DIABETES/METABOLISM RESEARCH AND REVIEWS

RESEARCH ARTICLE
Diabetes Metab Res Rev 2010; 26: 109114.
Published online 7 January 2010 in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/dmrr.1060

Are the associations of plasma leptin and

adiponectin with type 2 diabetes independent
of obesity in older Chinese adults?

Zhiqiang Wang1 *
Qin Zhuo2
Ping Fu2
Jiahua Piao2
Yuan Tian2
Jie Xu2
Xiaoguang Yang2
1

Centre for Chronic Disease, School of

Medicine, University of Queensland,
Herston QLD 4029, Australia
2

Institute of Nutrition and Food

Safety, Chinese Center for Disease
Control and Prevention, Beijing,
China
*Correspondence to:
Zhiqiang Wang, School of Medicine,
University of Queensland, Room
317, Edith Cavell Building, Royal
Brisbane and Womens Hospital,
Herston QLD 4029, Australia.
E-mail: [email protected]

Abstract
Background China has experienced a rapid increase in diabetes. In this
study, we assessed whether the associations of two adipocyte-derived
hormones, leptin and adiponectin, with type 2 diabetes are independent
of obesity in older Chinese adults.
Methods In this matched casecontrol study, each of the 619 diabetes and
impaired fasting glucose (IFG) cases aged 6096 years was matched to a
control by age, sex, waist circumference and body mass index (BMI).
Results Before matching, IFG and diabetes cases had significantly lower
adiponectin and higher leptin concentrations than the participants with
normal glucose. After matching for age, sex, waist circumference and BMI,
the differences between cases and controls remained significant (p < 0.001)
in adiponectin but not in leptin (p = 0.77). Adjusted odds ratios for the
combined outcome of diabetes and IFG were 1.03 (95% confidence interval:
0.88, 1.21; p = 0.71) for one standard deviation increase in plasma leptin
and 0.79 (95% confidence interval: 0.69, 0.91; p < 0.001) for one standard
deviation increase in plasma adiponectin.
Conclusion Without adjustment for obesity related body size measurements
of waist circumference and BMI, both adiponectin and leptin are associated
with diabetes and IFG. After adjustment, adiponectin is independently
associated with diabetes and IFG, but there is no independent association
between leptin and either diabetes or IFG. Our findings suggest that
adiponectin provides extra-predictive power beyond obesity while leptin
does not independently predict the risk of diabetes and IFG in older Chinese
adults. Copyright 2010 John Wiley & Sons, Ltd.
Keywords
matching

leptin; adiponectin; diabetes; impaired fasting glucose; obesity;

Introduction

Received: 17 June 2009

Revised: 25 November 2009
Accepted: 30 November 2009

Copyright 2010 John Wiley & Sons, Ltd.

Since the identification of the obese gene that encodes leptin [1], the possible role of leptin in type 2 diabetes in humans has attracted researchers
interests [25]. Increased plasma leptin concentrations strongly correlate
with elevated total adipose tissue and are considered to be a risk factor for type 2 diabetes. It is still not clear if the increased circulation
leptin level is a risk or protective factor for diabetes independent of obesity in human populations [6,7]. Some studies have shown that leptin is

110

an independent risk predictor of diabetes in men but

not in women in both Mauritian and Japanese American
populations [3,4]. Schmidt et al. concluded that highleptin levels are associated with an increased risk
of developing diabetes but they revealed a protective
association between leptin and diabetes after adjusting for
body mass index (BMI), waist to hip ratio, fasting insulin,
inflammation score, hypertension and triglycerides [6].
However, in the Hoorn study, adjusting for a similar
set of variables, Snijder et al. did not find a significant
negative association between leptin levels and diabetes
[7]. On the other hand, adiponectin, a dominant secretory
product of adipocytes, has been consistently found to be
associated with the development of type 2 diabetes in
several populations [817].
China has experienced a rapid increase in obesity,
diabetes, hypertension and dislipidemia in recent years
[1820]. However, little is known whether the plasma
leptin concentration is associated with diabetes and if
so, whether the association is independent of obesity in
Chinese population. The 2002 National Nutrition and
Health Survey (CHNS) was a comprehensive survey
in nutrition and health in China [21]. We conducted
an individually matched casecontrol study using data
derived from the survey to assess the associations of
plasma leptin and adiponectin with type 2 diabetes in
older Chinese adults aged 6096 years. We particularly
focussed on assessing whether these associations were
independent of obesity in older Chinese adults.

