Original Article

|
doi: 10.1111/j.1365-2796.2008.02007.x

Circulating serum adiponectin levels in patients with

coronary artery disease: relationship to atherosclerotic
burden and cardiac function
J. V. Patel1, A. Abraheem2, O. Dotsenko1, J. Creamer2, M. Gunning2, E. A. Hughes1 & G. Y. H. Lip1
From 1Haemostasis Thrombosis and Vascular Biology Unit, University Department of Medicine, City Hospital, Birmingham;
and 2North Staffordshire Cardiac Centre, University Hospital of North Staffordshire, Stoke-on-Trent, North Staffordshire, UK

Abstract. Patel JV, Abraheem A, Dotsenko O, Creamer

J, Gunning M, Hughes EA, Lip GYH (City Hospital
and University Hospital of North Staffordshire; UK).
Circulating serum adiponectin levels in patients with
coronary artery disease: relationship to atherosclerotic burden and cardiac function. J Intern Med 2008;
264: 593598.
Background. Abnormal adipocyte function is implicated
in both the pathophysiology of coronary heart disease
(CHD) and cardiac function, where adiponectin provides a putative link. However, the utility of adiponectin as a discriminator of CHD severity is unclear and
may be confounded by cardiac function. We hypothesized that plasma adiponectin would relate to indices of
coronary artery disease severity (coronary atheroma
scores, CAS), ejection fraction (EF) and regional wall
motion abnormalities (RWMA) therein.
Method. We measured adiponectin using a cross-sectional approach, we measured plasma adiponectin
enzyme-linked immunosorbent assay in 204 consecutive patients (aged 3481 years) undergoing elective
coronary angiography.

Introduction
Coronary heart disease (CHD) continues to be the
leading cause of premature mortality and morbidity
worldwide [1]. The stratication of CHD risk amongst
asymptomatic individuals as well as those with overt
disease continues to be one of the major priorities of

Results. Levels of adiponectin decreased in an ordinal fashion across tertiles of increasing CAS
(P = 0.047), but were nonsignicantly correlated to
absolute values of CAS (P = 0.06). Adiponectin levels were unrelated to left ventricular dysfunction
related measures of RWMA or EF. On multivariate
analysis, (including factors relating to CHD risk,
history and medication) adiponectin levels were
independently inversely associated with triglycerides
(P = 0.001), CAS tertiles (P = 0.01) and positively
with age (P < 0.001).
Conclusion. Levels of adiponectin decreased with
coronary artery disease severity, without impact
from systolic dysfunction, but levels may be
moderated through established CHD risk factors
such as smoking and triglycerides. Further work is
warranted as to the clinical prognostic utility of this
marker amongst CHD patients.
Keywords:
adiponectin, atheroma score, cardiac
function, coronary heart disease, ejection fraction.

clinically orientated research, where efforts focus on

approaches to improve our estimates of atherosclerotic
disease burden. Specically, there is increasing interest in the evaluation of circulating biological markers
biomarkers measurable and quantiable biological
parameters that are used to enrich or even substitute
standard diagnostic invasive tests such as coronary
angiography [2].

Drs Patel and Abraheem contributed equally to this work.

2008 Blackwell Publishing Ltd

593

J. V. Patel et al.

Adipocytokines are hormones produced by adipocytes, which elicit a diverse range of metabolic
changes that are implicated in the pathophysiology of
atherosclerosis. Importantly, circulating concentrations
of adipocytokines are of both diagnostic and prognostic potential with respect to the cardiovascular disease
continuum [38]. Of the adipocytokines, the collagenlike protein adiponectin has received the most intense
evaluation as a potential cardiovascular biomarker [9],
where low levels are used to explicate increased CHD
risk [10, 11] in both healthy [12, 13] and high-risk
populations [14]. There are also emerging data that
adiponectin elicits a favourable effect on postmyocardial infarction left ventricular (LV) remodelling [15],
with the potential to attenuate cardiac myocyte hypertrophy [16] as well as to protect against the progression of heart failure [17]. Given that cardiac function
deteriorates in the presence of longstanding CHD, we
tested the hypothesis that plasma adiponectin levels
would be signicantly correlated to indices of coronary artery disease (CAD) severity [coronary atheroma scores (CAS)], ejection fraction (EF) and
regional wall motion abnormalities (RWMA) therein.

