Blood Cells, Molecules, and Diseases 43 (2009) 230–234

Contents lists available at ScienceDirect

Blood Cells, Molecules, and Diseases

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / y b c m d

Soluble endothelial adhesion molecules and inflammation markers in patients with

β-thalassemia intermedia
Ino Kanavaki a, Periklis Makrythanasis a, Christina Lazaropoulou a, Maria Tsironi b, Antonis Kattamis c,
Ioannis Rombos d, Ioannis Papassotiriou a,⁎
a
Department of Clinical Biochemistry, “Aghia Sophia” Children's Hospital, 115 27 Athens, Greece
b
Department of Medicine, University of Peloponnesus, School of Nursing, Sparta, Greece
c
First Department of Paediatrics, Athens University Medical School, Athens, Greece
d
First Department of Internal Medicine, Athens University Medical School, Athens, Greece

a r t i c l e i n f o a b s t r a c t

Article history: The term thalassemia intermedia, indicates a clinical condition of intermediate severity between thalassaemia
Submitted 7 May 2009 minor, the asymptomatic carrier, and thalassaemia major, the transfusion-dependent, severe form.
Revised 19 June 2009 Thromboembolic events frequently complicate the outcome of thalassemia intermedia patients, reflecting a
Available online 4 August 2009
hypercoagulable state to which endothelial activation is believed to play an important role. The aim of this
(Communicated by G. Stamatoyannopoulos,
study was to evaluate the levels of soluble endothelial adhesion molecules that reflect endothelial activation
M.D., Dr. Sci., 25 June 2009) and dysfunction and levels of chronic inflammation markers in the serum of β-thalassemia intermedia
patients. Thirty-five Greek patients with β-thalassemia intermedia that have received different types of
Keywords: treatment (Hydroxyurea, splenectomy, untreated), aged 8–63 years, were included in the study. Twenty
Thalassemia intermedia apparently healthy individuals matched for age and sex, formed the control group. Measurements of sVCAM-1,
Endothelium sICAM-1, sTM, P-selectin, E-selectin and CRP levels were performed using immunoassays. We found that all
Adhesion molecules endothelial adhesion molecules and CRP were significantly increased in patients (p b 0.001) and not influenced
Inflammation by treatment. A negative correlation was observed between levels of sICAM-1 and sTM and this finding agrees
with the results of studies, which propose this correlation as a predictive marker of increased risk for vascular
damage. No correlation was observed between endothelial adhesion molecules and inflammation markers.
These findings support the hypothesis that a serious degree of endothelial activation and damage along with a
state of chronic inflammation underlie the pathophysiology of β-thalassemia intermedia. Furthermore, these
findings are of particular importance in patients who can otherwise be characterized by a subtle clinical
phenotype and may have an important role in their clinical care.
© 2009 Elsevier Inc. All rights reserved.

Introduction some patients with a β-thalassemia intermedia genotype are treated

The term thalassemia intermedia is used to define a group of
patients with β thalassemia in whom the clinical severity of the
disease is somewhere between the mild symptoms of the β
thalassemia trait and the severe manifestations of β thalassemia
major. The diagnosis is a clinical one that is based on the patient
maintaining a satisfactory hemoglobin (Hb) level of at least 6–7 g/
dL at the time of diagnosis without the need for regular blood
transfusions. This initial definition of thalassemia intermedia, which
was based on clinical observation alone, retained its validity even
after some of the specific mutations associated with thalassemia
intermedia were recognized because severity of the clinical course
remains unpredictable even in known genotypes. For this reason,
⁎ Corresponding author. Fax: +30 210 746 7171.
E-mail addresses: biochem@paidon-agiasofia.gr, ipapassotiriou@gmail.com
(I. Papassotiriou).
as if they have thalassemia major because they present with severe
manifestations; similarly, others with a thalassemia intermedia
genotype are considered to have thalassemia minor because of the
mild or even asymptomatic nature of their condition. This variability
is most likely related to the presence or absence of modifying genes.
Because of the significant overlap in clinical severity among the 3
types of β thalassemia and despite the fact that several genotypes
are associated with the β thalassemia intermedia picture, the
diagnosis continues to be a clinical one, regardless of the genotype
involved [1]. Complications in thalassemia intermedia result from
chronic hemolytic anemia, ineffective erythropoiesis and iron
overload [2,3]. Therapeutic approaches, such as iron chelation
therapies, hydroxyurea and blood transfusions, have significantly
increased the life expectancy of thalassemic patients, leading to the
identification and description of previously underestimated condi-
tions which complicate the life of such patients, such as throm-
boembolic phenomena [4].

