diabetes research and clinical practice 111 (2016) 5865

Contents available at ScienceDirect

Diabetes Research
and Clinical Practice
jou rnal hom ep ag e: w ww.e l s e v i er . c om/ loca te / d i ab r es

The effects of adiponectin and inflammatory

cytokines on diabetic vascular complications
in obese and non-obese patients with type 2
diabetes mellitus
Seong Bin Hong a,1, Jung Jin Lee a,b,1, So Hun Kim a, Young Ju Suh c,
Ju Young Han a, Yong Seong Kim a, Moonsuk Nam a,*
a

Department of Endocrinology, Inha University School of Medicine, Incheon, Republic of Korea

Department of Internal Medicine, Hallym General Hospital, Incheon, Republic of Korea
c
Department of Biomedical Sciences, Inha University School of Medicine, Incheon, Republic of Korea
b

article info

abstract

Article history:

Aims: To evaluate the associations between inflammatory cytokines and adiponectin and

Received 16 April 2015

various vascular complications in type 2 diabetes mellitus (T2DM).

Received in revised form

Methods: A total of 761 patients with T2DM were divided into a non-obese group and an

4 October 2015

obese group to enable the effects of obesity and T2DM on vascular complications to be

Accepted 11 October 2015

differentiated. The serum levels of circulating inflammatory cytokines, that is, tumor

Available online 21 October 2015

necrosis factor (TNF)-a, and interleukin (IL)-6, total adiponectin, and high molecular weight
(HMW) adiponectin were measured, and carotid intima media thickness (IMT), the presence

Keywords:

of carotid plaque, and the severities of retinopathy and nephropathy, were assessed.

Type 2 diabetes mellitus, Adipokines

Results: The obese group had significantly lower serum total and HMW adiponectin levels

Vascular complication

than the non-obese group. In the obese group, serum levels of total and HMW adiponectin,
and TNF-a were significantly higher in patients with proliferative retinopathy than in those
without retinopathy after adjusting for covariates. In the non-obese group, only IL-6 levels
were significantly higher in patients with proliferative retinopathy than in those without.
Serum levels of total and HMW adiponectin were significantly higher in patients with
macroalbuminuria than in those with normoalbuminuria. No significant difference of three
cytokines levels were observed depending on the carotid IMT or the presence of plaque.
Logistic regression analysis revealed that serum total adiponectin (OR = 1.209, P = 0.038),
diabetes duration (OR = 1.230, P = 0.014), and HbA1c (OR = 2.359, P = 0.006) were significantly
associated with proliferative retinopathy in the obese group.
Conclusion: The study shows total adiponectin may influence proliferative retinopathy in
obese patient with T2DM.
# 2015 Elsevier Ireland Ltd. All rights reserved.

* Corresponding author at: Department of Endocrinology, Inha University School of Medicine, 7-206 3rd St. Shinheung-Dong, Jung-gu,
Incheon 400-711, Republic of Korea. Tel.: +82 32 890 1101; fax: +82 32 890 1120.
E-mail address: namms@inha.ac.kr (M. Nam).
1
Seong Bin Hong and Jung Jin Lee contributed equally to this work.
http://dx.doi.org/10.1016/j.diabres.2015.10.017
0168-8227/# 2015 Elsevier Ireland Ltd. All rights reserved.

diabetes research and clinical practice 111 (2016) 5865

1.

Introduction

Diabetes mellitus and obesity are the major causes of various

vascular complications, and chronic low grade systemic
inflammation may constitute a pivotal link between obesity
and diabetes. The majority of studies on the development
and progression of diabetic vascular complications have been
performed on obese patients with type 2 diabetes mellitus
(T2DM), which makes it difficult to determine how obesity and
T2DM individually affect vascular complications. Adiponectin
is an adipose specific protein with antiatherogenic properties
that is abundantly present in the circulation [1]. In general,
serum adiponectin levels have been reported to be low in
obese subjects and in individuals with T2DM or cardiovascular
disease [13]. However, in one study, it was concluded that
elevated adiponectin levels did not have a significant protective effect on the development of coronary heart disease [4].
Many studies have addressed the association between
vascular complications and serum adiponectin. Of the studies
that investigated the effects of adiponectin on diabetic
microvascular complications, one found adiponectin levels
in diabetics with proliferative retinopathy were significantly
lower than in patients without [5]. However, in another study,
serum adiponectin levels were found to be elevated in patients
with progressed retinopathy and nephropathy [6]. It is possible
that race and demographic factors explain these inconsistencies.
Although, obesity alone or in combination with blood
pressure and other clinical factors can be risk factor of
vascular complications, few comparative studies have examined the effects of adipokines on diabetic vascular complications in obese and non-obese type 2 diabetes patients. In the
present study, we compared the effects of inflammatory
cytokine and adipokine on vascular complications, namely,
retinopathy, nephropathy, carotid intima media thickness
and the presence of carotid artery plaque in obese and nonobese patients with T2DM.

2.

Materials and methods

2.1.