Methods
Cases and controls
Study subjects included all survey participants aged
6096 years in 18 major cities in China. One fasting blood
sample was taken from each participant. Diabetes cases
included 389 subjects with a fasting glucose 7.0 mmol/L
or with a previous diagnosis of diabetes. Another 271
subjects who did not meet the above criteria for diabetes
but with a fasting glucose >5.6 mmol/L and <7.0 mmol/L
(271) were categorized as having impaired fasting glucose
(IFG). Each case was matched to a participant (control),
who was free from diabetes and IFG, by sex, age, BMI
and waist circumference. We divided the age variable
into four groups: 6064, 6569, 7075 and 75 years,
and each of waist circumference and BMI variables into
five groups according to their quintiles. A control must
have matched to the case in all four categorical variables:
age, sex, waist circumference and BMI. When multiple
eligible controls were identified for a specific case, we
calculated the differences in standard deviation scores in
age, BMI and waist circumference between the case and
all potential controls, and chose the one with the smallest
sum of the differences in all three continuous variables.
A total of 619 casecontrol pairs (361 pairs for diabetes
and 258 pairs for IFG) were established. This project was
approved by the Academic Committee of the Institute of
Copyright 2010 John Wiley & Sons, Ltd.

Z. Wang et al.

Nutrition and Food Safety of Chinese Center for Disease

Control and Prevention.

Leptin and adiponectin measurements

The plasma samples collected during the survey were
stored at 70 C and retrieved for testing leptin and
adiponectin concentrations. Plasma leptin (ng/mL) and
adiponectin (g/mL) were determined with the commercially available ELISA kits (Phoenix Pharmaceuticals, Inc.,
Belmont, CA, USA). All procedures as described in the
manufacturers instructions were followed with quality
control parameters within the expected range recommended by the manufacturer. Every tenth sample was
duplicated on the same plate. The minimum detectable
concentration of leptin kit was 0.25 ng/mL with the assay
coefficient of variation (CV) <3% and the inter-assay
CV <10%. The minimum detectable concentration of
adiponectin kit was 0.15 ng/mL, with the intra-assay CV
ranging from 3% to 6% and the inter-assay CV <10%.
The adiponectin values in this study represented the total
adiponectin measurements of trimer, hexamer and highmolecular weight (HMW) forms in plasma.
The measurements of body weight, height, BMI,
triglycerides, high-density lipoprotein (HDL) cholesterol
have been described elsewhere [22]. High triglycerides
were defined as triglycerides 1.7 mmol/L and low
HDL as HDL <1.03 mmol/L for men and 1.29 mmol/L
for women. Current self-identified tobacco smoking and
alcohol drinking data were collected through a structured
questionnaire.

Statistical analysis
As some previous studies have shown that increased
leptin levels associated with diabetes in men but not
in women [3,4], we conducted analyses for men and
women separately. We used the paired t-test to assess the
differences in means and the McNemar test to assess
the differences in proportions. As the plasma leptin
and adiponectin values were skewed, the logarithmic
transformed values were used and the geometric means
were presented. Conditional logistic regression was used
to examine the associations of adiponectin and leptin
with diabetes. Crude and adjusted odds ratios and their
95% confidence intervals (CI) corresponding to one
standard deviation increase in adiponectin and leptin
were calculated. As each case was matched to a control
with the same sex, the variable of sex cannot be a possible
confounder. However, the potential residual confounding
effects of age, waist circumference and BMI were further
assessed in the multiple conditional logistic regressions
along with other non-matching variables such as systolic
blood pressure, cigarette smoking and alcohol drinking.
Interactions between the exposures of interest (leptin and
adiponectin) and other variables were tested. All analyses
were performed using Stata 10 [23].
Diabetes Metab Res Rev 2010; 26: 109114.
DOI: 10.1002/dmrr

