Serum adipokine levels in chronic liver diseases: Association of resistin levels with fibrosis severity

This article was downloaded by:[HEAL-Link Consortium]On: 15 May 2008Access Details: [subscription number 772810551]Publisher: Informa HealthcareInforma Ltd Registered in England and Wales Registered Number: 1072954Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Scandinavian Journal ofGastroenterologyPublication details, including instructions for authors and subscription information:http://www.informaworld.com/smpp/title~content=t713690387

Serum adipokine levels in chronic liver diseases:Association of resistin levels with fibrosis severityEmmanuel Tsochatzis a; George V. Papatheodoridis a; Emilia Hadziyannis a;Anastasia Georgiou a; Georgia Kafiri b; Dina G. Tiniakos c; Emanuel K. Manesis a;Athanasios J. Archimandritis aa Second Department of Internal Medicine, Athens University Medical School,Hippokration General Hospital, Athens, Greeceb Department of Pathology, Hippokration General Hospital, Athens, Greecec Laboratory of Histology & Embryology, Athens University Medical School, Greece

First Published on: 30 April 2008

To cite this Article: Tsochatzis, Emmanuel, Papatheodoridis, George V., Hadziyannis, Emilia, Georgiou, Anastasia,Kafiri, Georgia, Tiniakos, Dina G., Manesis, Emanuel K. and Archimandritis, Athanasios J. (2008) 'Serum adipokinelevels in chronic liver diseases: Association of resistin levels with fibrosis severity', Scandinavian Journal ofGastroenterology,

To link to this article: DOI: 10.1080/00365520802085387URL: http://dx.doi.org/10.1080/00365520802085387

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

This article maybe used for research, teaching and private study purposes. Any substantial or systematic reproduction,re-distribution, re-selling, loan or sub-licensing, systematic supply or distribution in any form to anyone is expresslyforbidden.

The publisher does not give any warranty express or implied or make any representation that the contents will becomplete or accurate or up to date. The accuracy of any instructions, formulae and drug doses should beindependently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings,demand or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with orarising out of the use of this material.

Dow

nloa

ded

By:

[HE

AL-

Link

Con

sorti

um] A

t: 17

:37

15 M

ay 2

008

ORIGINAL ARTICLE

Serum adipokine levels in chronic liver diseases: Association of resistinlevels with fibrosis severity

EMMANUEL TSOCHATZIS1, GEORGE V. PAPATHEODORIDIS1,

EMILIA HADZIYANNIS1, ANASTASIA GEORGIOU1, GEORGIA KAFIRI2,

DINA G. TINIAKOS3, EMANUEL K. MANESIS1 & ATHANASIOS J. ARCHIMANDRITIS1

1Second Department of Internal Medicine, Athens University Medical School, Hippokration General Hospital, Athens,

Greece, 2Department of Pathology, Hippokration General Hospital, Athens, Greece, and 3Laboratory of Histology &

Embryology, Athens University Medical School, Greece

AbstractObjective. Leptin and adiponectin have been implicated in the pathogenesis and progression of non-alcoholicsteatohepatitis (NASH) and chronic hepatitis C (CHC), but little is known about the role of resistin in chronic liverdiseases. The objective of this study was to investigate serum levels of the above three adipokines in relation to the etiology ofliver disease and to determine their associations with histological severity. Material and methods. We prospectivelyevaluated 146 patients (HBeAg-negative chronic hepatitis B (CHB): 52, CHC: 70, NASH: 24) who consecutivelyunderwent liver biopsy. Detailed epidemiological, anthropometric and laboratory data were recorded. Histological lesionswere evaluated blindly according to the Ishak and the Brunt classifications for CHB/CHC and NASH, respectively.Results. Serum adipokine levels were similar between CHB and CHC patients, while CHB/CHC patients had significantlylower leptin levels compared with NASH patients (8.397.3 versus 17.6916.6 ng/ml, p�0.012) and higher adiponectin(10.295.1 versus 7.594 mg/ml, p�0.018) and resistin levels (7.192.5 versus 5.792.8 ng/ml, p�0.016). In CHB/CHC,there was no significant association between steatosis or necroinflammation and levels of adipokines, while the presence ofmoderate/severe fibrosis (stages 4�6) was associated with higher leptin and adiponectin levels in male but not in femalepatients and with lower resistin levels irrespective of gender or other factors (adjusted odds ratio�0.788, p�0.035).Conclusions. Serum adipokine levels depend on the etiology of liver disease differing between chronic viral hepatitis andNASH, but not between CHB and CHC. In CHB/CHC, resistin levels are independently associated with fibrosis severity,whereas in the association of leptin and adiponectin levels with fibrosis, it seems to be a gender effect.

Key Words: Adipokines, adiponectin, chronic hepatitis B, chronic hepatitis C, leptin, non-alcoholic steatohepatitis, resistin

Introduction

In recent years, adipose tissue has gained attention

as a metabolically active tissue that secretes bioactive

proteins, called adipokines. The adipokines, which

include leptin, adiponectin and resistin, participate

in energy homeostasis and the inflammatory re-

sponse. Leptin is an anorexigenic hormone and is

increased in obesity as a consequence of leptin

resistance [1]. It has pro-inflammatory actions and

is involved in fibrogenesis in various in vitro and

animal models [2]. Adiponectin is an anti-inflam-

matory cytokine that reduces body fat, ameliorates

insulin resistance (IR) and has hepatoprotective

actions [2]. Finally, resistin primarily causes hepatic

IR in rodents but its role in human disease is still

unclear [3].