Methods
We measured plasma adiponectin levels in 204 consecutive patients (3481 years) with symptomatic, but
stable CAD, who were undergoing diagnostic coronary angiography at University Hospital of North
Staffordshire (Stoke, UK). Those patients with a
recent clinical history (within the last 3 months) of
acute coronary syndrome, coronary intervention or
cerebral vascular accident were excluded. Patients
were also excluded if they had a history of malignant
disease, chronic inammatory disease or major concomitant noncardiovascular disease (e.g. renal impairment, liver impairment, connective tissue disease,
etc). All the patients underwent a complete cardiovascular assessment on the day of their attendance for
cardiac catheterization. Left heart catheters were
inserted via the femoral or radial approach, and
images were digitally recorded.
Left ventricular function and the severity of CAD
were assessed by an experienced cardiologist blinded
594

Adiponectin and atherosclerotic burden

to the clinical details using methods described by the

Ad Hoc Committee on Grading Coronary Artery
Disease, Council on Cardiovascular Surgery, American Heart Association [15]. For LV function analysis,
the LV walls were divided into seven regions; each
region was scored as normal (= 1), hypokinetic (= 2),
akinetic (= 3), dyskinetic (= 4), aneurysmal (= 5) and
none (= 0) (if not seen, the sum of LV walls score
was added up). LV EF was measured by Phillips software (Eindoven, Holland). CAD severity was assessed
in two ways: rst by the number of diseased vessels
(03) with at least one signicant stenosis (>50% stenosis) and second by a CAS where 15 proximal segments of the major coronary arteries were examined
for atheromatous lesions. Atheromatous lesions in
each segment were evaluated for extension (number
of plaques) and size (% of vessel diameter involved).
The scores for extension and size were then multiplied together for each segment and the total score
was calculated as the sum of individual segment
scores divided by the number of segments analysed.
Patients were stratied by CAS tertiles (<3; 37; >7).
The study was approved by the local Research Ethics
Committee of Shropshire and Staffordshire Health
Authority (Stoke, UK) and all participants gave
written informed consent.

Laboratory
Blood samples were collected before angiography.
Plasma was separated by centrifugation (4 C,
20 min, 3000 rpm), aliquoted and stored at )70 C
for batch analysis. Adiponectin levels were measured
by enzyme-linked immunosorbent assay (ELISA)
using commercially available antibodies (R&D Systems, Abingdon, UK). Intra-assay coefcients of variation were 3% and inter-assay variance was 5%. The
lower limit of detection by this ELISA was
62.5 pg mL)1.

Statistical analysis and power calculation

The study was of a cross-sectional design. We
hypothesized signicant differences in levels of adiponectin between CHD patients stratied by disease
severity and cardiac function. A sample size of 60 per

2008 Blackwell Publishing Ltd Journal of Internal Medicine 264; 593598

J. V. Patel et al.

Adiponectin and atherosclerotic burden

group was needed to detect a standard deviation in

adiponectin means, at 80% power, P < 0.05 using a
two-sided test; hence, given our primary analysis of
CAS tertiles, we recruited in excess of n = 180 to
allow condence in our comparisons.
Statistical analysis was undertaken using spss version14 (SPSS Inc., Chicago, IL, USA). Two-tailed
bivariate correlations were determined using Spearmans correlation coefcient. Partial correlation analysis (two-tailed) was used to adjust for the effects of
age A comparison of risk factors between patient
groups across CAS tertiles was performed by oneway anova and KruskalWallis test for parametric
and nonparametric variables respectively. Central tendencies and variation were reported as means with
standard deviations for parametric data and medians
Table 1 Patient characteristics by
coronary atheroma score tertile

and interquartile ranges for nonparametric data. A

stepwise multiple regression analysis was used to
determine independent predictors of variation in
plasma adiponectin, where beta (95% CI) is reported
to reect the strength of association from independent
predictors. PLUM ordinal regression analysis was
used to determine the association between adiponectin
levels and ordinal variables such as CAS and NYHA,
where a pseudo r2 was reported to reect the strength
of the association. A P-value <0.05 was considered as
statistically signicant.

