[Evaluation of serum adiponectin levels in patients with heart failure and relationship with functional capacity].

Abstract

We aimed to evaluate serum adiponectin levels in relation to the NYHA functional capacity class in patients with heart failure (HF).
The study included 49 patients (40 males, 9 females; mean age 63 years) with HF, whose functional capacity was NYHA class II to IV. Echocardiographic examination was performed and serum adiponectin levels were measured. The results were compared in relation to the NYHA classes and with those of 41 control subjects (24 males, 17 females; mean age 54.2 years) without HF.
Functional capacity was NYHA class II in 13 patients (26.5%), class III in 23 patients (46.9%), and class IV in 13 patients (26.5%). Compared to the control group, the HF group exhibited a significantly higher mean age (p=0.001), lower body mass index (p=0.004), decreased left ventricular ejection fraction (EF) (33.2+/-7.7% vs. 64.9+/-4.3%; p=0.0001), and increased serum adiponectin level (4.0+/-3.2 micromg/dl vs. 2.4+/-2.3 micromg/dl; p=0.009). Both EF (p=0.001) and adiponectin level (p=0.004) showed significant differences between the NYHA groups, with the latter showing a sharp increase from 2.6+/-2.6 micromg/dl in class II to 6.8+/-3.7 micromg/dl in class IV. In all paired comparisons between the three NYHA groups, EF and serum adiponectin level exhibited significant differences except for the serum adiponectin level for NYHA class II and III (for NYHA class II and IV, p=0.003; for class III and IV, p=0.008). In correlation analysis, serum adiponectin level was in a significantly inverse correlation with EF (r=-0.380, p=0.0001), and a positive correlation with the NYHA class (r=0.423, p=0.0001).
Serum adiponectin levels significantly increase in patients with HF, in parallel with deterioration in functional capacity and with significant decreases in EF.

