Metab 2004; 89: 150–156.
37 Ukkola O, Santaniemi M. Adiponectin: a link between
excess adiposity and associated comorbidities? J Mol
Med 2002; 80: 696–702.
#2005 Blackwell Publishing Ltd Diabetes, Obesity and Metabolism, 7, 2005, 406–413 413
M. S. Farvid et al. Association of plasma adiponectin and resistin concentrations with adipose tissue compartments OA