Study population

This study was conducted using a cross-sectional, observational design. The patient population consisted of 761 subjects
with type 2 diabetes recruited from the outpatient clinic at the
Diabetes Center of Inha University Hospital from March 2007
to December 2009. Inclusion criteria for the patients with
T2DM included; older than 20 year of age, newly diagnosed
with T2DM using American Diabetes Association criteria
(2003) [7] and/or was being treated with oral hypoglycemic
agents or insulin, or lifestyle modification for known T2DM. All
patients were Korean. Patients were allocated to an obese
or non-obese groups using the classification used in the
redefined Asian-Pacific Region guideline [8]; non-obese group
(body mass index (BMI) <25 kg/m2) and obese group (BMI
25 kg/m2). The exclusion criteria applied were as follows:
congestive heart failure, severe infection, uncontrolled hypertension, severe dyslipidemia (total cholesterol >400 mg/d), or

59

a medical condition requiring active management. The study

protocol was approved by the institutional review board of
Inha University Hospital and all participants provided written
informed consent prior to participation.

2.2.

Measurements

Evaluations comprised detailed medical history taking, anthropometric measurements, physical examination, laboratory measurements, and diabetic vascular complication
studies.

2.2.1.

Anthropometric and laboratory measurements

Height and weight were measured to the nearest 0.1 cm and

0.1 kg, respectively. Body mass index (BMI) was defined as
weight (kg) divided by height (m) squared. Waist circumference was measured at the midpoint between the lower
borders of the rib cage and the iliac crest. Blood pressure was
measured after the subject had rested for at least 10 min in a
sitting position. Blood samples were collected after an
overnight fast of at least 10 h and stored at 70 8C for
subsequent assay. Laboratory measurements including fasting serum glucose, lipid subfractions, HbA1c, 75 g oral glucose
tolerance testing (not performed on all patients) were
conducted in a fasted state. Low density lipoprotein (LDL)
cholesterol levels were calculated using the Friedewald
formula [9]. Hs-CRP level was measured using a particle
enhanced immunoturbidimetric assay (Hitachi High-Technologies Corp., Tokyo, Japan). Estimated glomerular filtration rate
(GFR) values were calculated using the Modified of diet in Renal
Disease Study (MDRD) equation [10]. HOMA-IR (homeostasis
model assessment of insulin resistance) levels were calculated
by dividing the product of insulin (microunits per milliliter)
and glucose (millimoles per liter) concentrations by 22.5 [11].

2.2.2.

Carotid B-mode ultrasound measurements

Carotid intima media wall thickness (cIMT) and plaques

number were measured in right and left common carotid
arteries (CCA) using a GE. LOGIQ 7 ultrasonograph (GE
Healthcare, Wauwatosa, WI) equipped with a 47 MHz linear
transducer. A sonographer unaware of subjects characteristics scanned both common carotid arteries (CCAs), the
carotid bulb, and the proximal portions of the internal and
external carotid arteries in two planes (anterior oblique and
lateral), and then focused on the interfaces required to
measure IMT and on any areas of focal plaque. Plaque was
defined as either a distinct thickening of >1.5 mm into the
vessel lumen or as definite echogenicity with a posterior
echogenic shadow. IMT was measured on the far walls of
distal CCAs at three IMT points, that is, at the site of greatest
thickness and at points 1 cm upstream and downstream this
site. Computer-assisted acquisition, processing, storage of Bmode images, and cIMT calculations were performed using
Intima Scope software (MediaCross, Tokyo, Japan).

2.2.3.

Measurement of cytokine levels

Serum total adiponectin levels were measured using a

commercially available ELISA kit (ALPCO Diagnostics, Salem,
NH, USA). Intra- and interassay coefficients of variation (CVs)
for this kit were 5.3 and 5.0%, respectively. High molecular

60

diabetes research and clinical practice 111 (2016) 5865

weight (HMW) adiponectin serum levels were measured using

an ELISA kit (ALPCO Diagnostics, Salem, NH, USA). Intra- and
interassay CVs were 3.3 and 5.7%, respectively. Serum tumor
necrosis factor-a (TNF-a) levels were measured using an
enzyme immunometric assay kit (R&D Systems, Minneapolis,
MN), and intra- and interassay CVs were 3.18.5% and 7.4
10.6%, respectively. Serum interleukin-6 (IL-6) levels were
measured using an ELISA kit (R&D Systems), which had intraand interassay CVs of 6.97.4% and 6.59.6%, respectively.

2.2.4.

Definitions of diabetic retinopathy and nephropathy

Diabetic retinopathy was diagnosed and was classified into

five stages by ophthalmologists according to the international
classification of diabetic retinopathy, that is, no apparent
retinopathy, mild non-proliferative retinopathy, moderate
non-proliferative retinopathy, severe non-proliferative retinopathy, or proliferative retinopathy [12]. On the other hand,
diabetic nephropathy was classified using three stages: normoalbuminuria (spot urine albumin/creatinine ratio (UACR)
<30 mcg/mg Cr), microalbuminuria (UACR 30 and <299
mcg/mg Cr), and macroalbuminuria (UACR 300 mcg/mg Cr).