Full text

Page 1

384
Türk Kardiyol Dern Arfl - Arch Turk Soc Cardiol 2009;37(6):384-390
The English version of this article is prepared for online access only.
Objectives: We aimed to evaluate serum adiponectin levels
in relation to the NYHA functional capacity class in patients
with heart failure (HF).
Study design: The study included 49 patients (40 males, 9
females; mean age 63 years) with HF, whose functional ca-
pacity was NYHA class II to IV. Echocardiographic examina-
tion was performed and serum adiponectin levels were me-
asured. The results were compared in relation to the NYHA
classes and with those of 41 control subjects (24 males, 17
females; mean age 54.2 years) without HF.
Results: Functional capacity was NYHA class II in 13 pati-
ents (26.5%), class III in 23 patients (46.9%), and class IV in
13 patients (26.5%). Compared to the control group, the HF
group exhibited a significantly higher mean age (p=0.001), lo-
wer body mass index (p=0.004), decreased left ventricular
ejection fraction (EF) (33.2±7.7% vs. 64.9±4.3%; p=0.0001),
and increased serum adiponectin level (4.0±3.2 μmg/dl vs.
2.4±2.3 μmg/dl; p=0.009). Both EF (p=0.001) and adiponec-
tin level (p=0.004) showed significant differences between the
NYHA groups, with the latter showing a sharp increase from
2.6±2.6 μmg/dl in class II to 6.8±3.7 μmg/dl in class IV. In all
paired comparisons among the three NYHA groups, EF and
serum adiponectin level exhibited significant differences ex-
cept for the serum adiponectin level for NYHA class II and III
(for NYHA class II and IV, p=0.003; for class III and IV,
p=0.008). In correlation analysis, serum adiponectin level was
in a significant inverse correlation with EF (r=-0.380,
p=0.0001), and a positive correlation with the NYHA class
(r=0.423, p=0.0001).
Conclusion: Serum adiponectin levels significantly increase
in patients with HF, in parallel with deterioration in functional
capacity and with significant decreases in EF.
Amaç: Kalp yetersizli¤i (KY) olan hastalarda serum adiponek-
tin düzeyleri NYHA fonksiyonel kapasite s›n›flar›na göre de-
¤erlendirildi.
Çal›flma plan›: Fonksiyonel kapasitesi NYHA s›n›f II-IV olan,
KY’li 49 hasta (40 erkek, 9 kad›n; ort. yafl 63) çal›flmaya al›n-
d›. Tüm hastalarda ekokardiyografik inceleme yap›ld› ve se-
rum adiponektin düzeyleri ölçüldü. Sonuçlar NYHA s›n›flar›
aras›nda ve klinik ve ekokardiyografik olarak KY saptanmayan
41 kiflilik (24 erkek, 17 kad›n; ort. yafl 54.2) kontrol grubuyla
karfl›laflt›r›ld›.
Bulgular: Fonksiyonel kapasite 13 hastada (%26.5) NYHA s›n›f
II, 23 hastada (%46.9) s›n›f III, 13 hastada (%26.5) s›n›f IV idi.
Kontrol grubu ile karfl›laflt›r›ld›¤›nda, KY hastalar›nda ortalama
yafl anlaml› derecede daha yüksek (p=0.001), beden kütle indek-
si daha düflük (p=0.004), sol ventrikül ejeksiyon fraksiyonu (EF)
daha düflük (%33.2±7.7 ve %64.9±4.3; p=0.0001), serum adipo-
nektin düzeyi daha yüksek (4.0±3.2 Ìmgr/dl ve 2.4±2.3 μmgr/dl;
p=0.009) bulundu. Sol ventrikül EF ve serum adiponektin düzeyi
NYHA gruplar› aras›nda anlaml› farkl›l›k gösterdi (p=0.001 ve
p=0.004). NYHA s›n›f II’de ortalama 2.6±2.6 Ìmgr/dl olan serum
adiponektin düzeyi s›n›f IV’de 6.8±3.7 μmgr/ dl’ye yükseldi.
NYHA s›n›flar›n›n ikili karfl›laflt›rmas›nda, EF ikili tüm gruplar ara-
s›nda anlaml› farkl›l›k gösterirken, serum adiponektin düzeyi,
NYHA II-III gruplar› d›fl›nda, di¤er gruplar aras›nda anlaml› dere-
cede farkl›yd› (NYHA II-IV, p=0.003; NYHA III-IV, p=0.008). Ko-
relasyon analizinde, serum adiponektin düzeyi EF ile anlaml› ne-
gatif iliflki (r=-0.380, p=0.0001), NYHA s›n›f› ile anlaml› pozitif ilifl-
ki gösterdi (r=0.423, p=0.0001).
Sonuç: Kalp yetersizli¤i olan hastalarda serum adiponektin
düzeyi anlaml› art›fl göstermekte ve bu art›fl fonksiyonel kapa-
site kötülefltikçe, EF’deki azalma ile birlikte, daha belirgin
olmaktad›r.
Key words: Adiponectin/blood; biological markers; heart failure;
ventricular function, left.
Anahtar sözcükler: Adiponektin/kan; biyolojik belirteç; kalp
yetersizli¤i; ventrikül fonksiyonu, sol.
Received: 27.12.2008; Accepted: 13.05.2009
Corresponding address: Dr. Dursun Dursuno¤lu. Pamukkale Üniversitesi T›p Fakültesi, Kardiyoloji Anabilim Dal›, K›n›kl› Kampüsü, 20200 Denizli.
Tel: +90 - 258 - 444 07 28 e-mail: dursundursunoglu@yahoo.com
Evaluation of serum adiponectin levels in patients with heart failure and
relationship with functional capacity
Kalp yetersizli¤i olan hastalarda serum adiponektin düzeylerinin
de¤erlendirilmesi ve fonksiyonel kapasite ile iliflkisi
Mehmet Ozturk, M.D., Dursun Dursunoglu, M.D., Hidayet Goksoy, M.D., Simin Rota, M.D.,
1Sukru Gur, M.D.
Pamukkale University, Faculty of Medicine, Department of Cardiology,
1Department of Biochemistry, Denizli