Hepatic steatosis is a hallmark of non-alcoholic

fatty liver disease (NAFLD), but its role seems to be

crucial as a co-factor of disease progression in other

liver diseases, most importantly in chronic hepatitis

C (CHC), whereas it does not seem to influence

fibrosis severity in chronic hepatitis B (CHB) [4�6].

Since adipokines regulate lipogenesis, lipolysis and

fat distribution, they were initially studied in NAFLD

and suggested to participate in its pathogenesis.

Correspondence: George V. Papatheodoridis, MD, Second Department of Internal Medicine, Athens University Medical School, Hippokration General

Hospital of Athens, 114 Vas. Sophias Ave., GR-115 27 Athens, Greece. Tel: �30 210 7774 742. Fax: �30 210 7706 871. E-mail: [email protected]

Scandinavian Journal of Gastroenterology

2008, 1�9, iFirst article

(Received 10 December 2007; accepted 26 March 2008)

ISSN 0036-5521 print/ISSN 1502-7708 online # 2008 Taylor & Francis

DOI: 10.1080/00365520802085387

Dow

nloa

ded

By:

[HE

AL-

Link

Con

sorti

um] A

t: 17

:37

15 M

ay 2

008

When their involvement in inflammatory and fibro-

genic response became evident, leptin and adiponec-

tin were studied as markers of fibrosis and disease

severity mainly in CHC, but also in cholestasis and

cirrhosis [2]. Results, however, were contradictory,

possibly reflecting the heterogeneity of study popula-

tions or small-size samples. Resistin has not been

adequately studied in chronic liver diseases to date

[7,8].

In this study, we evaluated the metabolic and

adipokine profiles in patients with CHB, CHC and

non-alcoholic steatohepatitis (NASH). In particular,

our aims were a) to study the serum adipokine levels

in relation to the etiology of liver disease and b) to

clarify the associations of leptin and adiponectin

levels and to investigate the potential associations of

resistin levels with liver histology and with para-

meters of IR and metabolic syndrome in a well-

defined cohort of patients with chronic liver disease.

Material and methods

Patient population

We prospectively included 146 treatment-naı̈ve pa-

tients who consecutively underwent liver biopsy at

our Department between August 2004 and August

2006. Fifty-two patients had HBeAg-negative CHB,

70 CHC and 24 had NASH. Patients with HBeAg-

positive CHB, NAFLD and histological findings of

simple fatty liver, any type of therapy for their liver

disease, alcohol abuse, concomitant chronic HBV

and HCV infection, antibodies against hepatitis D

virus (anti-HDV), antibodies against human immu-

nodeficiency virus (anti-HIV), evidence of autoim-

mune liver disease or an inadequate liver biopsy were

excluded. None of the patients had decompensated

liver disease (evidence or history of ascites, variceal

bleeding, hepatic encephalopathy or jaundice from

hepatic failure) or evidence of hepatocellular carci-

noma (HCC).

The local ethics committee approved the study

protocol and all patients gave informed consent.

HBeAg-negative CHB was diagnosed in patients

with positive hepatitis B surface antigen (HBsAg),

negative hepatitis B e antigen (HBeAg) and positive

for its antibody (anti-HBe) for at least 6 months,

increased alanine aminotransferase (ALAT) on at

least two monthly occasions, detectable serum

hepatitis B virus (HBV) DNA, histological lesions

compatible with chronic viral hepatitis (CVH) and

absence of any other possible cause of liver injury.

CHC was diagnosed in patients with positive anti-

bodies against HCV (anti-HCV), detectable serum

hepatitis C virus (HCV) RNA and histological

lesions compatible with CVH. NASH was diagnosed

in patients with NAFLD and histological lesions

compatible with NASH. The diagnosis of NAFLD

was made in patients with increased ALAT activity

(�40 IU/l) on ]3 separate monthly determinations

within the last 6 months and evidence of hepatic

steatosis on liver biopsy provided that any other

possible cause of liver injury including alcohol abuse

had been excluded.

Epidemiological data were obtained from all

patients. Alcohol consumption during the past 5

years was taken into account and alcohol abuse was

excluded by careful questioning conducted by two

independent physicians, including interrogation of

family members, where possible. Alcohol abuse was

considered as a mean daily alcohol consumption of

�30 g for males and �20 g for females for CVH and

�20 g/day regardless of gender for NAFLD patients.

The presence of diabetes mellitus was also recorded

in known diabetics or patients with fasting glucose

�126 mg/dl on more than one occasion. Weight and

height were measured and body mass index (BMI)

(kg/m2) was calculated on the liver biopsy day. Waist

circumference was measured at the midpoint be-

tween the lower border of the rib cage and the iliac

crest, whereas hip circumference was measured at

the widest point between hip and buttock.

Metabolic syndrome was diagnosed in the pre-

sence of 3 or more out of 5 criteria, as defined by the

Adult Treatment Panel III [9], namely: 1) fasting

glucose ]110 mg/dl, 2) central adiposity, defined as

waist circumference �102 cm in men or �88 cm in

women, 3) hypertension, defined as systolic blood

pressure ]135 mmHg, diastolic blood pressure ]

85 mmHg or antihypertensive treatment 4) serum

triglycerides �150 mg/dl and 5) serum high-density

lipoprotein (HDL) cholesterol B40 mg/dl in men or

B50 mg/dl in women.