111

Leptin, Adiponectin and Type 2 Diabetes

Results
Before matching diabetes and IFG cases to their normal
glucose controls, IFG and diabetes cases had significantly
lower adiponectin and higher leptin concentrations than
the participants with normal glucose (Figure 1). After
matching for age, sex, waist circumference and BMI,
the differences in adiponectin remained significant (p <
0.001), whereas the difference in leptin was no longer
significant (p = 0.77) (Figure 2).
The characteristics of 619 casecontrol pairs are shown
in Table 1. The matching variables of age, sex, waist
circumference and BMI were similar between the case
and control groups. The participants in the case group
in women had a significantly higher level of systolic
blood pressure as well as significantly higher prevalence
of clinically diagnosed hypertension than their matched
controls, but similar prevalence in cigarette smoking
and alcohol drinking. The case group also had higher
prevalence of high triglycerides and low HDL than the

Table 1. Characteristics of cases and controls

Male
Number
Age, years
BMI, kg/m2
Waist circumference, cm
Weight, kg
Height, cm
Systolic BP, mmHg
Diastolic BP, mmHg
Hypertension, n(%)a
High triglycerides, n(%)b
Low HDL, n(%)c
Tobacco smoking, n(%)
Alcohol drinking, n(%)
Women
Number
Age, years
BMI, kg/m2
Waist circumference, cm
Weight, kg
Height, cm
Systolic BP, mmHg
Diastolic BP, mmHg
Hypertension, n(%)a
High triglycerides, n(%)b
Low HDL, n(%)c
Tobacco smoking, n(%)
Alcohol drinking, n(%)

Controls

Cases

p value

274
68.6 (5.7)
25.4 (3.1)
89.4 (9.2)
70.1 (10.2)
166.0 (5.9)
140.6 (22.0)
84.5 (11.4)
86 (31.4)
41 (15.0)
49 (17.9)
95 (34.7)
106 (38.7)

274
68.6 (5.7)
25.4 (3.2)
89.6 (9.4)
69.4 (10.7)
165.3 (6.3)
141.3 (20.9)
82.6 (11.9)
107 (39.1)
48 (17.5)
55 (20.1)
101 (36.9)
97 (35.4)

0.97
0.94
0.86
0.47
0.16
0.72
0.053
0.06
0.41
0.51
0.59
0.43

345
67.4 (5.6)
25.9 (3.7)
85.9 (9.5)
61.0 (9.8)
153.4 (5.6)
140.3 (22.2)
81.3 (10.9)
101 (29.3)
69 (20.0)
148 (17.9)
30 (8.7)
14 (4.1)

345
67.4 (5.6)
25.9 (3.7)
86.0 (9.7)
60.9 (9.4)
153.2 (5.8)
145.8 (24.5)
81.7 (10.6)
163 (47.2)
88 (25.5)
55 (20.1)
24 (7.0)
19 (5.5)

0.97
0.91
0.90
0.91
0.77
0.002
0.57
<0.001
0.08
0.51
0.40
0.37

BMI, body mass index; HDL, high-density lipoprotein.

a Clinically diagnosed hypertension.
b Triglycerides 1.7 mmol/L.
c HDL cholesterol <1.03 mmol/L for men and <1.27 mmol/L for women.

Figure 1. Geometric means of adiponectin and leptin according

to outcome status in older Chinese adults before matching. NG,
normal glucose; IFG, impaired fasting glucose

Figure 2. Geometric means of adiponectin and leptin according

to outcome status in older Chinese adults after matching. NG,
normal glucose; IFG, impaired fasting glucose
Copyright 2010 John Wiley & Sons, Ltd.