Results
Patient characteristics by CAS tertile strata are presented in Table 1. Individuals with higher score were
more likely to be male and they smoke and have

Coronary atheroma score tertile

Variable

Bottom

Middle

Top

P-value

Age (years)

60.0 (9.8)

63.5 (8.6)

63.0 (8.4)

0.073

Male gender

15 (23.8)

21 (47.7)

55 (77.5)

<0.001

Current smoker

2 (3.2)

11 (25)

11 (15.7)

0.005

Hypertension

22 (36.1)

17 (39.5)

34 (49.3)

0.289

Diabetes mellitus

14.5 (9)

22.7 (10)

31.4 (22)

0.072

Systolic blood pressure (mmHg)

141 (18.9)

142 (18.1)

143 (19.9)

0.878

Diastolic blood pressure (mmHg)

81 (11.2)

77.2 (9.6)

77.6 (12.3)

0.122

Body mass index (kg m)2)

28.9 (6.4)

29.3 (5.59)

28.4 (4.39)

0.719

Ejection fraction (%)

61 (14.7)

58.1 (15.4)

51.5 (18.9)

0.005

Total cholesterol (mmol l)1)

4.89 (1.12)

4.91 (1.07)

4.41 (1.05)

0.021

Triglycerides (mmol l)1)

1.4 (1.01.9)

1.8 (1.52.6)

1.6 (1.32.5)

0.003a

Adiponectin (ng ml )

287 (195409)

240 (182404)

206 (153302)

0.047a

Statin therapy (%)

37 (60.7)

37 (86.0)

55 (78.6)

0.008

Beta blocker therapy (%)

36 (59.0)

29 (67.4)

54 (77.1)

0.08

ACE inhibitor therapy (%)

17 (27.9)

19 (44.2)

22 (31.4)

0.20

Diuretic therapy (%)

9 (14.5)

9 (20.9)

14 (20.3)

0.62

Aspirin therapy (%)

41 (67.2)

32 (74.4)

52 (74.3)

0.61
0.29

)1

New York Heart Association classication (%)

I

13 (21.0)

6 (13.6)

10 (14.3)

II

14 (22.6)

12 (27.3)

10 (14.3)

III

1 (1.6)

3 (6.8)

2 (2.9)

23 (36.5)

25 (56.8)

49 (71.0)

Presence of any RWMA (%)

<0.001

ACE, angiotensin converting enzyme; IQR, interquartile range; RWMA, regional wall motion abnormalities.
One-way anova, aKruskallWallis test. Data are mean (SD), number (%) or median (IQR).

2008 Blackwell Publishing Ltd Journal of Internal Medicine 264; 593598 595

J. V. Patel et al.

Adiponectin and atherosclerotic burden

Left Ventricle RWMA Index, n (%)

Coronary atheroma score tertile

Bottom

60 (95.2)

0 (0)

2 (3.2)

0 (0)

1 (1.6)

Middle

40 (90.9)

3 (6.8)

1 (2.3)

0 (0)

0 (0)

Top

57 (82.6)

4 (5.8)

5 (7.2)

2 (2.9)

1 (1.4)

Signicance (chi-squared test)

NS

NS

NS

NS

NS

Table 2 Left ventricle regional

wall motion analysis (RWMA)
index by coronary atheroma
score tertiles

NS, not signicant.

lower total cholesterol and higher triglycerides.