Laboratory methods

Blood chemistry values including complete blood

count, prothrombin time and levels of serum glu-

cose, cholesterol, HDL and low-density lipoprotein

(LDL), triglycerides and liver enzymes (ALAT,

aspartate aminotransferase (ASAT), alkaline phos-

phatase (ALP), gamma-glutamyl transpeptidase

(GGT)) were determined using commercially avail-

able assays on the liver biopsy day. Insulin levels were

also measured with the INS-EASIA kit (BioSource

Europe S.A., Nivelles, Belgium). IR was determined

by the homeostasis model assessment (HOMA)

method using the following equation: Insulin resis-

tance (HOMA)�Fasting insulin (mU/ml)�fasting

glucose (mmol/l)/22.5 [10]; IR was considered to be

present in cases with HOMA ]3.0.

2 E. Tsochatzis et al.

Dow

nloa

ded

By:

[HE

AL-

Link

Con

sorti

um] A

t: 17

:37

15 M

ay 2

008

Commercially available enzyme immunoassays

were used for detection of HBsAg, HBeAg, anti-

HBe, anti-HCV, anti-HDV and anti-HIV. In patients

with CHB, serum HBV DNA levels were deter-

mined with a commercially available quantitative

polymerase chain reaction (PCR) assay (Amplicor

HBV Monitor; Roche Molecular Diagnostics Sys-

tems, Branchburg, N.J., USA). In patients with

CHC, serum HCV RNA was detected using a

commercially available qualitative PCR assay (Am-

plicor; Roche Molecular Diagnostic Systems, sensi-

tivity: 30 IU/ml) and HCV genotype was determined

with a commercially available assay (HCV Genotype

Assay; LIPA, Versant, Bayer Healthcare, Tarrytown,

N.Y., USA).

All serum adipokine levels were determined in

stored serum samples using immunoenzymometric

assays. Serum samples were obtained after overnight

fasting on the liver biopsy day and were stored in

aliquots at �608C until assayed. Leptin was mea-

sured with the LEPTIN EASIA kit (BioSource

Europe S.A.), adiponectin with the Adiponectin

Elisa kit (Asbach Medical Products (AMP), Obrigh-

eim, Germany) and resistin with the Human Resistin

ELISA kit (Biovendor, Heidelberg, Germany).

Liver histology

All 146 liver biopsy specimens were of adequate

length (]1.5 cm). All biopsies from CVH patients

were evaluated blindly by one liver pathologist

(G.K.) in accordance with the classification system

proposed by Ishak et al. [11]. All biopsies from

NAFLD patients were evaluated blindly by another

liver pathologist (D.G.T.) according to the classifica-

tion system proposed by Brunt et al. [12]. NASH

was considered to be present in NAFLD cases with

grade score ]2 and/or fibrosis score ]1. In both

CVH and NASH, hepatic steatosis was assessed as

the percentage of hepatocytes containing fat droplets

and it was graded as 0 (B5% steatosis), 1 (mild, 5�33% of hepatocytes affected), 2 (moderate, 33�66%

of hepatocytes affected) and 3 (severe, �66% of

hepatocytes affected).

Statistical analysis

All data were analyzed using the statistical package

SPSS (version 14.0; SPSS Inc., Chicago, Ill., USA).

Statistical analysis was performed using the t-test,

ANOVA, the Mann-Whitney test or the Kruskal-

Wallis test for comparisons of continuous variables

between or among groups, the corrected x2 method

or the Fisher two-tailed exact test for comparisons of

qualitative data and Spearman’s coefficient for

correlations of quantitative data, when appropriate.

A two-tailed p-value of less than 0.05 was considered

to be statistically significant. In the univariate

analysis including multiple (K) comparisons, a

p-value of 0.05/K was considered to be statistically

significant. The multivariate analysis was performed

using backward logistic regression models. Only

variables with a p-value 50.10 at univariate analysis

were entered in the multivariate analysis models.

Results

Baseline characteristics

The patients’ baseline characteristics are presented

in Table I. The patients were middle-aged and

overweight, with a slight predominance of males.

Males and females were comparable in terms of age,

BMI, steatosis, necroinflammation grade and fibro-

sis stage. NASH patients had significantly higher

BMI values, higher values of triglycerides, more

severe steatosis and a more frequent presence of

glucose intolerance, central adiposity and metabolic

syndrome compared with patients with CVH. In-

sulin resistance was present in 44% and 46% of

patients with CVH and NASH, respectively.

Associations of serum adipokines with demographic and

metabolic parameters

The associations between each adipokine level and

all patients’ characteristics are summarized in

Table II. Higher serum leptin levels were signifi-

cantly associated with increasing age (r�0.173, p�0.029), increasing BMI (r�0.417, pB0.001), fe-

male gender (PB0.001), presence of IR (p�0.012),

central adiposity (pB0.001), diabetes mellitus

(p�0.025), presence of metabolic syndrome (pB

0.001), increasing serum adiponectin (r�0.263,

p�0.002) and decreasing serum resistin levels

(r��0.164, p�0.049). Higher serum adiponectin

levels were significantly associated with increasing

age (r�0.298, pB0.001), female gender (pB0.001)

and increasing leptin levels. Finally, higher serum

resistin levels significantly correlated only with

female gender (p�0.001) and decreasing leptin

levels. No association was noted between any

adipokine and viral load in CHB or CHC patients.

Serum adipokines and etiology of chronic liver disease

Serum adipokine levels did not significantly differ

between patients with CHB and CHC (Table III).