control group but the difference did not reach statistical

significance.
As shown in Table 2, in both men and women,
cases had significantly lower levels of adiponectin than
their matched controls. We also stratified the study
participants by the case status. Both the cases of IFG
and diabetes had lower adiponectin concentrations than
their corresponding controls. However, no significant
differences in leptin concentrations were observed
between cases and controls in all strata.
Odds ratios of diabetes corresponding to one standard
deviation increase in leptin and adiponectin concentrations are shown in Table 3. No significant associations between leptin concentrations and diabetes were
observed in either males or females. After further adjustment for age, BMI, waist circumference, systolic blood
pressure, cigarette smoking and alcohol drinking, the
association between leptin and diabetes remained not
significant. As previous studies showed that leptin is an
independent predictor of diabetes in men but not in
women [3,4], an interaction term between sex and leptin as well as the interaction terms of leptin with other
variables were assessed. No significant interactions were
identified. The adjusted odds ratios of diabetes corresponding to one standard deviation increment in leptin
were 0.93 (95% CI: 0.72, 1.21) and 1.11 (95% CI: 0.90,
1.37) for men and women, respectively. The association
of leptin with IFG was also not statistically significant.
Diabetes Metab Res Rev 2010; 26: 109114.
DOI: 10.1002/dmrr

112

Z. Wang et al.

Table 2. Geometric means of adiponectin and leptin concentrations among cases and controls
Stratum
By sex
Male
Female

By case status
IFG
Diabetes

Variable

Controls

Cases

Number
Adiponectin, g/mL
Leptin, ng/L
Number
Adiponectin, g/mL
Leptin, ng/L

274
10.7 (9.9, 11.7)
4.5 (3.9, 5.2)
345
13.1 (12.2, 14.1)
13.7 (12.6, 15.0)

274
9.2 (8.4, 10.0)
4.4 (3.8, 5.2)
345
11.3 (10.5, 12.1)
14.3 (13.2, 15.5)

Number
Adiponectin, g/mL
Leptin, ng/L
Number
Adiponectin, g/mL
Leptin, ng/L

258
12.4 (11.5, 13.5)
8.0 (6.9, 9.3)
361
11.7 (10.8, 12.7)
8.6 (7.6, 9.7)

258
11.1 (10.1, 12.1)
8.2 (7.0, 9.6)
361
9.8 (9.1, 10.6)
8.7 (7.8, 9.8)

p value

0.024
0.86
0.0058
0.40

0.022
0.78
0.0059
0.82

IFG, impaired fasting glucose.

Table 3. Crude and adjusted odds ratios of diabetes and impaired

fasting glucose stratified by sex and outcome status
Adjusted a

Crude

Leptin
Sex
Maleb
Femaleb
Case status
IFGc
Diabetesc
Combined
Adiponectin
Sex
Maleb
Femaleb
Case status
IFGc
Diabetesc
Combinedd

OR (95% CI)

p value

OR (95% CI)

p value

0.98 (0.76, 1.26)

1.09 (0.89, 1.34)

0.86
0.40

0.93 (0.72, 1.21)

1.11 (0.90, 1.37)

0.61
0.33

1.11 (0.87, 1.42)

1.00 (0.81, 1.24)
1.05 (0.89, 1.23)

0.40
1.00
0.58

1.10 (0.86, 1.41)

0.99 (0.80, 1.23)
1.03 (0.88, 1.21)

0.46
0.90
0.71

0.79 (0.64, 0.97)

0.79 (0.66, 0.94)

0.026 0.77 (0.62, 0.96)

0.0069 0.81 (0.68, 0.97)

0.020
0.023

0.77 (0.61, 0.97) 0.025 0.77 (0.61, 0.98) 0.031

0.80 (0.68, 0.94) 0.0069 0.80 (0.68, 0.94) 0.0082
0.79 (0.69, 0.90) <0.001 0.79 (0.69, 0.91) <0.001

IFG, impaired fasting glucose; BMI, body mass index; CI, confidence
interval; OR, odds ratio.
a Adjusted for age, BMI, waist circumference, systolic blood pressure,
cigarette smoking and alcohol drinking.
b Both IFG and diabetes.
c Both sexes.
d Cases included both IFG and diabetes with both sexes combined.

Adiponectin concentrations were negatively and significantly associated with diabetes. Both crude and adjusted
odds ratios show a protective effect of a higher adiponectin
level. The adjusted odds ratios were 0.77 (95% CI:
0.62, 0.96) and 0.81 (95% CI: 0.68, 0.91) for men and
women, respectively, for one standard deviation increase
in adiponectin. There were no significant interactions of
adiponectin with sex and other variables.