Although EF signicantly declined across CAS strata,
there were no differences in RWMA (Table 2) and in
NYHA class were found between patients allocated to
three CAS tertile groups. No differences in cardiovascular medications use were found between patients in
CAS groups.
Levels of adiponectin decreased in an ordinal fashion
across tertiles of increasing CAS (P = 0.047)
(Table 1), but were neither nonsignicantly related to
absolute CAS values (P = 0.06) (Table 3) nor to the
number of the vessels with more than 50% stenosis of
the luminal diameter (data not shown).
Correlation analysis (Table 3) revealed that plasma
adiponectin levels were negatively correlated with
male gender, weight, body surface area, presence of
diabetes mellitus and statin therapy positively with
age; no correlations were observed with EF, RWMA
or NYHA class (data not shown). There was a trend
to positive correlation between adiponectin levels and
diuretic therapy, which may be either related to hidden chronic heart failure (CHF) or reects the tendency of weight reduction associated with diuretic
use.
On multivariate regression analysis (including variables from Tables 1 and 3, which were signicantly
associated with CAS and with plasma adiponectin
levels), the factors which independently predicted circulating adiponectin levels were triglycerides
(P < 0.001), current smoking status (P = 0.001), age
(P = 0.002) and CAS tertiles (P = 0.04). On partial
correlation analysis, the relationship between CAS
and adiponectin remained after adjusting for age (partial correlation coefcient = )0.21, P = 0.006).
596

Table 3 Correlations between serum adiponectin (ng ml)1)

levels and other laboratory and clinical variables
Spearmans correlation
Variable
Age (years)
Male gender (%)

coefcient (P-value)
0.265 (0.001)
)0.186 (0.009)

Weight (kg)

)0.185 (0.009)

Body mass index (kg m)2)

)0.124 (0.083)

Body surface area (m2)

)0.191 (0.007)

Hypertension (%)

0.032 (0.661)

Current smoking (%)

)0.236 (0.001)

Diabetes mellitus (%)

)0.159 (0.026)

Family history of ischaemic

)0.013 (0.86)

heart disease (%)

Serum total cholesterol (mmol l)1)
Serum triglycerides (mmol l)1)
Systolic blood pressure (mmHg)
Diastolic blood pressure (mmHg)

0.006 (0.933)
)0.279 (0.001)
0.111 (0.123)
0.076 (0.292)

Ejection fraction (%)

)0.047 (0.516)

Heart rate (beats min)1)

)0.025 (0.733)

Beta blocker therapy (%)

)0.120 (0.095)

Calcium antagonist therapy (%)

)0.021 (0.768)

Nitrate therapy (%)

0.082 (0.254)

Aspirin therapy (%)

)0.078 (0.276)

Clopidogrel therapy (%)

)0.002 (0.982)

Statin therapy (%)

)0.166 (0.02)

Angiotensin converting enzyme

)0.047 (0.098)

inhibitor therapy (%)

Diuretic therapy (%)
Coronary atheroma score

0.139 (0.053)
)0.134 (0.061)

(absolute numbers)

Discussion
Given the diverse metabolic remit relating to the regulatory role of adiponectin (insulin-sensitization, free

2008 Blackwell Publishing Ltd Journal of Internal Medicine 264; 593598

J. V. Patel et al.

Adiponectin and atherosclerotic burden

fatty acid oxidation, gluconeogenesis, anti-inammatory modulation of the endothelium and counter
action of pro-inammatory cytokines to name just a
few [18]), there is increasing interest in the utility of
this biomarker in CHD. Indeed, many of these
effects could contribute to protection against atherosclerosis. Moreover, several other vascular and myocardial effects of adiponectin have been described,
including the modulation of angiogenesis and apoptosis, attenuation of LV hypertrophy, brosis and postmyocardial infarction remodelling as well as possible
regulation of myocardial ischaemia and reperfusion
injury [15, 16].
Whilst this biomarker appears to be physiologically
specic for CHD, there is confusion as to the clinical
signicance of circulating levels especially across
the cardiovascular disease spectrum from asymptomatic individuals to patients with end stage heart failure. For example, several studies conducted in
patients with CHF have found a direct correlation
between adiponectin and brain-type natriuretic peptides [1921], where high, not low adiponectin levels
independently predicted mortality [20, 21]. There are
also the lack of compelling prospective data on adiponectin levels as a predictor of recurrent cardiovascular
events, although relatively small studies in patients
with end stage renal disease [22], diabetes mellitus
[23] and CHD [24] have been published some with
conicting ndings [2527]. However, a recent prospective study in Japanese CHD patients reported that
the measurement of adiponectin was of clinical utility
with respect to predicting future clinical outcomes
following percutaneous coronary intervention [22].
The positive association between age and adiponectin
was not expected. However, amongst hospital
patients, adiponectin has been implicated in the ageing process, where increased levels in the elderly may
reect a physiological protective response to combat
low level inammation and oxidative stress [28]. The
absence of association between absolute scores of
CAS and adiponectin levels may reect the presence
of a nonlinear relationship, that is, levels of adiponectin falls dramatically amongst extreme thresholds of
CAS. Alternatively, we may have been underpowered