Patients with CVH had significantly lower leptin

levels (8.397.3 versus 17.6916.6 ng/ml, p�0.012)

and higher adiponectin (10.295.1 versus 7.594.0

mg/ml, p�0.018) and resistin levels (7.192.5 versus

5.792.8 ng/ml, p�0.016) compared with patients

Adipokines in chronic liver diseases 3

Dow

nloa

ded

By:

[HE

AL-

Link

Con

sorti

um] A

t: 17

:37

15 M

ay 2

008

with NASH. Leptin, adiponectin and resistin levels

in patients with CHC, CHB and NASH are depicted

in Figure 1. Among patients with CHC, 11 had

infection with genotype 3 and 59 with genotype non-

3. Serum leptin levels were significantly higher and

serum resistin significantly lower in genotype non-3

compared with genotype 3 CHC patients, while

no difference was noted for adiponectin levels

(Figure 2). In CHB patients without steatosis

(n�16), who may be considered as cases with

chronic liver disease unrelated to disturbances in

fat metabolism, serum adiponectin levels were

higher than those in the remaining CVH patients

(12.693.8 versus 9.995.2, p�0.018), while no

significant difference was noted for leptin and

resistin levels. Insulin resistance did not significantly

differ among patients with CHB, CHC and NASH.

All the above associations did not significantly

change after adjustment for age, gender and BMI.

Serum adipokines and liver histology

As serum adipokine levels did not significantly differ

between CHB and CHC patients, we addressed

their potential associations with liver histology in the

whole cohort of 122 CVH patients (Table IV).

Severe fibrosis (stages 4�6) was detected in 33

(27%) of the CVH patients. In the univariate

analysis, severe fibrosis was associated with advanced

age, increased BMI, higher necroinflammation

grade, higher HOMA index, increased serum

ASAT, ALAT, ALP and GGT levels and decreased

serum resistin levels, but not with leptin or adipo-

nectin levels. In the multivariate analysis, lower

serum resistin was, among others, an independent

predictor of severe fibrosis (OR: 0.788, 95% CI:

0.631�0.984, p�0.035) (Table IV). No significant

association between steatosis or necroinflammation

and serum adipokine levels was noted (data not

shown). Similar findings were observed when an

analysis was done separately for CHB and CHC or

for genotype 3 and non-3 CHC or when patients

with CHC and diabetes mellitus were excluded (data

not shown).

When males and females were analyzed separately,

some additional associations between adipokine

levels and histological parameters were found. In

particular, leptin levels were higher only in males

with severe fibrosis than in those without severe

fibrosis (7.495.1 versus 4.593.5 ng/ml, p�0.013)

but not in females. Similarly, serum adiponectin

levels correlated with the severity of necroinflamma-

tion only in males (r�0.29, p�0.016) and were

higher only in male patients with than in those

without severe fibrosis (11.195.0 versus 8.194.3

mg/ml, p�0.021). In contrast, serum resistin levels

Table I. Baseline characteristics of 146 patients with chronic liver disease, chronic viral hepatitis (CVH): 122, non-alcoholic steatohepatitis

(NASH): [24].

Patient characteristics CHB (n�52) CHC (n�70) CVH (n�122) NASH (n�24) p-value1

Gender, males (%) 36 (69) 36 (51) 72 (59) 13 (54) NS

Age, years 41.5912.9 44.5911.9 43.2912.4 45.2916.5 NS

BMI, kg/m2 26.593.9 25.393.6 25.893.8 29.294.5 0.002

ALAT, IU/l 1089129 95964 101997 74956 NS

ASAT, IU/l 65990 56933 60964 44924 NS

ALP, U/l 94964 84943 88952 101949 0.057

GGT, U/l 57961 54944 55951 999107 0.061

Cholesterol, mg/dl 194946 179944 185945 202942 0.085

HDL, mg/dl 48918 46915 47916 46912 NS

LDL, mg/dl 126945 114934 119939 134943 NS

Triglycerides, mg/dl 90943 88963 89955 118942 0.006

HOMA-IR 5.598.7 3.492.5 4.396 492.9 NS

Insulin resistance, n (%) 26 (50) 27 (39) 53 (44) 11 (46) NS

Central adiposity, n (%) 9 (17) 15 (21) 29 (24) 13 (54) 0.005

Fasting glucose ]110 mg/dl, n (%) 1 (2) 6 (9) 7 (6) 6 (25) 0.008

Metabolic syndrome, n (%) 2 (4) 5 (7) 7 (6) 9 (37) B0.001

Necroinflammation* 6.192.7 6.592 6.492.3 2.090.7 �Fibrosis* 2.991.8 2.991.4 2.991.5 2.391.1 �Steatosis, n (%) 36 (71) 56 (80) 92 (76) 24 (100) 0.004

Severe steatosis, n (%) 8 (16) 19 (27) 27 (22) 15 (62) B0.001

Abbreviations: CHB�HBeAg-negative chronic hepatitis B; CHC�chronic hepatitis C; BMI�body mass index; ALAT�alanine

aminotransferase; ASAT�aspartate aminotransferase; ALP�alkaline phosphatase; GGT�gamma-glutamyl transpeptidase; HDL�high-density lipoprotein; LDL�low-density lipoprotein; HOMA-IR�homeostasis model assessment for insulin resistance; NS�non-

significant.

*Necroinflammation and fibrosis were evaluated using the Ishak classification system [11] in CVH patients and the Brunt classification

system [12] in NASH patients; 1p-values for comparisons between CVH and NASH patients.