2.3.

Statistical analysis

The x2 test or Fishers exact test were used to compare

categorical variables between two groups. Because all continuous variables were non-normally distributed, the Mann
Whitney U test was used to compare pairs of groups.
Spearmans partial correlation test was used to compare
cytokine levels with respect to vascular complications after
adjusting for covariates. Multiple logistic regression analysis
was used to assess the independency and strengths of
associations between clinical factors at each stage of

retinopathy and nephropathy. This analysis was performed

using retinopathy and nephropathy stages as dependent
variables, and cytokine level, age, gender, diabetes duration,
HbA1c, GFR (in cases of nephropathy) as independent
variables. Results were analyzed using SPSS ver.19.0 (SPSS
Inc., Chicago, IL, USA). Bonferronis correction were additionally applied to Spearmans partial correlation test to access
relationships between cytokine levels and vascular complications, and to subgroup analysis with respect to obesity to
evaluate association between clinical factors and retinopathy
or nephropathy. Statistical significance was accepted for P
values <0.05.

3.

Results

The study population comprised 468 men and 293 women with
type 2 diabetes. Study subjects were divided into two groups:
335 non-obese (BMI <25 kg/m2) patients and 426 obese (BMI
25 kg/m2) patients. The baseline clinical characteristics of
the two groups are shown in Table 1. Systolic and diastolic
blood pressures were higher in the obese group (P = 0.017 and
0.002, respectively). Triglyceride, HOMA-IR, and high sensitivity C reactive protein (hs-CRP) were higher (P < 0.0001,
P < 0.0001 and P = 0.001, respectively), and high density
lipoprotein (HDL) cholesterol and total and HMW adiponectin
levels were lower in the obese group (P = 0.01 and P < 0.0001,
respectively) (Table 1).
Mean and maximum cIMT, and the prevalence of diabetic
nephropathy were similar in the obese and non-obese groups.
More non-obese patients had retinopathy (30.2 vs. 21.2%,
P = 0.044), and more obese patients had carotid plaque (49.7 vs.
57.1%, P = 0.054) (Table 2).

Table 1 Baseline clinical and laboratory characteristics.

N
Age (years)
Gender (M/F)
BMI (kg/m2)
Waist to hip ratio
Diabetes duration (years)
Systolic BP (mmHg)
Diastolic BP (mmHg)
Fasting glucose (mg/dl)
HbA1c (%)
Total cholesterol (mg/dl)
Triglyceride (mg/dl)
HDL (mg/dl)
LDL (mg/dl)
hsCRP (mg/L)
HOMA-IR
Baseline levels of adipokines (N)
TNF-a (pg/ml)
IL-6 (pg/ml)
Total adiponectin (ng/ml)
HMW adiponectin (ng/ml)

Total

Non-obese

Obese

761
52 (4760)
468/293
25.4 (23.527.4)
0.91 (0.890.94)
4(19)
125 (114135)
77 (7085)
139 (117168)
7.4 (6.88.6)
179 (156204)
142 (96201)
45 (3952)
117 (95144)
0.1 (0.050.21)
2.7 (1.674.15)
750
1.29 (0.842.07)
0.95 (0.581.63)
3.94 (2.786.12)
1.29 (0.612.39)

335
52 (4660)
207/128
23.2 (22.024.1)
0.90 (0.870.92)
5 (111)
122 (113133)
76 (6982)
139 (116178)
7.6 (6.89.0)
177 (155202)
126 (18380)
46 (4054)
115 (94143)
0.08 (0.040.19)
2.11 (1.253.42)
328
1.37 (0.872.09)
0.92 (0.561.47)
4.21 (3.186.84)
1.52 (0.832.85)

426
53 (4861)
261/165
27.1 (25.928.8)
0.92 (0.900.96)
4 (18)
126 (116136)
79 (7286)
138 (118163)
7.35 (6.78.4)
180 (157208)
153 (109215)
44 (3851)
118 (96146)
0.11 (0.060.23)
3.13 (2.114.78)
422
1.25 (0.832.06)
0.96 (0.591.70)
3.64 (2.585.46)
1.07 (0.511.97)

P valuea
0.247
0.883
<0.0001
<0.0001
0.002
0.017
0.002
0.65
0.22
0.22
<0.0001
0.01
0.15
0.001
<0.0001
0.778
0.068
<0.0001
<0.0001

Results are medians (interquartile ranges (IQRs)).

P values were obtained using the x2 test for categorical variables and the MannWhitney U test for continuous variables and represent the
significances of differences between obese and non-obese patients.
a

61

diabetes research and clinical practice 111 (2016) 5865

Table 2 Comparison of vascular complications.