Discussion
In this study, we found that plasma adiponectin
concentrations independently predict the risk of diabetes
and IFG in older Chinese adults. Although leptin is
significantly higher among people with diabetes and IFG
Copyright 2010 John Wiley & Sons, Ltd.

than among those with normal glucose before matching,

leptin is not independently associated with diabetes or IFG
after adjusting for age, sex, BMI and waist circumference.
Our findings suggest that plasma leptin concentrations
do not provide additional predictive value other than
obesity as measured using BMI and waist circumference
in predicting the risk of diabetes and IFG in older Chinese
adults. Therefore, the contribution of leptin to the risk
diabetes is likely to be in the same causal chain as obesity.
On the other hand, adiponectin may contribute to the
risk of diabetes and IFG beyond the effect of adiposity as
measured by BMI and waist circumference.
Several recent studies have examined the predictive
values of leptin to the risk of diabetes using different study
designs. In a population-based study, Soderberg et al.
found that leptin predicts the development of diabetes
in Mauritian men but not in women [4]. In a Japanese
American study, McNeely et al. also reported that men
with higher leptin had an increased risk of diabetes
(relative risk: 1.78 per standard deviation change). On
the other hand, a casecohort study using data from
the Atherosclerosis Risk in Communities Study, Schmidt
et al. concluded a protective association after adjusting
for factors purportedly related to leptin [6]. A small
casecontrol study of Mexican Americans found that
leptin concentrations were not different in diabetic and
non-diabetic subjects [24]. Some population-based cohort
studies do not support the protective effect of leptin
[7,25]. In our study, we focussed on the populationbased data in older Chinese adults for the first time. The
cases and controls were individually matched according
to age, sex, BMI and waist circumference at the design
stage. We found that the sex-specific associations between
leptin and diabetes were strikingly similar, and there were
no independent associations between leptin and diabetes
or IFG in both sexes.
Contradicting to several previous studies reporting
leptin as a risk factor in men but not in women
[3,4], our data showed that men and women had a
similar association between leptin and diabetes. Our
findings are consistent with those in the Atherosclerosis
Risk in Communities Study [6] and the Hoorn study
Diabetes Metab Res Rev 2010; 26: 109114.
DOI: 10.1002/dmrr

113

Leptin, Adiponectin and Type 2 Diabetes

[7]. There are several possible explanations for the

discrepancies among different studies. It is possible that
the ethnic differences in the association between leptin
and diabetes exist. Although the discrepancies may be
due to differences in study populations and study designs,
special attention should be given to confounding and
residual confounding of body size measurements. Since
leptin is strongly correlated with obesity, it is a tricky
task to identify an appropriate model to separate the
independent effect using multiple regression techniques.
Several issues such as non-linear relationship between
body size measurements and diabetes and categorization
of the body size measurements should be carefully
addressed. To address those issues, we used the individual
matching method in combination with multiple logistic
regressions.
Unlike leptin, adiponectin was significantly associated
with diabetes and IFG. In this study, using the
large population-based matched casecontrol data, we
confirmed the independent associations of adiponectin
with IFG and diabetes in older Chinese adults. Our
findings are consistent with those from other populations
[817,25]. Studies of middle-aged and elderly Chinese
adults reported a negative association of adiponectin and
metabolic syndrome [22,26].
These results have potential implications in predicting
and preventing IFG and diabetes. When predicting an
individual risk of diabetes, body size measurements
carry all the information of leptin concentration in
predicting diabetes risk. However, the predictive value
of adiponectin levels cannot be fully represented by body
size measurements. Therefore, adiponectin is important
independent predictor of diabetes and IFG. Although
leptin and obesity may be in the same causal chain for
diabetes, they are strongly correlated. Further studies are
needed to assess whether modifying circulating leptin
levels beyond the effect of body size will reduce the risk
of diabetes. On the other hand, adiponectin can be useful
for predicting the risk of diabetes even for people with
the same body size. The interventions that increase the
adiponectin levels may also improve glycaemic status.
There are some potential limitations to our study.
First, this is a matched casecontrol study nested in
a population-based survey. As the cases of IFG and
diabetes were identified at the same time as the exposure
measurements were taken, we reported the associations
but were unable to establish causal relationships. Second,
we only used a relative homogenous group - older adults
in the major cities. Further studies are needed to validate
our findings in other populations. Third, we focussed on
assessing whether the associations are independent of
obesity (body size). Other potential confounding factors
could bias the observed associations. Fourth, as we
ascertained diabetes without using a glucose tolerance
test, the control group could be contaminated by some
unidentified cases of diabetes and impaired glucose
tolerance. As a result, the estimated association would
underestimate the true association. Finally, adiponectin
exists as multimers (trimer, hexamer and HMW forms)
Copyright 2010 John Wiley & Sons, Ltd.