4 E. Tsochatzis et al.

Dow

nloa

ded

By:

[HE

AL-

Link

Con

sorti

um] A

t: 17

:37

15 M

ay 2

008

were lower in patients with than in those without

severe fibrosis, irrespective of gender. In the multi-

variate analysis including only male patients, in-

creased serum leptin levels but not adiponectin levels

were independently associated with severe fibrosis

(OR: 1.231, 95%CI 1.029�1.473, p�0.023) No

Table II. Associations of serum adipokine levels with demographic and metabolic parameters in 146 patients with chronic liver disease

(chronic viral hepatitis: 122; non-alcoholic steatohepatitis: 24).

Patient characteristic

Serum leptin

(ng/ml) p-value

Serum adiponectin

(mg/ml) p-value

Serum resistin

(ng/ml) p-value

Gender B0.001 B0.001 0.001

Males 5.894.8 8.895.2 8.895.2

Females 15.7914.2 11.895.2 11.895.2

Age, years 0.002 0.05 NS

]50 13912.3 10.894.7 6.992.7

B50 8.197.9 9.195.2 6.992.5

BMI, kg/m2 0.001 NS NS

]25 11.8911.3 9.595.3 6.892.5

B25 6.996.5 1094.6 7.192.7

Insulin resistance 0.012 NS NS

HOMA ]3 12.6912.2 9.394.4 6.692.5

HOMA B3 8.297.7 10.595.6 6.792.7

Central adiposity B0.001 NS NS

Yes 16.4911.9 10.195.5 7.292.5

No 7.698.1 10.195.4 7.192.1

Hypertension 0.031 NS NS

Yes 16.5915.3 8.994.4 6.292.8

No 8.998.6 10.295.6 6.892.6

Fasting glucose ]110 mg/dl 0.025 NS NS

Yes 18.3918 9.394.3 6.493.4

No 9.298.6 10.195.5 6.792.6

Triglycerides 0.04 NS NS

�150 mg/dl 15.4914.2 10.194.3 5.992.3

5150 mg/dl 9.399.4 9.795.3 6.992.5

HDL NS NS NS

]40/50 mg/dl M/F 11.4911.7 9.394.5 7.192.4

B40/50 mg/dl M/F 7.897.2 9.995.7 6.692.5

Metabolic syndrome B0.001 NS NS

Yes 23.3916.7 10.194.3 5.892.7

No 8.297.4 10.195.6 6.892.6

Steatosis NS NS NS

Yes 8.697.9 9.894.8 7.292.5

No 7.495.5 11.295.7 7.192.1

Severe steatosis NS NS 0.033

Yes 12.5912.4 9.194.6 6.292.3

No 8.898.7 9.995.1 7.292.6

Abbreviations: BMI�body mass index; HOMA�homeostasis model assessment for insulin resistance; NS�non-significant.

Table III. Baseline serum adipokine levels of 146 patients with chronic liver disease (chronic viral hepatitis (CVH): 122; non-alcoholic

steatohepatitis (NASH): 24).

Serum adipokine CHB (n�52) CHC (n�70) CVH (n�122) NASH (n�24) p-value1

Leptin, ng/ml 8.696.7 8.197.8 8.397.3 17.6916.6 0.012

Adiponectin, mg/ml 10.294.4 10.295.6 10.295.1 7.594 0.018

Resistin, ng/ml 6.892.6 7.392.3 7.192.5 5.792.2 0.016

Abbreviations: CHB�chronic hepatitis B; CHC�chronic hepatitis C; NS�non-significant.1p-values are for comparisons between CVH and NASH patients.

Adipokines in chronic liver diseases 5

Dow

nloa

ded

By:

[HE

AL-

Link

Con

sorti

um] A

t: 17

:37

15 M

ay 2

008

association was noted between leptin or adiponectin

and presence or severity of steatosis, irrespective of

gender.

In patients with NASH, severe fibrosis (stage 3�4)

was present in 5 (21%) patients and was associated

with older age (p�0.006), lower serum resistin (p�0.011), higher serum leptin (p�0.012) and ASAT

levels (p�0.015), increasing IR (p�0.036) and the

presence of metabolic syndrome (p�0.003). More-

over, moderate/severe steatosis (grades 2�3) was

associated with lower resistin levels (7.292.6 versus

6.292.3, p�0.033) but with neither leptin nor

adiponectin levels or with IR. Since multiple com-

parisons were made, a p-value of less than 0.005 was

considered to be statistically significant.

Discussion

Serum adipokines and their potential associations

with the pathogenesis and severity of liver diseases

have been a focus of research in recent years [2].

Leptin and adiponectin have been studied in CHC

and NAFLD but with contradictory results, whereas

there have been few studies on resistin [13�16].

Moreover, data on the potential impact of adipo-

kines in patients with CHB suggest a possible role of

adiponectin in fibrosis progression [17,18]. For

these reasons, we studied all three adipokines in a

fairly large cohort that included patients with CHB,

CHC and NASH. In order to preclude potential

hidden confounding factors, which have partly con-

tributed to the contradictory results of previous

studies, we assessed in detail various metabolic,

anthropometric, demographic and epidemiologic

factors, while excluding alcohol abusers.