to chart the ordinal association between the extended

range of CAS scores in comparison to tertiles.
Because of lipolytic inuence on triglyceride rich proteins [18], the coordination between adiponectin and
triglyceride levels was as anticipated, and may underline the benets of supplementation with long-chain
omega-3 polyunsaturated fatty acids and other interventions that lower the serum triglycerides in the secondary prevention of fatal and nonfatal myocardial
infarction [29]. In our study, we were interested to
dene whether cardiac function parameters had an
impact on circulating adiponectin levels in unselected
symptomatic (but stable) CAD patients with varying
severity. In our studied patient group, the levels of
adiponectin decreased with CAD severity without signicant impact from cardiac function. Therefore, we
are tempted to conclude that the adiponectin measurement might potentially add to the existing clinical
tools to dene individuals with the highest CHD burden, who might possibly benet from targeted interventional and surgical treatment modalities.
In conclusion, serum adiponectin levels decreased
with CAD severity without impact from LV systolic
dysfunction, but levels may be moderated through
established CHD risk factors such as smoking and triglycerides. Further work is warranted as to the clinical
prognostic utility of this marker amongst CHD
patients.

Conflict of interest statement

All authors declare no conicts of interest.

Acknowledgements
We thank Dr R Butler and Ms Julie Machin for their
help. We also acknowledge the support of the Sandwell
and West Birmingham Hospitals NHS Trust for the
Haemostasis Thrombosis and Vascular Biology Unit.

References
1 Reddy KS, Yusuf S. Emerging epidemic of cardiovascular
disease in developing countries. Circulation 1998; 97: 596601.

2008 Blackwell Publishing Ltd Journal of Internal Medicine 264; 593598 597

J. V. Patel et al.

2 Vasan RS. Biomarkers of cardiovascular disease: molecular basis

and practical considerations. Circulation 2006; 11: 233562.
3 Ouchi N, Walsh K. Adiponectin as an anti-inammatory factor.
Clinical Chimica Acta 2007; 380: 2430.
4 Iwashima Y, Katsuya T, Ishikawa K et al. Hypoadiponectinemia
is an independent risk factor for hypertension. Hypertension
2003; 43: 131823.
5 Schulze PC, Kratzsch J, Linke A et al. Elevated serum levels of
leptin and soluble leptin receptors in patients with advanced
chronic heart failure. Eur J Heart Fail 2003; 5: 3340.
6 Arita Y, Kihara S, Ouchi N et al. Paradoxical decrease of an
adipose-specic protein, adiponectin, in obesity. Biochem Biophys Res Commun 1999; 257: 7983.
7 Hotta K, Funahashi T, Arita Y et al. Plasma concentrations of a
novel, adipose-specic protein, adiponectin, in type 2 diabetic
patients. Arterioscler Thromb Vasc Biol 2000; 20: 15951599.
8 Lim HS, Tayebjee MH, Tan KT, Patel JV, Macfadyen RJ, Lip
GYH. Severity differences and relation to coronary artery disease serum adiponectin in coronary heart disease. Heart 2005;
91: 16056.
9 Patel JV, Lim HS, Hughes EA, Lip GY. Adiponectin and hypertension: a putative link between adipocyte function and atherosclerotic risk? J Hum Hypertens 2007; 21: 14.
10 von Eynatten M, Schneider JG, Humpert PM et al. Serum adiponectin levels are an independent predictor of the extent of coronary artery disease in men. J Am Coll Cardiol 2006; 47: 212426.
11 Hara K, Yamauchi T, Imai Y, Manabe I, Nagai R, Kadowaki T.
Reduced adiponectin level is associated with severity of coronary artery disease. Int Heart J 2007; 48: 14953.
12 Pischon T, Girman CJ, Hotamisligil GS, Rifai N, Hu FB, Rimm
EB. Plasma adiponectin levels and risk of myocardial infarction
in men. JAMA 2004; 291: 173037.
13 Frystyk J, Berne C, Berglund L, Jensevik K, Flyvbjerg A,
Zethelius B. Serum adiponectin is a predictor of coronary heart
disease: a population-based 10-year follow-up study in elderly
men. J Clin Endocrinol Metab 2007; 92: 5716.
14 Schulze MB, Shai I, Rimm EB, Li T, Rifai N, Hu FB. Adiponectin and future coronary heart disease events among men with
type 2 diabetes. Diabetes 2005; 54: 5349.
15 Shibata R, Izumiya Y, Sato K et al. Adiponectin protects against
the development of systolic dysfunction following myocardial
infarction. J Mol Cell Cardiol 2007; 42: 106574.
16 Shibata R, Ouchi N, Ito M et al. Adiponectin-mediated modulation of hypertrophic signals in the heart. Nat Med 2004; 10:
13849.
17 Liao Y, Takashima S, Maeda N et al. Exacerbation of heart failure in adiponectin-decient mice due to impaired regulation of