in plasma. We only examined the associations of total

adiponectin with diabetes and IFG. Total adiponectin
and HMW adiponectin are highly correlated [27]. Recent
studies have shown that HMW adiponectin values have
a better predictive power for the risk of type 2 diabetes
than total adiponectin values [16,28,29]. Therefore, the
true association between HMW adiponectin and diabetes
in Chinese older adults is likely to be stronger than the
observed association of total adiponectin with diabetes in
this study.
In conclusion, in older Chinese adults, high-leptin levels
do not provide extra-predictive value beyond body size
(obesity) for the risk of diabetes. Low-adiponectin levels
can predict the risk of diabetes and IFG independent of
body size. Similar patterns of associations exist in both
men and women.

Acknowledgements
This work was supported by research grant from National Natural
Science Foundation of China (30671750). Zhiqiang Wang is
supported by National Health and Medical Research Council of
Australia (511013). The authors thank all staff and participants
of China National Nutrition and Health Survey in 2002.

Competing interests
The authors have no conflicts of interest.

References
1. Zhang Y, Proenca R, Maffei M, Barone M, Leopold L,
Friedman JM. Positional cloning of the mouse obese gene and
its human homologue. Nature 1994; 372: 425432.
2. Taylor SI, Barr V, Reitman M. Does leptin contribute to diabetes
caused by obesity? Science 1996; 274: 11511152.
3. McNeely MJ, Boyko EJ, Weigle DS, et al. Association between
baseline plasma leptin levels and subsequent development of
diabetes in Japanese Americans. Diabetes Care 1999; 22: 6570.
4. Soderberg S, Zimmet P, Tuomilehto J, et al. Leptin predicts the
development of diabetes in Mauritian men, but not women: a
population-based study. Int J Obes (Lond) 2007; 31: 11261133.
5. Malmstrom R, Taskinen MR, Karonen SL, Yki-Jarvinen H.
Insulin increases plasma leptin concentrations in normal subjects
and patients with NIDDM. Diabetologia 1996; 39: 993996.
6. Schmidt MI, Duncan BB, Vigo A, et al. Leptin and incident type 2
diabetes: risk or protection? Diabetologia 2006; 49: 20862096.
7. Snijder MB, Dekker JM, Bouter LM, et al. Comment on: Schmidt
MI, Duncan BB, Vigo A, et al. (2006) Leptin and incident type
2 diabetes: risk or protection? Diabetologia 49: 20862096.
Diabetologia 2007; 50: 234236.
8. Snijder MB, Heine RJ, Seidell JC, et al. Associations of
adiponectin levels with incident impaired glucose metabolism
and type 2 diabetes in older men and women: the hoorn study.
Diabetes Care 2006; 29: 24982503.
9. Mather KJ, Funahashi T, Matsuzawa Y, et al. Adiponectin,
change in adiponectin, and progression to diabetes in the
Diabetes Prevention Program. Diabetes 2008; 57: 980986.
10. Snehalatha C, Mukesh B, Simon M, Viswanathan V, Haffner SM,
Ramachandran A. Plasma adiponectin is an independent
predictor of type 2 diabetes in Asian Indians. Diabetes Care
2003; 26: 32263229.
11. Lindsay RS, Funahashi T, Hanson RL, et al. Adiponectin and
development of type 2 diabetes in the Pima Indian population.
Lancet 2002; 360: 5758.
Diabetes Metab Res Rev 2010; 26: 109114.
DOI: 10.1002/dmrr