The main finding of our study was that serum

adipokines differ significantly among patients with

NASH and CHB or CHC. In previous studies leptin

and adiponectin levels were compared between

healthy controls and patients with NAFLD or

CHC [13,14,19�22], but the studies did not provide

a direct comparison between NAFLD and CHC

patients. Moreover, resistin has not been evaluated

in relation to the etiology of liver disease. The

different levels of adipokines in NAFLD probably

reflect their direct implication in the pathogenesis of

this disease. NAFLD may reflect the failure of leptin

Figure 1. Serum levels of adipokines (A. leptin; B. adiponectin;

C. resistin) in 70 patients with chronic hepatitis C, 52 patients

with HBeAg-negative chronic hepatitis B and 24 patients with

non-alcoholic steatohepatitis. Box-and-whiskers plots express

medians, and interquartile and overall ranges; the outlying values

are plotted individually. Abbreviations: CHC�chronic hepatitis

C; CHB�chronic hepatitis B; NASH�non-alcoholic steatohe-

patitis. Figure 1 (Continued)

6 E. Tsochatzis et al.

Dow

nloa

ded

By:

[HE

AL-

Link

Con

sorti

um] A

t: 17

:37

15 M

ay 2

008

to prevent ectopic lipid accumulation, referred to as

lipotoxicity, and is thus a leptin-resistant state [14].

Furthermore, the hepatoprotective adiponectin that

improves insulin sensitivity and regulates triglyceride

metabolism seems to be down-regulated and its

effects thus negated [20]. Although resistin has

been associated with IR, no experimental data exist

on its potential association with the pathogenesis of

liver disease. In a recent study of 80 Asian patients

with NAFLD and 48 controls, serum resistin levels

did not differ between NAFLD and controls or

NAFLD and NASH, although unmatched controls

were used and adjustments were not made [7]. Our

findings suggest that resistin may participate in the

pathogenesis of NAFLD, as its levels were found to

correlate with both steatosis and fibrosis severity in

our NASH patients. It is not clear whether the

adipokine deregulation is part of the ‘‘first’’ or

‘‘second’’ hit in NASH development [23], as few

studies have directly compared controls, patients

with simple steatosis and patients with NASH [20].

Adipokine profiles and all their associations did

not differ significantly between patients with CHB

and CHC. Of note is the difference observed in

serum leptin and resistin levels between CHC

patients with genotype 3 and non-3. Genotype-3

CHC has unique virological features, as it is

steatogenic per se and does not cause IR, in sharp

contrast to genotype non-3 [4,6,24,25]. We and

others have implied that patients with genotype 3

have different adipokine profiles and this is probably

attributed to the absence of IR [2,26]. Our findings,

although restricted in a small group of genotype-3

patients, further support this hypothesis.

Another finding of our study is a possible gender

effect in the associations between serum adipokines

and liver histology. Interestingly, we found positive

correlations between serum leptin and adiponectin

levels and fibrosis only in male patients with CHB or

CHC. On the one hand, this finding cannot be

directly attributed to confounding factors, as male

and female patients did not differ significantly in any

of the metabolic and histological parameters. On the

other hand, gender has previously been found to

affect the serum levels of adipokines. In particular, a

gender effect in the association between leptin [27]

and adiponectin levels [15] with the severity of

steatosis has been demonstrated in CHC patients.

This observation, although recorded, was not further

analyzed. Estrogen effects or gender-mediated adi-

pokine-receptor activities might be a possible ex-

planation of this finding, which needs further study

and verification.

Serum resistin levels were found to be indepen-

dently associated with fibrosis severity in patients

with CHB or CHC. They were also associated with

the severity of steatosis and fibrosis in NASH, but

only a small number of these patients were included

in the multivariate analysis, and multiple compar-

isons errors might have affected our results from

Figure 2. Serum levels of adipokines (A. leptin; B. adiponectin;

C. resistin) in relation to the hepatitis C virus (HCV) genotype.

Box- and-whiskers plots express medians, and interquartile and

overall ranges; the outlying values are plotted individually.

Adipokines in chronic liver diseases 7

Dow

nloa

ded

By:

[HE

AL-

Link

Con

sorti

um] A

t: 17

:37

15 M

ay 2

008

univariate analyses. Therefore, evaluation of larger

cohorts of NAFLD patients is required to verify our

findings. A recent study in cirrhotic patients showed

correlation of serum resistin levels with Child-Pugh

score and survival irrespective of the etiology of

cirrhosis [8], while two small studies on CHC and

NAFLD failed to show any association between

serum resistin and liver histology [7,27]. However,

the study in CHC included fewer patients and used a

different scoring system, while the study in NAFLD

included exclusively patients of Chinese ethnicity

who usually have normal or low BMI values. In

contrast to leptin and adiponectin, there is little

experimental background to support a pathogenetic

role of resistin in liver fibrogenesis [28]. Therefore, it

is not yet clear whether lower levels of resistin in

advanced fibrosis represent a marker of disease

severity or whether resistin is directly implicated in

disease progression. Of note is that, although resistin

has been termed a pro-inflammatory cytokine, its

levels are lower in severe fibrosis. This could reflect a

down-regulation by anti-inflammatory cytokines or a

direct anti-inflammatory effect in liver tissue. Similar

paradoxical effects have been noted in adiponectin

levels in cirrhotic patients and have been attributed

to an anti-inflammatory response to liver injury [29].

Finally, we found no independent association of

adiponectin with liver histology in CHC or CHB,

whereas an independent association was found for

leptin, but only in males. Our findings are not

surprising, since most studies failed to show inde-

pendent associations of adiponectin in CHC

[15,16,26,29,30], while few data exist for CHB

[17,18]. The gender effect on the association of

leptin and fibrosis might explain the contradictory

results of previous trials [22,31�33]. Our findings in

CHB patients support the previously analyzed simi-

lar kinetics of serum adipokines in CHB and CHC.