1079-9796/$ – see front matter © 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.bcmd.2009.06.002
 I. Kanavaki et al. / Blood Cells, Molecules, and Diseases 43 (2009) 230–234
Thromboembolic events are frequently described in thalassemia
intermedia patients, clinically manifested as ischemic cerebral lesions,
deep venous thrombosis and pulmonary hypertension [5,6]. Autopsy
findings confirm the presence of a high incidence of thromboembolic
events [4] and Magnetic Resonance Imaging (MRI) studies have
shown brain damage in otherwise asymptomatic thalassemic patients
[7]. The above mentioned clinical, imaging and autopsy evidence
define thalassemia as a hypercoagulable state [5]. Thalassemia
intermedia patients suffer more frequently from thromboembolic
events than patients with thalassemia major [4]. Pathogenesis of
hypercoagulability in thalassemia is based on several factors, such as
erythrocyte cell membrane alterations, platelet and endothelial cells
activation [4].
Therapeutic approaches like splenectomy and transfusions seem to
influence the manifestation of thromboembolic complications. Thus,
splenectomized patients present higher risks for thrombosis than
non-splenectomized patients, due probably to the presence of
increased numbers of damaged erythrocytes as well as high
thrombocyte counts [4,8]. Regularly transfused patients present a
lower incidence of thromboembolic phenomena than transfusion
independent patients [8].
Inflammation is known to have an important role in the
pathogenesis of thalassemia and a chronic inflammatory state is
present in these patients. Pro-inflammatory cytokines such as IL-6 and
inflammation markers such as CRP are known to be elevated in
thalassemic patients [9,10]. Endothelial activation is also believed to
play an important role in the pathophysiology of thalassemia, through
inflammation and thrombosis [4,9,11]. Endothelial activation is
reflected by the increased serum levels of soluble endothelial
adhesion molecules such as inter-cellular adhesion molecule 1
(sICAM-1), vascular cell adhesion molecule 1 (sVCAM-1), E-selectin
(E-sel) and von Willebrand factor (vWF) [12,13]. Increased levels of
soluble thrombomodulin (sTM) have also been demonstrated in α
and β thalassemia patients, indicating endothelial injury [14].
The aim of this study was to evaluate the levels of soluble
endothelial adhesion molecules, that are used as markers of
endothelial activation and dysfunction, in the serum of β-thalassemia
intermedia patients. Levels of CRP as chronic inflammation marker
were also evaluated.
Patients and methods
Thirty-five patients with thalassemia intermedia, 13 men and 22
women, aged 8–63 years, selected from the Laboratory of Human
Genetics of Athens University and the First Department of Internal
Medicine of the Medical School of Athens, were included in the study.
The blood samples were collected in an outpatient basis, as patient's
clinically steady state did not require hospitalization.
Seven (7/35) patients were smokers, while one (1/35) presented
cardiac insufficiency, two (2/35) presented diabetes mellitus and one
(1/35) suffered from rheumatoid arthritis. Eight patients (8/35)
received hydroxyurea (HU) and only 4 (4/35) had been transfused
occasionally. Twenty-five (25/35) patients had been splenectomized.
Twenty healthy age and sex-matched individuals were included in the
control group. Patients' demographic and clinical characteristics are
shown in Table 1.
Patients were divided in different study groups according to their
treatment (hydroxyurea, splenectomy), in order to evaluate the effect
of these therapeutic approaches on the expression of endothelial
adhesion molecules.
Blood was collected into sterile EDTA vacuum tubes, transported
on ice to the laboratory, and immediately centrifuged at 1500 g for
15 min at 4 °C to avoid cytokine synthesis or breakdown in vitro.
Serum samples were stored at − 70 °C for no longer than 15 days
before assay.
231
Table 1
β-thalassemia intermedia patients' characteristics (mean ± SD, median in parentheses).
β-thalassemia intermedia patients (n = 35)
Age (years) 34
(mean, range) 8–63
Sex (men, women) 13/22
Transfusions 4/35
Hydroxyurea 8/35
Splenectomy 23/35
Hb (g/dl) 8.3 ± 1.5
(8.0)
Hb F (%) 55.4 ± 29.0
(61.5)
PLT (× 103/ml) 496.3 ± 259.3.
(533.0)
Ferritin (ng/ml) 650.6± 523.7(489.5)
Non-Smokers 14/35
Serum levels of sVCAM-1, sICAM-1, P-selectin, and E-selectin were
measured in duplicate using enzyme-linked immunosorbent assays
(R and D Systems, Minneapolis, MN, USA), while sTM concentration
was measured with an enzyme-linked immunosorbent assay pro-
vided by Asserachrom, Diagnostica Stago, France. Assays were
performed on serum specimens as recommended by the manufac-
turers. The performance characteristics of these assays, have already
been established by the manufacturer, with inter-assay and intra-
assay coefficient of variation each b10%. CRP levels were measured
with routine nephelometry methods.
All measures were performed in the Department of Clinical
Biochemistry of “Aghia Sophia” Children's Hospital.
The statistical procedures were performed using the STATGRAFICS
PLUS version 5.1 for Windows program (Graphic Software System)
while the regression plots were prepared using the Sigma plot version
8.0 program (Sigma-Aldrich Chemical Co). The software's ability to
choose between t-test and Wilcoxon test according to the population's
distribution was used. The graphs were done with R.
Results
We compared serum levels of soluble endothelial molecules in β-
thalassemia intermedia patients with healthy controls. As shown in
Fig. 1, levels of sICAM-1 (p b 0.0001), sVCAM-1 (p b 0.0001), P-selectins
(p b 0.0001) and E-selectins (p b 0.0001) are significantly higher
compared to controls. Levels of sTM are also higher but results do
not reach statistical significance (p = 0.65).
Patients were then divided in groups according to different
therapeutic approaches. Eight patients (8/35) received hydroxyurea
(HU) and twenty-five (25/35) patients had been splenectomized. We
compared levels of endothelial adhesion molecules in patients
receiving HU to those who did not. No significant differences are
observed in those two groups (Fig. 2). We also compared levels of
endothelial adhesion molecules in splenectomized patients to those
who were not splenectomized. No significant differences were
observed in levels of sICAM-1, sTM and E-sel in the two groups.
However, levels of sVCAM-1 are found significantly increased in non-
splenectomized patients (p = 0.02), whereas statistically significant
differences are observed in plasma levels of P-sel in the splenecto-
mized patients' group (p b 0.0001) (Fig. 3).
We extended our analysis searching possible interactions between
soluble adhesion molecules in our different groups of patients. In
patients who were not receiving hydroxyurea we observed a negative
correlation between sICAM-1 and sTM (p = 0.04) and betweenVCAM-
1 and P-sel (p = 0.0004), whereas a positive correlation was observed
between sICAM-1 and sVCAM-1 (p = 0.02), sTM and sVCAM-1
(p = 0.009) and between P-sel and E-sel (p = 0.003). On the contrary,
in patients treated by HU no statistically significant relationships
between soluble adhesion molecules were observed, besides a
232 I. Kanavaki et al. / Blood Cells, Molecules, and Diseases 43 (2009) 230–234