114
12. Spranger J, Kroke A, Mohlig M, et al. Adiponectin and protection
against type 2 diabetes mellitus. Lancet 2003; 361: 226228.
13. Choi KM, Lee J, Lee KW, et al. Serum adiponectin concentrations
predict the developments of type 2 diabetes and the metabolic
syndrome in elderly Koreans. Clin Endocrinol (Oxf) 2004; 61:
7580.
14. Duncan BB, Schmidt MI, Pankow JS, et al. Adiponectin and the
development of type 2 diabetes: the atherosclerosis risk in
communities study. Diabetes 2004; 53: 24732478.
15. Daimon M, Oizumi T, Saitoh T, et al. Decreased serum levels of
adiponectin are a risk factor for the progression to type 2 diabetes
in the Japanese Population: the Funagata study. Diabetes Care
2003; 26: 20152020.
16. Nakashima R, Kamei N, Yamane K, Nakanishi S, Nakashima A,
Kohno N. Decreased total and high molecular weight
adiponectin are independent risk factors for the development of
type 2 diabetes in Japanese-Americans. J Clin Endocrinol Metab
2006; 91: 38733877.
17. Nakatani K, Noma K, Nishioka J, et al. Adiponectin gene
variation associates with the increasing risk of type 2 diabetes
in non-diabetic Japanese subjects. Int J Mol Med 2005; 15:
173177.
18. Liu M, Wang Z, Sun X, Chen Y, Zhang Q. Rapid increase in the
incidence of clinically diagnosed type 2 diabetes in Chinese in
Harbin between 1999 and 2005. Prim Care Diabetes 2007; 1:
123128.
19. Wang Y, Mi J, Shan XY, Wang QJ, Ge KY. Is China facing an
obesity epidemic and the consequences? The trends in obesity
and chronic disease in China. Int J Obes (Lond) 2007; 31:
177188.
20. Ma GS, Li YP, Wu YF, et al. The prevalence of body overweight
and obesity and its changes among Chinese people during 1992
to 2002. Zhonghua Yu Fang Yi Xue Za Zhi [Chin. J Prev Med]
2005; 39: 311315.

Copyright 2010 John Wiley & Sons, Ltd.

Z. Wang et al.
21. Li LM, Rao KQ, Kong LZ, et al. A description on the Chinese
national nutrition and health survey in 2002. Zhonghua Liu Xing
Bing Xue Za Zhi 2005; 26: 478484.
22. Zhuo Q, Wang Z, Fu P, et al. Comparison of adiponectin, leptin
and leptin to adiponectin ratio as diagnostic marker for
metabolic syndrome in older adults of Chinese major cities.
Diabetes Res Clin Pract 2009; 84: 2733.
23. StataCorp. Stata Statistical Software: Release 10. StataCorp LP:
College Station, TX, 2007.
24. Haffner SM, Stern MP, Miettinen H, Wei M, Gingerich RL.
Leptin concentrations in diabetic and nondiabetic MexicanAmericans. Diabetes 1996; 45: 822824.
25. Wannamethee SG, Lowe GD, Rumley A, Cherry L, Whincup PH,
Sattar N. Adipokines and risk of type 2 diabetes in older men.
Diabetes Care 2007; 30: 12001205.
26. Wang J, Li H, Franco OH, Yu Z, Liu Y, Lin X. Adiponectin and
metabolic syndrome in middle-aged and elderly Chinese. Obesity
2008; 16: 172178.
27. Aso Y, Yamamoto R, Wakabayashi S, et al. Comparison of
serum high-molecular weight (HMW) adiponectin with total
adiponectin concentrations in type 2 diabetic patients
with coronary artery disease using a novel enzyme-linked
immunosorbent assay to detect HMW adiponectin. Diabetes
2006; 55: 19541960.
28. Satoh H, Fujii S, Tsutsui H. The high-molecular weight (HMW)
form of adiponectin is a significant surrogate marker for the
prediction of type 2 diabetes mellitus in the Japanese population.
Nutr Metab Cardiovasc Dis 2009; epub ahead of print: DOI:
10.1016/j.numecd.2009.1005.1009.
29. Heidemann C, Sun Q, van Dam RM, et al. Total and highmolecular-weight adiponectin and resistin in relation to the
risk for type 2 diabetes in women. Ann Intern Med 2008; 149:
307316.

Diabetes Metab Res Rev 2010; 26: 109114.

DOI: 10.1002/dmrr