Nephropathy
Normoalbuinuria
Microalbuminuria
Macroalbuminuria

Retinopathy
No apparent
Retinopathy (+)
Mild NPDR
Moderate NPDR
Severe NPDR
PDR

cIMT and plaque

Plaque (+)
Plaque ( )
Mean IMT (mm)
Maximal IMT (mm)

Total patients

Non-obese patients

Obese patients

n = 726

n = 315

n = 411

418 (57.7%)
274 (37.8%)
33 (4.6%)

186 (59.0%)
116 (36.8%)
13 (4.1%)

232 (56.4%)
158 (38.4%)
20 (4.9%)

Total patients

Non-obese patients

Obese patients

n = 634

n = 285

n = 348

474 (74.8%)
160 (34%)
86 (13.6%)
23 (3.6%)
30 (4.7%)
21(3.3%)

199
86
43
16
15
12

(69.8%)
(30.2%)
(15.1%)
(5.6%)
(5.3%)
(4.2%)

274
74
43
7
15
9

(78.8%)
(21.2%)
(12.3%)
(2.0%)
(4.3%)
(2.6%)

Total patients

Non-obese patients

Obese patients

n = 678

n = 298

n = 380

365 (53.8%)
313 (46.2%)
0.68 (0.610.79)
0.80 (0.710.93)

148 (49.7%)
150 (50.3%)
0.68 (0.600.78)
0.79 (0.690.91)

217 (57.1%)
163 (42.9%)
0.67 (0.610.78)
0.79 (0.710.93)

P valuea

0.867

P valuea

0.044

P valuea

0.054
0.54
0.48

Results are numbers or patients (percentages).

P values represent the significances of differences between obese and non-obese patients.

Cytokine levels were compared at different stages of

diabetic retinopathy after adjusting for covariates (age,
gender, diabetes duration, HbA1c, history of hypertension
and medications (angiotensin converting enzyme inhibitor
(ACEi) or angiotensin receptor blocker (ARB) and thiazolidinedione (TZD))). As compared with patients with no apparent
retinopathy, study subjects with proliferative diabetic retinopathy had significantly higher total and HMW adiponectin,
IL-6, and TNF-a levels (P = 0.004, 0.032, 0.004 and 0.004,
respectively). When the same analysis was done in the obese
and non-obese groups, total and HMW adiponectin and TNF-a
levels were significantly higher in obese patients and IL-6 was
significantly higher in non-obese patients (Table 3).
In study subjects with diabetic nephropathy, those with
macroalbuminuria had higher total and HMW adiponectin
levels than those with normoalbuminuria. When this analysis
was performed on the obese group, total and HMW adiponectin levels in patients with macroalbuminuria were higher
than patient with normoalbuminuria (P = 0.006 and 0.001
respectively). However, in the non-obese group, these relations were non-significant (P = 0.111 and 0.115, respectively)
(Table 4).
To examine the effects of cytokines on proliferative
retinopathy, subgroup analysis was performed to compare
their levels in patients with apparent retinopathy and
proliferative retinopathy (Table 5). Multivariate logistic regression analysis on all study subjects revealed that diabetes
duration and HbA1c significantly influenced proliferative
retinopathy. On the other hand, in the obese group, total
adiponectin, HbA1c, and duration of diabetes were found to
be significantly related to the presence of proliferative
retinopathy. However, these relations were not significant
in the non-obese group. No single cytokine was found to

influence diabetic nephropathy after adjusting for age, gender,

diabetes duration, and renal function in either the obese or
non-obese group. Furthermore, age, diabetes duration were
related to macroalbuminuria in all study subjects and each
group (Table S1). In obese and on-obese group, no significant
relation was found between cytokine levels and the mean IMT
(Table S2).

4.

Discussion

Adiponectin is considered a beneficial adipokine, and its

levels have been shown to be negatively correlated with
insulin resistance, dyslipidemia, and inflammatory markers
[13]. Adiponectin is multifunctional and known to be involved
in energy metabolism, atherosclerosis, inflammation, and
cell proliferation [14]. The findings of the present study
indicate that adiponectin levels are higher in obese and
non-obese patients with proliferative retinopathy than in
those without apparent retinopathy. Lifestyle modifications,
weight reductions and several medical therapies have been
reported to increase adiponectin levels in obese patients
[15,16], and treatment with thiazolidinedione, ACEIs, or ARBs
increases adiponectin levels to a certain extent [18]. In the
present study, we adjusted for these medications, and thus,
excluded the effects of drugs on adiponectin levels during
the analysis.
Our findings are in agreement with those of a previous
study [6], but disagree with the reported observation that
patients with proliferative retinopathy have lower adiponectin
levels than patients without [5]. Clinical studies about
relationships between adiponectin levels and diabetic retinopathy have been inconclusive. In an experimental study of

62

diabetes research and clinical practice 111 (2016) 5865

Table 3 Comparison of cytokines levels at different stages of diabetic retinopathy in obese and non-obese patients.
TNF-a (pg/ml)

Non-obese patients
No apparent
n = 199
Mild NPDR
N = 43
Moderate NPDR
n = 16
Severe NPDR
n = 15
PDR
n = 12
Obese patients
No apparent
n = 274
Mild NPDR
N = 43
Moderate NPDR
n=7
Severe NPDR
n = 15
PDR
n=9