Page 2

385
Evaluation of serum adiponectin levels in patients with heart failure and relationship with functional capacity
The English version of this article is prepared for online access only.
In recent years, adipocytokines such as interleukin-
6 (IL-6), interleukin-10 (IL-10), tumor necrosis factor-
alpha (TNF-·) are also thought to be responsible for the
pathogeneses of heart failure (HF) and its possible cau-
ses such as coronary artery disease (CAD), hypertensi-
on, diabetes mellitus, and obesity.
physiological role of adiponectin which is released
from adipose tissue has not been yet established, its an-
ti-atherogenic and anti-inflammatory effects on endot-
helial cells and macrophages have been reported.
rum adiponectin level has also been reported to decrea-
se in the presence of diabetes mellitus, metabolic
syndrome, obesity and CAD.
adiponectin levels have been suggested to prevent the
development of cardiovascular diseases and complica-
tions in healthy individuals.
also demonstrated that serum adiponectin levels are pa-
radoxically higher in patients with chronic heart failu-
re, and that they were an independent predictor of mor-
tality.
[1,2] Although the
[3-5]Se-
[6-8] Normal or high serum
[9]Some recent studies have
[10,11]
In this study, we evaluated serum adiponectin levels
according to the functional capacity classification in
patients with HF.
PATIENTS AND METHODS
The study included 49 patients (40 males, 9 fema-
les; mean age 63.0±10.6 years) with ischemic or hyper-
tensive HF, idiopathic cardiomyopathy or heart valve
disease whose functional capacity was NYHA class II
to IV, and 41 control subjects without HF as shown
through clinical and electrocardiographic evaluation
(24 males, 17 females; mean age 54.2±11.3). Of the 49
patients with HF, 16 had hypertension and 25 had di-
abetes mellitus, whereas 14 developed both.
Physical examination was performed and history
obtained from both the patient group and controls. Pa-
tients with functional capacity of NYHA class I, chro-
nic renal failure, and those with a known pulmonary di-
sease such as chronic obstructive pulmonary disease
were excluded from the study. All patients were recei-
ving treatment for HF including diuretics, angiotensin
converting enzyme (ACE) inhibitor or angiotensin re-
ceptor blockers (ARB), and beta-blockers.
The 12-lead ECG of the patients was performed at
rest and transthoracic echocardiographic assessment
was obtained in the left lateral decubitus position in ac-
cordance with recommendations of the American Soci-
ety of Echocardiography.
7 echocardiography device (General Electric, Horton,
Norway) was used during assessment. Two-dimensio-
nal echocardiography was used to estimate left ventri-
cular ejection fraction (EF) by the Simpson’s rule.
[12]A 2.5 mHz probe and Vivid
Routine biochemical analyses of all subjects were
performed using 12-hour fasting venous blood samp-
les. The collected samples for adiponectin were centri-
fuged at 4000 rpm for 7 minutes, and the supernatant
obtained was stored at -20
ons. Levels of serum total adiponectin were measured
by ELISA method (Adiponectin ELISA, BioVendor
Laboratory Medicine Inc., Czech Republic).
oC under laboratory conditi-
All patients were informed about the purpose of the
study and informed consents were obtained from all
patients. The study was approved by the local ethical
committee of our hospital.
Statistical analysis. The SPSS program (Statistical
Package for Social Sciences, v. 11.0 for Windows) was
used for statistical analysis and the results were expres-
sed as mean ± standard deviation. The Student t-test
was used for comparison of quantitative data and in the
assessment of subjects with/without HF, while Krus-
kal-Wallis test was used in the comparison of subgro-
ups of HF patients according to their functional capa-
city (NYHA class II, III, IV). The Mann-Whitney U-
test was used in paired comparison of control group
and patient subgroups according to their functional ca-
pacity. Pearson correlation test was also used in the as-
sessment of the correlation between serum adiponectin
levels and certain parameters. P<0.05 was considered
to be statistically significant.
RESULTS
CAD in 39 patients (79.6%), idiopathic cardiomyo-
pathy in six (12.2%), hypertensive heart disease in two
(4.1%), and other factors in two (12.2%) were respon-
sible for heart failure. On the other hand, 13 patients
(26.5%) were in NYHA class II, 23 patients (46.9%)
were in NYHA class III, and 13 patients (26.5%) were
in NYHA class IV.
The main characteristics of the patients and controls
are summarized in Table 1. Compared to the control
group, the HF group exhibited a significantly higher
mean age (p=0.001) and a lower body mass index
(p=0.004). On the other hand, no significant difference
was found between the groups in terms of hypertensi-
on, diabetes mellitus and smoking history.
Left ventricular ejection fraction was within normal
range in the controls, while it was significantly lower in
the patient group (p=0.0001). The serum adiponectin
levels were significantly higher in patients with HF
(4.0±3.2 μmg/dL) compared to the controls (2.4±2.3
μmg/dL) (p=0.009). Fasting blood glucose level was
significantly higher, while the total and HDL-cholesterol
levels were significantly lower in the patient group,