Fig. 1. Boxplot showing the distribution of measured values for ICAM-1, VCAM-1, TM, P-sel and E-sel, in thalassemia intermedia patients (TI patients) and controls. (⁎) denotes
statistical significance.

positive correlation of P-sel and E-sel (p = 0.006). In the group of
splenectomized patients a positive correlation is observed between
sTM and sVCAM-1 (p = 0.02) as well as between P-sel and E-sel
(p = 0.029), whereas sVCAM-1 and P-sel are negatively correlated
(p = 0.005). In the non-splenectomized group of patients, we only
observed a negative correlation between sTM and P-sel (p = 0.035).
We also studied the levels of CRP in β-thalassemia intermedia
patients compared to healthy individuals. CRP levels were signifi-
cantly increased in our group of patients, (p b 0.0001) regardless of
therapy (Fig. 4).
Discussion
Previous studies have shown higher levels of endothelial adhesion
molecules, such as sICAM-1, sVCAM-1, E-sel and vWF, in the serum of
patients with α and β thalassemia [12,14]. In vitro studies came to the
same conclusion [12]. In other clinical studies concerning transfusion-
dependent thalassemia major patients, plasma levels of sICAM-1 and
sVCAM-1 were found significantly increased [9,13]. Plasma levels of
sTM were also elevated in β-thalassemia patients and this team
proposes the use of sTM as a marker of endothelial damage [14].
Few studies are focused on thalassemia intermedia. In an in vitro
study authors found increased adhesion of erythrocytes on the
endothelial membrane, however no specific adhesion molecules are
mentioned [15]. The overall mechanism seems however similar to that
described in all thalassemia syndromes and endothelial dysfunction is
now considered one of the pathophysiologic axes of thalassemia
intermedia [16].
In this study, the levels of soluble endothelial adhesion molecules
such as sICAM-1, sVCAM-1, P- and E-selectins were found significantly
higher compared to healthy individuals. Levels of thrombomodulin
are also higher than those of healthy subjects, without however