Pa

IL-6 (pg/ml)

1.28 (0.851.93)

Pa

Total
adiponectin
(ng/ml)

0.88 (0.541.42)

Pa

4.16 (3.116.36)

HMW
adiponectin
(ng/ml)

Pa

1.48 (0.692.73)

1.42 (0.752.14)

1.000

0.77 (0.491.40)

1.000

4.69 (3.417.06)

1.000

1.60 (0.983.07)

1.000

2.24 (1.072.88)

0.652

1.17 (0.781.50)

1.000

4.60 (2.956.57)

1.000

1.54 (0.722.21)

1.000

1.42 (1.032.33)

1.000

1.12 (0.782.57)

0.508

3.97 (3.096.12)

1.000

1.36 (1.033.45)

1.000

2.41 (1.073.00)

0.208

1.66 (0.992.69)

0.02

4.97 (3.777.86)

1.000

1.89 (1.234.88)

1.000

1.21 (0.822.00)

0.93 (0.581.63)

3.47 (2.424.86)

1.03 (0.451.82)

1.29 (0.892.07)

1.000

0.99 (0.621.54)

1.000

4.27 (2.848.49)

0.124

1.20 (0.592.40)

0.756

1.07 (0.852.29)

1.000

1.23 (0.961.85)

0.840

3.61 (1.8412.52)

1.000

1.61 (0.487.22)

1.000

1.25 (0.862.67)

1.000

1.86 (1.022.36)

0.148

3.65 (2.924.64)

0.996

1.04 (0.821.77)

0.428

2.27 (1.728.64)

0.012

1.58 (1.042.34)

0.196

8.98 (6.6711.35)

<0.001

2.52 (1.605.75)

0.001

Results are medians (interquartile ranges).

Covariates: age, sex, diabetes duration, HbA1c, presence of hypertension, medications (ACEi or ARB, TZD), creatinine, GFR.
a
P values were calculated with respect to patients without apparent retinopathy.

ischemic retina, adiponectin knockout mice showed marked

increases in retinal TNF-a mRNA levels vs. wild type mice, and
adiponectin administration suppressed pathological neovascularization, which suggests adiponectin has a protective
effect on retinopathy. Furthermore, the protective effect of
adiponectin was found to be reduced after TNF-a ablation,
indicating TNF-a suppression plays a functionally significant
role in the retinal protection afforded by adiponectin [19]. In

recent study, Costagliola et al. demonstrated that adiponectin

and vascular endothelial growth factor level in aqueous
humor obtained from patients with T2DM, proliferative
retinopathy and macular edema is higher than control [20].
The mechanism how adiponectin effects on proliferative
retinopathy might be explained by a counter-regulatory
response to retinal vessel formation and inflammation. In a
previous study, adiponectin increases in aqueous humor and

Table 4 Comparison of cytokines levels at different stages of diabetic nephropathy in obese and non-obese patients.
TNF-a (pg/ml)

Non-obese patients
Normoalbuminuria
n = 186
Microalbuminuria
n = 116
Macroalbuminuria
n = 13
Obese patients
Normoalbuminuria
n = 232
Microalbuminuria
n = 158
Macroalbuminuria
n = 21

Pa

1.22 (0.871.85)

IL-6 (pg/ml)

Pa

0.87 (0.541.42)

Total
adiponectin
(ng/ml)

Pa

4.15 (3.176.80)

HMW
adiponectin
(ng/ml)

Pa

1.52 (0.812.75)

1.44 (0.862.33)

1.000

0.95 (0.581.63)

1.000

4.20 (3.096.23)

1.000

1.44 (0.732.79)

0.576

2.33 (0.953.63)

0.053

1.27 (0.895.52)

0.120

5.90 (4.429.41)

0.111

2.64 (1.435.36)

0.115

1.15 (0.821.96)

0.94 (0.611.81)

3.61 (2.545.29)

1.07 (0.521.90)

1.37 (0.872.29)

0.176

1.02 (0.581.53)

1.000

3.69 (2.585.73)

1.000

1.05 (0.511.96)

1.000

1.48 (0.892.34)

1.000

1.19 (0.622.16)

1.000

3.95 (3.047.69)

0.006

1.42 (0.842.43)

0.001

Results are medians (interquartile ranges).

Covariates: age, sex, diabetes duration, HbA1c, presence of hypertension, medications (ACE inhibitor, ARB or TZD), creatinine, GFR.
a
P values were obtained using normoalbuminuria as referent.

63

diabetes research and clinical practice 111 (2016) 5865

Table 5 Factors significantly-influencing proliferative diabetic retinopathy in the study group.