Page 3

386
Türk Kardiyol Dern Arfl
The English version of this article is prepared for online access only.
compared to the control group. Plasma LDL-cholesterol
level was not significantly different between the groups,
whereas triglyceride level was significantly lower in pa-
tients with HF (Table 1).
Table 2 shows comparison of the left ventricular EF
and serum adiponectin level based on NYHA classes.
The EF (p=0.001) and adiponectin (p=0.004) level
showed significant differences between the NYHA
groups. The increase in adiponectin levels was also
more apparent as the functional capacity worsened.
In all paired comparisons between the left ventricu-
lar EF and serum adiponectin level according to the
NYHA groups, a significant difference was found
between the groups in terms of left ventricular EF
(p=0.04, p=0.001 and p=0.01 for NYHA II-III, II-IV
and III-IV, respectively). On the other hand, no signi-
ficant difference was found between the NYHA II and
III groups in terms of serum adiponectin level, where-
as there was a significant difference between the
NYHA II-IV and III-IV groups (p=0.003 and
p=0.008).
In all paired comparisons between the controls and
each NYHA group in terms of age, left ventricular EF
and serum adiponectin level, a significant difference
was found in respect of all three parameters, apart
from the age and serum adiponectin level in the
NYHA II group (Table 2).
Age
Sex
Male
Female
Body mass index (kg/m2)
Hypertension
Diabetes Mellitus
Cigarette smoking history
Previous myocardial infarction
Coronary artery bypass grafting
Percutaneous coronary intervention
Left ventricular ejection fraction (%)
Adiponectin (μmg/dL)
Glucose (mg/dL)
Total cholesterol (mg/dL)
HDL-cholesterol (mg/dL)
LDL-cholesterol (mg/dL)
Triglyceride (mg/dL)
Creatinine (mg/dL)
Hemoglobin (g/dL)
NS: non-significant
40
9
16
25
15
39
18
10
24
17
17
14
18
-
-
-
58.5
41.5
41.5
34.2
43.9
81.6
18.4
32.7
51.0
30.6
79.6
36.7
20.4
63.0±10.6
26.9±4.3
33.2±7.7
4.0±3.2
138.4±53.1
169.0±.35.0
40.2±9.8
102.2±28.0
126.4±60.6
1.0±0.3
13.3±1.9
54.2±11.3
29.8±4.8
64.9±4.3
2.4±2.3
110.4±44.0
185.3±28.7
44.5±10.0
104.4±22.5
159.7±93.0
0.8±0.12
14.01±1.8
0.001
NS
0.004
NS
NS
NS
0.0001
0.009
0.009
0.019
0.045
NS
0.05
NS
NS
Number Percentage Mean±SD
Patient Group (n=49)
Number Percentage Mean±SD
Control Group (n=41)
p
Table 1. Comparison of clinical and laboratory findings and left ventricular ejection fraction of patients
Ejection fraction (%)
Adiponectin (μmg/dL)
Compared to controls
Age
Ejection fraction
Adiponectin
NS: non-significant
38.4±4.3
2.6±2.6
p
NS
0.0001
NS
NYHA II
(n=13)
33.9±7.4
3.2±2.1
p
0.0001
0.0001
0.03
NYHA III
(n=23)
27.0±7.1
6.8±3.7
p
0.005
0.0001
0.001
NYHA IV
(n=13)
0.001
0.004
p
Table 2. Evaluation of left ventricular ejection fraction and serum
adiponectin levels in patients with HF according to the NYHA classes and
paired comparison analysis of NYHA classes and the control group