Fig. 2. Boxplot showing the distribution of measured values for ICAM-1, VCAM-1, TM, P-sel and E-sel, in thalassemia intermedia patients (TI patients) under Hydroxyurea treatment
or not and controls. (⁎) denotes statistical significance.
 I. Kanavaki et al. / Blood Cells, Molecules, and Diseases 43 (2009) 230–234 233

Fig. 3. Boxplot showing the distribution of measured values for ICAM-1, VCAM-1, TM, P-sel and E-sel, in thalassemia intermedia patients (TI patients) that are underwent
splenectomy or not and controls. (⁎) denotes statistical significance.

reaching statistical significance. We did not find any significant In conclusion, the results of this study show that in thalassemia
difference in soluble endothelial adhesion molecule levels in patients intermedia the marked elevation of the measured molecules show
treated with hydroxyurea compared to patients who did not receive that there is a serious degree of endothelial activation and damage
this treatment. underlying the pathophysiologic mechanisms of this condition with
We found also that splenectomy did not result in any changes in otherwise mild clinical manifestations. CRP was also elevated,
levels of sICAM-1, sTM and E-sel. Levels of sVCAM-1 however were implying a chronic inflammatory state present in these patients.
found significantly higher in non-splenectomized patients compared More studies are needed correlating these findings with patients'
to the splenectomized group. In the study by Kyriakou et al. [13] the clinical state, in order to identify further implications with possible
authors found no difference in adhesion molecule expression in therapeutic protocols.
splenectomized β-thalassemia patients compared to non-splenecto-
mized patients, while Butthep et al. [14] mention the presence of Acknowledgments
higher levels of sTM in non-splenectomized patients compared to the
splenectomized group. Grant/funding support: Funding were received from Athens
Further analysis searching possible interactions between soluble University, (to I.P and J.R.). The funding sources played no role in
adhesion molecules in our group of patients was performed and we the study design; in the collection, analysis, and interpretation of data;
found that sICAM-1 and sVCAM-1, as well as P- and E-selectins, were in the writing of the report; or in the decision to submit the report for
positively correlated. We also observed a negative correlation publication.
between levels of sICAM-1 and sTM. This result agrees with the Financial disclosures: None declared.
study of Wu et al. who go further by suggesting the use of this
correlation as a predictive marker of increased risk for coronary heart
disease [17] based on the fact that sICAM-1, a known marker of
endothelial activation and damage, negatively interacts with sTM who
possesses vasoprotective properties. In our group of β-thalassemia
intermedia patients, we suggest the hypothesis that this interaction
between sICAM-1 and sTM reflects the important extent of vascular
damage presented in these patients, who are otherwise characterized
by a subtle clinical phenotype.
We included in our study measurement of CRP. CRP is believed to
play a critical role in vascular disease, by promoting a pro-inflamma-
tory reaction [18,19]. Little is known about the role of acute phase
proteins in hemoglobinopathies. In our study levels of CRP were found
significantly higher in patients compared to controls, regardless of
treatment (hydroxyurea or splenectomy), the result of a chronic
underlying inflammation of β-thalassemia intermedia patients. These
results agree with the study published by Archararit et al. [10], where
CRP levels were found elevated in β-thalassemia patients compared to
healthy individuals, especially in splenectomized patients [10].