Total patients

Age (years)
Gender (M/F)
Diabetes duration (years)
HbA1c (%)
Total adiponectin (ng/ml)
TNF-a (pg/ml)
IL-6 (pg/ml)

Obese

Non-obese

OR

P value

OR

P value

OR

P value

0.927
0.689
1.182
1.468
1.126
1.035
1.051

0.024
1.000
<0.0001
0.006
0.066
0.934
1.000

0.89
0.230
1.233
2.359
1.209
1.117
.949

0.074
0.532
0.014
0.006
0.038
0.184
1.000

0.938
1.177
1.181
1.186
.967
.950
1.095

0.208
1.000
<0.0001
0.608
1.000
1.000
1.000

Multiple logistic regression analysis was performed using no apparent retinopathy group as a reference.
OR: Odds ratio adjusting for other covariates in the logistic regression model.
Bonferronis multiple testing correction was applied for the subgroup analysis.

in the systemic circulation were proposed to explain its local

reparative mechanism of endothelial dysfunction [14]. In this
study, TNF-a levels were higher in patients with proliferative
retinopathy, but the significance of this relation was not
supported by multiple regression analysis. In another study,
TNF levels in tears were found to be highly correlated with
severity of retinopathy [21], which suggests local TNF-a
production has greater clinical significance.
Our study also shows elevated adiponectin levels in
patients with macroalbuminuria are independent of diabetic
duration, eGFR, HbA1c, medications, and hypertension.
Several authors have reported that adiponectin levels are
positively correlated with diabetic nephropathy stage independently of renal function in type 1 and 2 diabetes [2224].
Furthermore, biomarkers, such as, sVCAM-1, PAI-1, IL-18, and
total adiponectin, have been shown to have predictive values
for the progression of progression of kidney disease [2527].
However, in the present study, multiple logistic regression
analysis showed adiponectin had no independent effect on
macroalbuminuria. Small number of macroalbuminuria (n = 34)
might influenced different result with previous results. Further
studies are needed to determine the nature of the causative
relation between adiponectin and diabetic nephropathy.
Furthermore, we found inflammatory cytokine and adiponectin levels were not related to subclinical atherosclerosis
markers (cIMT, carotid plaque), whereas previous studies have
reported plasma adiponectin was negatively correlated with
carotid IMT in type 2 diabetes [28], but this inverse correlation
was not found in 112 non-obese diabetes with type 2 diabetes
[29]. These observations could be explained by difference
between study populations. For example, in our cohort,
median duration of diabetes was only 4 years and cardiovascular risk factors were unexceptional, that is, blood pressure
and lipid profiles were within normal ranges and glycemic
control was relatively good. Unfortunately, smoking histories
were not assessed in the present study due to missing data.
We suggest long term observational studies be undertaken to
investigate relationship between theses cytokines and atherosclerosis outcomes.
In general, studies about the impact of adipose tissue on
the development and progression of type 2 diabetes have
been conducted on obese patients and made comparisons
between type 2 diabetes and normoglycemic controls [3033].
Doupis et al. reported diabetes and obesity have different
effects on vascular reactivity, and suggested the possibility

that obesity and diabetes are related in different ways with

inflammatory cytokines and growth factors [34].
In the present study, total and HMW adiponectin levels
were higher in obese diabetes patients with macroalbuminuria, and total and HMW adiponectin levels were higher in
obese patients (n = 9) with proliferative retinopathy than nonobese patients (n = 12) with proliferative retinopathy (P = 0.011
and P = 0.236, respectively, data was not shown in the table).
Although number of patients with proliferative retinopathy
was small, the difference of serum total adiponectin between
non-obese and obese patients with proliferative retinopathy
(4.97 (3.777.86) vs. 8.98 (6.6711.35)), was bigger than one
between all obese and no-obese group (4.21 (3.186.84) vs. 3.64
(2.585.46)). It suggests that vascular complication affect the
total adiponectin level more significantly in obese group.
However, obesity was not considered in previous studies
[6,20].
Insulin resistance is a well-known characteristic of obesity
and type 2 diabetes, and has been associated with diminished
adiponectin receptor expression. Furthermore, it has been
suggested that increases in adiponectin levels reflect a
compensatory mechanism [35,36]. It has also been suggested
adiponectin expression might be upregulated in obese type 2
diabetes as compared with non-obese type 2 diabetes patients,
and that these patients show adiponectin resistance. This
suggestion implies adiponectin may be the missing link
between obesity, diabetes, and microvascular complications,
and that other factors are likely to play more prominent roles
in the pathogenesis of diabetic microvascular complications in
non-obese type 2 diabetes.
In this cross-sectional study, we investigated relationships
between cytokines and the severity of microvascular complications with respect to BMI in a cohort of 761 Korean patients
with T2DM. The obese group exhibit disturbed lipid profiles,
higher blood pressures, and a lower grade inflammatory state
higher hs-CRP and lower total and HMW adiponectin levels
than non-obese group. Furthermore, in the obese group, total
and HMW adiponectin levels were higher in patients with
macroalbuminuria, and total and HMW adiponectin and TNFa levels were higher in patients with proliferative retinopathy.
Of note, total adiponectin was found to contribute significantly
to proliferative retinopathy in obese patients.
Our study has several limitations. First, due to the crosssectional nature of this study, we could not assess the
causative roles of inflammatory cytokines and adipokines.