In conclusion, we have shown that the levels of

serum adipokines depend on the etiology of liver

disease and differ significantly between patients with

NASH and CVH. Their serum levels and associa-

tions are similar in CHB and CHC, where there

seems to be a gender effect in the associations of

leptin and adiponectin with liver histology. Most

importantly, serum resistin, a rarely studied adipo-

kine, appears to be independently associated with

the severity of fibrosis in CVH. Therefore, the

potential pathogenetic implication of resistin should

be further studied.

Declaration of interest: The authors report no

conflicts of interest. The authors alone are respon-

sible for the content and writing of the paper.

References

[1] Mantzoros CS. The role of leptin in human obesity and

disease: a review of current evidence. Ann Intern Med

1999;8:671�80.

[2] Tsochatzis E, Papatheodoridis GV, Archimandritis AJ. The

evolving role of leptin and adiponectin in chronic liver

diseases. Am J Gastroenterol 2006;11:2629�40.

Table IV. Univariate and multivariate analysis of factors associated with moderate/severe fibrosis in 122 patients with chronic viral hepatitis

(CVH).

Patient characteristic

Fibrosis 0�3(n�89)

Fibrosis 4�6(n�33)

p-value

Univariate

Adjusted odds ratio

(95% CI)

p-value

Multivariate

Gender, males (%) 49 (55) 23 (70) NS

Age, years 41912 50913 0.001 1.046 (1.001�1.094) 0.047

BMI, kg/m2 25.193.8 27.793 B0.001 NS

ALAT, IU/l 80963 1569141 B0.001 NS

ASAT, IU/l 44922 1009101 B0.001 NS

ALP, U/l 78943 111965 0.003 NS

GGT, U/l 45943 79963 B0.001 NS

Cholesterol, mg/dl 186946 183943 NS

Triglycerides, mg/dL 87989 92941 NS

Leptin, ng/ml 8.197.7 8.996.5 NS

Adiponectin, mg/ml 9.895 11.395.2 NS

Resistin, ng/ml 7.492.3 6.392.6 0.044 0.788 (0.631�0.984) 0.035

HOMA-IR 3.593.9 6.399.5 0.001 NS

Insulin resistance, n (%) 32 (37) 21 (64) 0.013

Central adiposity, n (%) 15 (17) 12 (42) 0.007 8.918 (2.268�35.068) 0.002

Fasting glucose ]110 mg/dl, n (%) 5 (6) 2 (6) NS

Necroinflammation 5.792 8.292 B0.001 2.064 (1.474�2.888) B0.001

Steatosis, n (%) 67 (75) 25 (78) NS

Severe steatosis, n (%) 17 (19) 10 (31) NS

Abbreviations: BMI�body mass index; ALAT�alanine aminotransferase; ASAT�aspartate aminotransferase; ALP�alkaline phospha-

tase; GGT�gamma-glutamyl transpeptidase; HOMA-IR�homeostasis model assessment for insulin resistance; NS�non-significant.

8 E. Tsochatzis et al.

Dow

nloa

ded

By:

[HE

AL-

Link

Con

sorti

um] A

t: 17

:37

15 M

ay 2

008

[3] McTernan PG, Kusminski CM, Kumar S. Resistin. Curr

Opin Lipidol 2006;2:170�5.

[4] Bugianesi E, McCullough AJ, Marchesini G. Insulin resis-

tance: a metabolic pathway to chronic liver disease. Hepa-

tology 2005;5:987�1000.

[5] Tsochatzis E, Papatheodoridis GV, Manesis EK, Chry-

santhos N, Kafiri G, Archimandritis AJ. Hepatic steatosis

in chronic hepatitis B develops due to host metabolic factors:

a comparative approach with genotype 1 chronic hepatitis C.

Digest Liver Dis 2007;39:936�42.

[6] Tsochatzis E, Papatheodoridis GV, Manesis EK, Chry-

santhos N, Kafiri G, Petraki K, et al. Hepatic steatosis in

genotype 4 chronic hepatitis C is mainly because of meta-

bolic factors. Am J Gastroenterol 2007;3:634�41.

[7] Wong VW, Hui AY, Tsang SW, Chan JL, Tse AM, Chan KF,

et al. Metabolic and adipokine profile of Chinese patients

with nonalcoholic fatty liver disease. Clin Gastroenterol

Hepatol 2006;4:1154�61.

[8] Yagmur E, Trautwein C, Gressner AM, Tacke F. Resistin

serum levels are associated with insulin resistance, disease

severity, clinical complications, and prognosis in patients

with chronic liver diseases. Am J Gastroenterol 2006;6:1244�52.

[9] Executive Summary of the Third Report of The National

Cholesterol Education Program (NCEP) Expert Panel on

Detection, Evaluation, and Treatment of High Blood Cho-

lesterol in Adults (Adult Treatment Panel III). JAMA

2001;19:2486�97.

[10] 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;7:412�9.

[11] Ishak K, Baptista A, Bianchi L, Callea F, De GJ, Gudat F, et

al. Histological grading and staging of chronic hepatitis.

J Hepatol 1995;6:696�9.

[12] Brunt EM, Janney CG, Di Bisceglie AM, Neuschwander-

Tetri BA, Bacon BR. Nonalcoholic steatohepatitis: a propo-

sal for grading and staging the histological lesions. Am J

Gastroenterol 1999;9:2467�74.