Additionally we observed no significant correlation between CRP and
endothelial adhesion molecules. Probably, despite the fact that
elevated levels of CRP reflect the overall alteration of the biochemical
profile of these patients, there is no direct interaction between Fig. 4. Boxplot showing the distribution of measured values for CRP in thalassemia
inflammation markers and endothelial adhesion molecules. intermedia depending on their treatment. (⁎) denotes statistical significance.
234 I. Kanavaki et al. / Blood Cells, Molecules, and Diseases 43 (2009) 230–234
References
[1] C. Camaschella, M.D. Cappellini, Thalassemia intermedia, Haematologica 80
(1995) 58–68.
[2] A. Taher, H. Isma'eel, M.D. Cappellini, Thalassemia intermedia: revisited, Blood
Cell. Mol. Dis. 37 (2006) 12–20.
[3] A. Aessopos, M. Kati, J. Meletis, Thalassemia intermedia today: should patients
regularly receive transfusions? Transfusion 47 (2007) 792–800.
[4] A.T. Taher, Z.K. Otrock, I. Uthman, M.D. Cappellini, Thalassemia and hypercoagul-
ability, Blood Rev. 22 (2008) 283–292.
[5] A. Eldor, E.A. Rachmilewitz, The hypercoagulable state in thalassemia, Blood 99
(2002) 36–43.
[6] M. Karimi, M. Khanlari, E.A. Rachmilewitz, Cerebrovascular accident in beta-
thalassemia major (beta-TM) and beta-thalassemia intermedia (beta-TI), Am. J.
Hematol. 83 (2008) 77–79.
[7] L. Manfre, E. Giarratano, A. Maggio, et al., MR imaging of the brain: findings in
asymptomatic patients with thalassemia intermedia and sickle cell-thalassemia
disease, Am. J. Roentgenol. 173 (1999) 1477–1480.
[8] M.D. Cappellini, L. Robbiolo, B.M. Bottasso, et al., Venous thromboembolism and
hypercoagulability in splenectomized patients with thalassemia intermedia, Br. J.
Haematol. 111 (2000) 467–473.
[9] C. Aggeli, C. Antoniades, C. Cosma, et al., Endothelial dysfunction and inflamma-
tory process in transfusion-dependent patients with beta-thalassemia major, Int. J.
Cardiol. 105 (2005) 80–84.
[10] N. Archararit, S. Chuncharunee, A. Pornvoranunt, et al., Serum C-reactive protein level
in postsplenectomized thalassemic patients, J. Med. Assoc. Thai. 83 (2000) S63–69.
[11] M.D. Cappellini, Coagulation in the pathophysiology of hemolytic anemias,
Hematol. Am. Soc. Hematol. Educ. Prog. (2007) 74–78.
[12] P. Butthep, A. Bunyaratvej, Y. Funahara, et al., Possible evidence of endothelial cell
activation and disturbance in thalassemia: an in vitro study, Southeast Asian J.
Trop. Med. Public Health 28 (1997) 141–148A.
[13] D.S. Kyriakou, M.G. Alexandrakis, E.S. Kyriakou, et al., Activated peripheral blood
and endothelial cells in thalassemia patients, Ann. Hematol. 80 (2001) 577–583.
[14] P. Butthep, A. Bunyaratvej, Y. Funahara, et al., Alterations in vascular endothelial
cell-related plasma proteins in thalassaemic patients and their correlation with
clinical symptoms, Thromb Haemost. 74 (1995) 1045–1049.
[15] T. Hovav, A. Goldfarb, G. Artmann, S. Yedgar, G. Barshtein, Enhanced adherence of
beta-thalassaemic erythrocytes to endothelial cells, Br. J. Haematol. 106 (1999)
178–181.
[16] A. Aessopos, M. Kati, D. Farmakis, Heart disease in thalassemia intermedia: a
review of the underlying pathophysiology, Haematologica 92 (2007) 658–665.
[17] K.K. Wu, N. Aleksic, C.M. Ballantyne, et al., Interaction between soluble
thrombomodulin and intercellular adhesion molecule-1 in predicting risk of
coronary heart disease, Circulation 107 (2003) 1729–1732.
[18] V. Pasceri, J.T. Willerson, E.T. Yeh, Direct proinflammatory effect of C-reactive
protein on human endothelial cells, Circulation 102 (2000) 2165–2168.
[19] S. Verma, S.H. Li, M.V. Badiwala, et al., Endothelin antagonism and interleukin-6
inhibition attenuate the proatherogenic effects of C-reactive protein, Circulation
105 (2002) 1890–1896.