64

diabetes research and clinical practice 111 (2016) 5865

Second, adiponectin levels are known to be gender

dependent, but we designed the study to investigate the
effect of BMI, not of gender. Nevertheless, the analysis was
adjusted for gender, and thus, the effect of gender on
adipokine levels was ruled out. Third, we used BMI to define
obesity, but body fat composition, especially, truncal adipose
tissue would have provided a more reliable basis for evaluating the metabolic effects of obesity. Fourth, we did not
consider the effect of habitual physical activity on inflammatory adipokine levels. Difference of inflammatory cytokine and
adipokine levels by the obesity suggest that inflammatory
cytokines and adiponectin influenced on vascular complication differently in obese and non-obese patients with diabetes.
In diabetes, effect of cytokine should be considered by obesity.
Long term follow up of our cohort could aid to evaluate effect
of these cytokines on vascular complication in type 2 diabetes.

[8]

[9]

[10]

[11]

[12]

Conflict of interest
All of the authors have read and approved this manuscript and
have no conflict of interest to disclose and industry relationship.

Acknowledgement

[13]

[14]

This study was supported by Inha University Research Grant.

[15]

Appendix A. Supplementary data

[16]

Supplementary data associated with this article can be

found, in the online version, at http://dx.doi.org/10.1016/j.
diabres.2015.10.017.

[18]

references
[19]
[1] Hotta K, Funahashi T, Arita Y, Takahashi M, Matsuda M,
Okamoto Y, et al. Plasma concentrations of a novel,
adipose-specific protein, adiponectin, in type 2 diabetic
patients. Arterioscler Thromb Vasc Biol 2000;20:15959.
[2] Arita Y, Kihara S, Ouchi N, Takahashi M, Maeda K,
Miyagawa J, et al. Paradoxical decrease of an adiposespecific protein, adiponectin, in obesity. Biochem Biophys
Res Commun 1999;257:7983.
[3] Kumada M, Kihara S, Sumitsuji S, Kawamoto T, Matsumoto
S, Ouchi N, et al. Coronary artery disease. Association of
hypoadiponectinemia with coronary artery disease in men.
Arterioscler Thromb Vasc Biol 2003;23:859.
[4] Kizer JR, Barzilay JI, Kuller LH, Gottdiener JS. Adiponectin
and risk of coronary heart disease in older men and
women. J Clin Endocrinol Metab 2008;93:335764.
[5] Yilmaz MI, Sonmez A, Acikel C, Celik T, Bingol N, Pinar M,
et al. Adiponectin may play a part in the pathogenesis of
diabetic retinopathy. Eur J Endocrinol 2004;151:13540.
[6] Kato K, Osawa H, Ochi M, Kusunoki Y, Ebisui O, Ohno K, et al.
Serum total and high molecular weight adiponectin levels
are correlated with the severity of diabetic retinopathy and
nephropathy. Clin Endocrinol (Oxf) 2008;68:4429.
[7] Expert Committee on the Diagnosis and Classification of
Diabetes Mellitus (2003) Report of the expert committee on

[20]

[21]

[22]

[23]

[24]

the diagnosis and classification of diabetes, mellitus.