Page 4

387
Evaluation of serum adiponectin levels in patients with heart failure and relationship with functional capacity
The English version of this article is prepared for online access only.
In correlation analysis, serum adiponectin level was
found to have a significant negative relationship with EF
(r=-0.380, p=0.0001), and a positive correlation with the
NYHA class (r=0.423, p=0.0001) (Figure 1). No signifi-
cant difference was found between serum adiponectin
level with age and BMI (r=0.204, p=0.06 and r=0.167,
p>0.05, respectively)
DISCUSSION
Heart failure is currently the major cause of morbi-
dity and mortality in adults. Several studies have de-
monstrated that neurohormonal activity resulting from
a compensatory mechanism and increased cytokine at
the beginning may have an adverse effect on prognosis
of the disease in the later stage.
creased plasma catecholamine has been shown to be
associated with increased morbidity and mortality in
patients with HF.
[13-16]For instance, in-
[13,14]
In our study, serum adiponectin levels were higher in
patients with HF predominantly (79%) in those with a
histology of ischemic disease compared to the controls
(4.0±3.2 μmg/dL vs. 2.4±2.3 μmg/dL; p=0.009). In ad-
dition, serum adiponectin levels were significantly diffe-
rent in the patient groups based on their functional capa-
city (NYHA II-III-IV). The increase in adiponectin level
was more apparent as functional capacity worsened. On
the other hand, paired comparisons showed no signifi-
cant difference but in the serum adiponectin levels bet-
ween the NYHA II and NYHA III groups. Therefore, se-
rum adiponectin levels can be suggested to increase sig-
nificantly in patients with a poor functional capacity
(NYHA III and IV), whereas no significant change was
observed in patients with a better functional capacity
(NYHA II). A recent study showed that serum adiponec-
tin levels increased in 175 consecutive patients with
chronic HF and that this increase was more apparent as
the functional capacity worsened (NYHA). During 2-ye-
ar-follow-up, serum adiponectin level was also shown to
be an independent predictor of morbidity and mortality
in the same study.
[11]
Given that no significant difference was observed
between the patient groups and controls in terms of the
incidence of hypertension and diabetes mellitus and that
there was no significant relation of serum adiponectin
levels with age and BMI, serum adiponectin levels can
be suggested to increase only due to HF. In other words,
the increase in serum adiponectin level is not relative in
patients with HF. On the other hand, several studies ha-
ve also shown that serum adiponectin the level is not
elevated, but is rather lower under such conditions.
According to the results obtained from several studies
suggesting that serum adiponectin levels decreased in
patients with CAD,
for the increased adiponectin levels in patients with HF
[6-9]
[1,4,6]it is obvious that HF accounted
Figure 1. The relationship of serum adiponectin level with (A) left ventricular ejection fraction and (B) NYHA class.