598

Adiponectin and atherosclerotic burden

18
19

20

21

22

23

24

25

26

27
28

29

AMPK and glucose metabolism. Cardiovasc Res 2005; 67:

70513.
Wolf G. Adiponectin: a regulator of energy homeostasis. Nutr
Rev 2003; 61: 2902.
Nakamura T, Funayama H, Kubo N et al. Association of
hyperadiponectinemia with severity of ventricular dysfunction in
congestive heart failure. Circ J 2006; 70: 155762.
Kistorp C, Faber J, Galatius S et al. Plasma adiponectin, body
mass index and mortality in patients with chronic heart failure.
Circulation 2005; 112: 175662.
Tamura T, Furukawa Y, Taniguchi R et al. Serum adiponectin
as an independent predictor of mortality in patients with congestive heart failure. Circ J 2007; 71: 62330.
Zoccali C, Tripepi G, Cambareri F et al. Adiponectin, metabolic
risk factors, and cardiovascular events among patients with endstage renal disease. J Am Soc Nephrol 2002; 13: 13441.
Costacou T, Zgibor JC, Evans RW et al. The prospective
association between adiponectin and coronary artery disease
among individuals with type 1 diabetes: the Pittsburgh Epidemiology of Diabetes Complications Study. Diabetologia 2005; 48:
418.
Inoue T, Kotooka N, Morooka T et al. High molecular weight
adiponectin as a predictor of long-term clinical outcome in
patients with coronary artery disease. Am J Cardiol 2007; 100:
56974.
Sattar N, Wannamethee G, Sarwar N et al. Adiponectin and coronary heart disease: prospective study and meta-analysis. Circulation 2006; 114: 6239.
Inami N, Nomura S, Inami O et al. Signicance of soluble
CD40 ligand, adiponectin and reactive oxygen metabolites in
aging. Arch Gerontol Geriatr 2008 (in press).
Ingelsson E, Riserus U, Berne C et al. Adiponectin and risk of
congestive heart failure. JAMA 2006; 295: 17724.
Wannamethee SG, Whincup PH, Lennon L, Sattar N. Circulating adiponectin levels and mortality in elderly men with and
without cardiovascular disease and heart failure. Arch Intern
Med 2007; 167: 15107.
GISSI-Prevention Investigators (Gruppo Italiano per lo Studio
della Sopravvivenza nellInfarto miocardico). Dietary supplementation with n-3 polyunsaturated fatty acids and vitamin E
after myocardial infarction: results of the GISSI-prevention trial.
Lancet 1999; 354: 447455.

Correspondence: Gregory Y. H. Lip, Haemostasis Thrombosis and

Vascular Biology Unit, University Department of Medicine, City
Hospital, Birmingham, UK.
(fax: 0121 554 4083; e-mail: g.y.h.lip@bham.ac.uk).

2008 Blackwell Publishing Ltd Journal of Internal Medicine 264; 593598