[13] Bugianesi E, Pagotto U, Manini R, Vanni E, Gastaldelli A,

de IR, et al. Plasma adiponectin in nonalcoholic fatty liver is

related to hepatic insulin resistance and hepatic fat content,

not to liver disease severity. J Clin Endocrinol Metab 2005;6:

3498�504.

[14] Chitturi S, Farrell G, Frost L, Kriketos A, Lin R, Fung C,

et al. Serum leptin in NASH correlates with hepatic steatosis

but not fibrosis: a manifestation of lipotoxicity? Hepatology

2002;2:403�9.

[15] Jonsson JR, Moschen AR, Hickman IJ, Richardson MM,

Kaser S, Clouston AD, et al. Adiponectin and its receptors in

patients with chronic hepatitis C. J Hepatol 2005;6:929�36.

[16] Liu CJ, Chen PJ, Jeng YM, Huang WL, Yang WS, Lai MY,

et al. Serum adiponectin correlates with viral characteristics

but not histologic features in patients with chronic hepatitis

C. J Hepatol 2005;2:235�42.

[17] Hui C, Zhang H, Lee N, Chan W, Yueng Y, Leung K, et al.

Serum adiponectin is increased in advancing liver fibrosis

and declines with reduction in fibrosis in chronic hepatitis B.

J Hepatol 2007;47:191�202.

[18] Siagris D, Vafiadis G, Michalaki M, Lekkou A, Starakis A,

Makri M, et al. Serum adiponectin in chronic hepatitis C

and B. J Viral Hepat 2007;14:577�83.

[19] Angulo P, Alba LM, Petrovic LM, Adams LA, Lindor KD,

Jensen MD. Leptin, insulin resistance, and liver fibrosis in

human nonalcoholic fatty liver disease. J Hepatol 2004;6:

943�9.

[20] Hui JM, Hodge A, Farrell GC, Kench JG, Kriketos A,

George J. Beyond insulin resistance in NASH: TNF-alpha or

adiponectin? Hepatology 2004;1:46�54.

[21] Crespo J, Rivero M, Fabrega E, Cayon A, Amado JA,

Garcia-Unzeta MT, et al. Plasma leptin and TNF-alpha

levels in chronic hepatitis C patients and their relationship to

hepatic fibrosis. Dig Dis Sci 2002;7:1604�10.

[22] Giannini E, Ceppa P, Botta F, Mastracci L, Romagnoli P,

Comino I, et al. Leptin has no role in determining severity of

steatosis and fibrosis in patients with chronic hepatitis C. Am

J Gastroenterol 2000;11:3211�7.

[23] Day CP, James OF. Steatohepatitis: a tale of two ‘‘hits’’?

Gastroenterology 1998;4:842�5.

[24] Fartoux L, Poujol-Robert A, Guechot J, Wendum D,

Poupon R, Serfaty L. Insulin resistance is a cause of steatosis

and fibrosis progression in chronic hepatitis C. Gut 2005;7:

1003�8.

[25] Muzzi A, Leandro G, Rubbia-Brandt L, James R, Keiser O,

Malinverni R, et al. Insulin resistance is associated with liver

fibrosis in non-diabetic chronic hepatitis C patients. J

Hepatol 2005;1:41�6.

[26] Wang AY, Hickman IJ, Richards AA, Whitehead JP, Prins JB,

Macdonald GA. High molecular weight adiponectin corre-

lates with insulin sensitivity in patients with hepatitis C

genotype 3, but not genotype 1 infection. Am J Gastro-

enterol 2005;12:2717�23.

[27] Lo Iacono O, Venezia G, Petta S, Mineo C, De LS, Di MV,

et al. The impact of insulin resistance, serum adipocytokines

and visceral obesity on steatosis and fibrosis in patients with

chronic hepatitis C. Aliment Pharmacol Ther 2007;10:

1181�91.

[28] Bertolani C, Sancho-Bru P, Failli P, Bataller R, Aleffi S,

DeFranco R, et al. Resistin as an intrahepatic cytokine:

overexpression during chronic injury and induction of

proinflammatory actions in hepatic stellate cells. Am J Pathol

2006;169:2042�53.

[29] Tietge UJ, Boker KH, Manns MP, Bahr MJ. Elevated

circulating adiponectin levels in liver cirrhosis are associated

with reduced liver function and altered hepatic hemody-

namics. Am J Physiol Endocrinol Metab 2004;1:E82�9.

[30] Sato S, Furuta K, Mishiro T, Miyake T, Kohge N, Akagi S,

et al. Serum adiponectin concentration in patients with

hepatitis C virus. J Clin Gastroenterol 2005;8:744�5.

[31] Piche T, Vandenbos F, Bakar-Mahamat A, Vanbiervliet G,

Barjoan EM, Calle G, et al. The severity of liver fibrosis is

associated with high leptin levels in chronic hepatitis C.

J Viral Hepatol 2004;1:91�6.

[32] Romero-Gomez M, Castellano-Megias VM, Grande L, Irles

JA, Cruz M, Nogales MC, et al. Serum leptin levels correlate

with hepatic steatosis in chronic hepatitis C. Am J Gastro-

enterol 2003;5:1135�41.

[33] Manolakopoulos S, Bethanis S, Liapi C, Stripeli F, Sklavos

P, Margeli A, et al. An assessment of serum leptin levels in

patients with chronic viral hepatitis: a prospective study.

BMC Gastroenterol 2007;1:17.

Adipokines in chronic liver diseases 9