Page 5

388
Türk Kardiyol Dern Arfl
The English version of this article is prepared for online access only.
of predominantly ischemic origin. Although mean BMI
was significantly lower in patients with HF compared to
the controls in our study, subjects in both groups were
overweight. Therefore, although serum adiponectin le-
vel was reported to decrease in obese patients and to
increase in the event of losing weight,
ted that adiponectin levels were not significantly influ-
enced by BMI in our study. The absence of a signifi-
cant relationship between BMI and serum adiponectin
levels also supports our conclusion. However, the ob-
jective and method of our study is insufficient to exp-
lain the reason for increased serum adiponectin level
in HF patients. On the other hand, it can be suggested
that increased serum adiponectin levels have protecti-
ve effects to provide adequate hemodynamic in pati-
ents with HF as seen with the compensatory increased
in plasma catecholamine. Whether the increase in se-
rum adiponectin level in patients with HF is hazardo-
us (toxic) similar to the increased plasma catecholami-
ne is still not known. Although increased serum adipo-
nectin levels did not predict the development HF in a
study including 946 male adults without HF at baseli-
ne,
levels increased in patients with chronic HF and that
increased levels were independent predictor of morta-
lity.
patients with chronic HF having various BMI (<21,
21-25 and >25 kg/m2) groups evaluated serum total
and high molecular weight (HMW) adiponectin levels,
atrial (ANP), brain (BNP) natriuretic peptides and NT-
proBNP levels, and total adiponectin levels were fo-
und to be independent prognostic predictor in all gro-
ups.
proBNP (p=0.017) and plasma total adiponectin levels
(p=0.003) to be prognostic predictor in only subjects
with normal BMI. Authors concluded that measure-
ment of total adiponectin level was more useful than
of HMW adiponectin level in establishing the morta-
lity risk and that plasma total adiponectin level was an
independent prognostic predictor in HF patients with
particularly normal BMI.
Similar to our study, Nakamura et al.
that serum adiponectin levels increased as the functional
capacity worsened in HF patients. In another study inc-
luding 138 patients with HF, Takano et al.
ted adiponectin levels through sampling blood from the
aorta, coronary sinus and peripheral vein and found that
adiponectin levels significantly increased from aorta to-
wards the coronary sinus. The authors also reported that
the difference of adiponectin levels in coronary sinus
and aorta was positively associated with adiponectin
levels in blood samples obtained from peripheral vein. In
[17]it was sugges-
[18]several studies showed that serum adiponectin
[10,11]In another study including 449 consecutive
[19]In addition, subgroup analysis found serum NT-
[19]
[20]also showed
[21]investiga-
this study, authors investigated whether adiponectin le-
vel affected cardiac tissue in HF patients. Another study
including patients with dilated cardiomyopathy investi-
gated cardiac adiponectin system through endomyocar-
dial biopsy.
ceptors in the cardiac tissue, and decreased adiponectin
mRNA and protein regulation. In the study, adiponectin
injected into neonatal rats in vitro was shown to led to
activation of pro-survival kinase and nitric oxide
synthase, thereby preventing stress-induced apoptosis
of cardiomyocytes.
suggested to be useful in the treatment of dilated cardi-
omyopathy in the future.
[22]The authors found intact adiponectin re-
[22]In this study, adiponectin was
Decreased left ventricular EF is well known as one of
the markers of poor prognosis of CAD.
we found significant difference between the groups in
terms of left ventricular EF and the EF decreased as the
functional capacity worsened. In addition, a significantly
negative relation was found between serum adiponectin
level and left ventricular EF (r=0.380, P=0.0001). On the
other hand, George et al.
relationship between serum adiponectin level and left
ventricular EF, but however found a significant associ-
ation with the functional capacity (NYHA) and serum
adiponectin level, similar to our study. Analyses perfor-
med to identify the possible relation between serum adi-
ponectin level and left ventricular EF among the patient
groups including those with functional capacity (NYHA
classes) and controls found no significant difference in
the serum adiponectin level only in patients with NYHA
II class and controls, whereas serum adiponectin levels
significantly increased in the other patient groups com-
pared to the controls. However, we observed a signifi-
cantly decreased left ventricular EF among all patient
groups compared to the controls. In conclusion, it can be
suggested that there is no significant difference between
the HF patients with functional capacity of NYHA II and
the controls in terms of serum adiponectin level, where-
as serum adiponectin levels markedly increased in HF
patients with NYHA III and IV.
[23]In our study,
[11]did not find any significant
A recent study showed that serum adiponectin levels
increased in the elderly (>70 years) with a significant in-
crease particularly in patients with HF of non ischemic
origin. The study also found increased serum adiponec-
tin levels to be increased mortality.
mental study on animals models suggested that adipo-
nectin improved systolic dysfunction following myocar-
dial infarction, suppressed cardiac hypertrophy and in-
terstitial fibrosis, and that it protect the myocardium aga-
inst myocyte and capillary loss.
[24]Another experi-
[25]
In conclusion, serum adiponectin levels significantly
increase in patients with HF compared to the controls