Diabetes Care 2003;26(Suppl. 1):S520.
Expert WHO, Consultation. Appropriate body-mass index
for Asian populations and its implications for policy and
intervention strategies. Lancet 2004;363:15763.
Friedewald WT, Levy RI, Fredrickson DS. Estimation of the
concentration of low-density lipoprotein cholesterol in
plasma, without use of the preparative ultracentrifuge. Clin
Chem 1972;18:499502.
Levey AS, Coresh J, Greene T, Marsh J, Stevens LA, Kusek
JW, et al. Expressing the Modification of Diet in Renal
Disease Study equation for estimating glomerular filtration
rate with standardized serum creatinine values. Clin Chem
2007;53:76672.
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. Diabetologia
1985;28:4129.
Wilkinson CP, Ferris 3rd FL, Klein RE, Lee PP, Agardh CD,
Davis M, et al. Proposed international clinical diabetic
retinopathy and diabetic macular edema disease severity
scales. Ophthalmology 2003;110:167782.
Hanley AJ, Bowden D, Wagenknecht LE, Balasubramanyam
A, Langfeld C, Saad MF, et al. Associations of adiponectin
with body fat distribution and insulin sensitivity in
nondiabetic Hispanics and African-Americans. J Clin
Endocrinol Metab 2007;9:266571.
Nigro E, Scudiero O, Monaco ML, Palmieri A, Mazzarella G,
Costagliola C, et al. New insight into adiponectin role in
obesity and obesity-related diseases. Biomed Res Int
2014;2014:658913.
Yang WS, Lee WJ, Funahashi T, Tanaka S, Matsuzawa Y,
Chao CL, et al. Weight reduction increases plasma levels of
an adipose-derived anti-inflammatory protein,
adiponectin. J Clin Endocrinol Metab 2001;86:38159.
Esposito K, Pontillo A, Di Palo C, Giugliano G, Masella M,
Marfella R, et al. Effect of weight loss and lifestyle changes
on vascular inflammatory markers in obese women: a
randomized trial. JAMA 2003;289:1799804.
Yu JG, Javorschi S, Hevener AL, Kruszynska YT, Norman RA,
Sinha M, et al. The effect of thiazolidinediones on plasma
adiponectin levels in normal, obese, and type 2 diabetic
subjects. Diabetes 2002;51:296874.
Akiko H, Koji O, Shinji K, Kenneth W, Noriyuki O.
Adiponectin suppresses pathological microvessel
formation in retina through modulation of tumor necrosis
factor-a expression. Circ Res 2009;104:105865.
Costagliola C, Daniele A, dellOmo R, Romano MR, Aceto F,
Agnifili L, et al. Aqueous humor levels of vascular
endothelial growth factor and adiponectin in patients with
type 2 diabetes before and after intravitreal bevacizumab
injection. Exp Eye Res 2013;110:504.
Costagliola C, Romano V, De Tollis M, Aceto F, dellOmo R,
Romano MR, et al. TNF-alpha levels in tears: a novel
biomarker to assess the degree of diabetic retinopathy.
Mediat Inflamm 2013;2013:629529.
Frystyk J, Tarnow L, Hansen TK, Parving HH, Flyvbjerg A.
Increased serum adiponectin levels in type 1 diabetic
patients with microvascular complications. Diabetologia
2005;48:19118.
Saraheimo M, Forsblom C, Fagerudd J, Teppo AM,
Pettersson-Fernholm K, Frystyk J, et al. Serum adiponectin
is increased in type 1 diabetic patients with nephropathy.
Diabetes Care 2005;28:14104.
Looker HC, Krakoff J, Funahashi T, Matsuzawa Y, Tanaka S,
Nelson RG, et al. Adiponectin concentrations are influenced
by renal function and diabetes duration in Pima Indians with
type 2 diabetes. J Clin Endocrinol Metab 2004;89:40107.

diabetes research and clinical practice 111 (2016) 5865

[25] Kollerits B, Fliser D, Heid IM, Ritz E, Kronenberg F. Genderspecific association of adiponectin as a predictor of
progression of chronic kidney disease: The Mild to Moderate
Kidney Disease Study. Kidney Int 2007;71:127986.
[26] Persson F, Rossing P, Hovind P, Stehouwer CD, Schalkwijk
CG, Tarnow L, et al. Endothelial dysfunction and
inflammation predict development of diabetic
nephropathy in the Irbesartan in Patients with Type 2
Diabetes and Microalbuminuria (IRMA 2) study. Scand J Clin
Lab Invest 2008;68:7318.
[27] Astrup AS, Tarnow L, Pietraszek L, Schalkwijk CG,
Stehouwer CD, Parving HH, et al. Markers of endothelial
dysfunction and inflammation in type 1 diabetic patients
with or without diabetic nephropathy followed for 10 years:
association with mortality and decline of glomerular
filtration rate. Diabetes Care 2008;31:11706.
[28] Dullaart R, de Vries R, van Tol A, Sluiter W. Lower plasma
adiponectin is a marker of increased intimamedia
thickness associated with type 2 diabetes mellitus and with
male gender. Eur J Endocrinol 2007;156:38794.
[29] Saif A, Abdelhamid A, Assem M, Mousa S. Plasma
adiponectin and carotid intimamedia thickness in nonobese patients with type 2 diabetes. J Diabetes Complicat
2015;29:80810.

65

[30] Al-Daghri N, Al-Rubean K, Bartlett WA, Al-Attas O, Jones

AF, Kumar S. Serum leptin is elevated in Saudi Arabian
patients with metabolic syndrome and coronary artery
disease. Diabet Med 2003;20:8327.
[31] Kim C, Park J, Park J, Kang E, Ahn C, Cha B, et al.
Comparison of body fat composition and serum
adiponectin levels in diabetic obesity and non-diabetic
obesity. Obesity 2006;14:116471.
[32] Sivitz WI, Wayson SM, Bayless ML, Larson LF, Sinkey C, Bar
RS, et al. Leptin and body fat in type 2 diabetes and
monodrug therapy. J Clin Endocrinol Metab 2003;88:154353.
[33] Tatti P, Masselli L, Buonanno A, Di Mauro P, Strollo F. Leptin
levels in diabetic and nondiabetic subjects. Endocrine
2001;15:3058.
[34] Doupis J, Rahangdale S, Gnardellis C, Pena SE, Malhotra A,
Veves A. Effects of diabetes and obesity on vascular
reactivity, inflammatory cytokines, and growth factors.
Obesity 2011;19(Silver Spring):72935.
[35] Kadowaki T, Yamauchi T. Adiponectin and adiponectin
receptors. Endocr Rev 2005;26:43951.
[36] Kadowaki T, Yamauchi T, Kubota N, Hara K, Ueki K, Tobe K.
Adiponectin and adiponectin receptors in insulin
resistance, diabetes, and the metabolic syndrome. J Clin
Invest 2006;116:178492.