Global vasomotor dysfunction and accelerated vascular aging in β-thalassemia major

Atherosclerosis 198 (2008) 448–457 Global vasomotor dysfunction and accelerated vascular aging in ␤-thalassemia major George Hahalis a,∗ , Dimitrios T. Kremastinos b , George Terzis c , Andreas P. Kalogeropoulos a , Athina Chrysanthopoulou c , Marina Karakantza d , Alexandra Kourakli d , Stamatis Adamopoulos e , Alexandros D. Tselepis f , Nikos Grapsas g , Dimitrios Siablis c , Nicholas C. Zoumbos d , Dimitrios Alexopoulos a b a Department of Cardiology, University Hospital of Patras, Greece Second Department of Cardiology, Attikon General Hospital, University of Athens, Greece c Department of Radiology, University Hospital of Patras, Greece d Department of Hematology, University Hospital of Patras, Greece e Second Department of Cardiology, Onassis Cardiac Surgery Center, Athens, Greece f Laboratory of Biochemistry, Department of Chemistry, University of Ioannina, Greece g First Department of Cardiology, “Agios Andreas” Hospital, Patras, Greece Received 25 April 2007; received in revised form 23 September 2007; accepted 24 September 2007 Available online 7 November 2007 Abstract Background: Patients with ␤-thalassemia major (␤-TM) demonstrate an increased incidence of vascular complications, which are thought to result from a procoagulant/proinflammatory environment. We investigated the arterial vasorelaxing capacity and sought for early carotid atherosclerosis and underlying pathophysiological correlates in these transfusion-dependent patients. Methods and results: The vasodilatory properties of the brachial artery and the carotid intima-media thickness (IMT) were examined with ultrasonography in 35 non-diabetic young adults with ␤-TM (patient group) and 35 control subjects (control group). Among thalassemic patients, both endothelium-dependent (FMD) and -independent dilatation (FID) as well as their ratio was impaired, whereas IMT was increased (p < 0.01). Patients on optimal, as compared with those on non-optimal chelation treatment had a non-significantly lower IMT. Vasodilatory capacity in the patient group was inversely correlated with IMT and independently associated either with the quality of chelation therapy (FMD) or serum ferritin levels (FID). Plasma concentrations of D-dimers, circulating markers of endothelial activation, inflammation and apoptosis were higher, while plasma cholesterol and fibrinogen levels were lower-than-normal in the patient group. Independent predictors of IMT among thalassemic patients were tumor necrosis factor-␣ levels and age. Conclusions: Young adults with ␤-TM exhibit both a global impairment of arterial vasorelaxation and early carotid atherosclerosis. A procoagulant/proinflammatory state in these transfusion-dependent patients may overwhelm atheroprotective mechanisms, including an optimal chelation regimen, and promote vascular injury and atherogenesis. © 2007 Elsevier Ireland Ltd. All rights reserved. Keywords: Thalassemia; Carotid atherosclerosis; Endothelium; Iron-overload; Inflammation ∗ Corresponding author. Tel.: +30 2610 435026/999281; fax: +30 2610 992941. E-mail address: ghahalis@otenet.gr (G. Hahalis). 0021-9150/$ – see front matter © 2007 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.atherosclerosis.2007.09.030 G. Hahalis et al. / Atherosclerosis 198 (2008) 448–457 1. Introduction Beta-thalassemias are monogenic hemolytic anemias, which result from reduced synthesis of ␤-globin chains. Frequent blood transfusions and increased gastrointestinal iron absorption pose thalassemic patients at risk for iron overload and damage of parenchymal organs [1,2]. Despite improved survival with chelation therapy, cardiovascular complications are still relatively common among patients who suffer from ␤-thalassemia major (␤-TM), i.e. the severe, transfusion-dependent phenotype of this disorder. Apart from heart disease, which remains the main cause of death in these patients, a higher-than-average incidence of vascular complications has been consistently reported and accounted for by both a hypercoagulable state and vascular endothelial activation [1–5]. Such thromboembolic events usually manifest in the 3rd decade of life [4] in ∼1% of patients with ␤-TM [3,4] and are responsible for ∼4% of deaths [3]. Prior investigations suggest a proatherothrombotic role for iron overload and ␤-TM. Potential mechanisms are believed to include enhanced platelet activation, LDL oxidation, macrophage activity stimulation, and increased nitric oxide destruction in the context of oxidative stress and hemolysis [6–10]. Iron-loading has been linked with endothelial function, circulating cholesterol oxidation products, and possibly coronary artery disease [11–16], whereas a procoagulant milieu in thalassemia may participate in accelerated atherogenesis [5,17–23]. Impaired flow-mediated dilatation of the brachial artery is thought to detect early stages of atherosclerosis because of its association with atherosclerotic risk factors and prognosis [24]. Likewise, intima-media thickness (IMT) of the carotid artery wall is linked with cardiovascular risk factors, including fibrinogen levels and blood viscosity as well as with iron-related oxidative stress [24–27]. Carotid IMT is related both with incident and prevalent cardiovascular disease and is an accepted measure of subclinical atherosclerosis [25]. The impact of low-level inflammation on subclinical atherosclerosis is a focus of recent research [28–33]. Patients with ␤-TM exhibit an altered immune response pattern, which has been ascribed among other causes, to antigenic stimulation related to transfusions and infectious agents, cytokine levels of the stored allogenic blood, iron overload, and stroma cells of the hyperplastic bone marrow [34–41]. In addition, activated leucocytes circulate in the peripheral blood in these patients [20]; and enhanced apoptosis, which is considered to participate in carotid plaque pathology, occurs in bone marrow erythroid precursors in ␤-TM [42,43]. Thus, it is conceivable from the current body of evidence that the interplay of iron overload, high oxidative stress, hemostatic disturbances, endothelial activation, proinflammatory state, and thalassemia itself may contribute to a proatherogenic environment in ␤-TM even on the background of a characteristic, rather favorable lipid pattern. The aim of this study was therefore to seek for possible prema- 449 ture atherosclerosis as well as potential pathophysiological correlates in these patients. 2. Methods 2.1. Study design Thirty-five patients with ␤-TM, who were on regular blood transfusions since their first years of life agreed to participate in the study. The diagnosis for ␤-TM was established in our patients if they met all of the following criteria, in addition to the necessity for regular transfusions since their first months of life: (1) both parents had to be known carriers of the ␤-thalassemia trait, and (2) severe anemia during infancy along with characteristic appearance of the peripheral blood film and unusually high levels of fetal hemoglobin had to be present. In all of the 19 patients, who underwent genotype evaluation the diagnosis of homozygous ␤-thalassemia could be confirmed. All patients were free from diabetes mellitus and cardiac disease and demonstrated normal biventricular systolic function on echocardiography. None of them exhibited either history of heart failure or asymptomatic systolic dysfunction on annual echocardiographic examinations over the last 5 years before the index study. All patients were receiving desferrioxamine approximately 40–50 mg/kg subcutaneously overnight and ascorbic acid orally. Target of desferrioxamine therapy was the achievement of serum ferritin concentrations ≤1000 ng/ml, unless toxic drug effects were noticed. Compliance with chelation therapy was considered optimal if patients were >90% adherent to the instructions given by the hematologists. Serum ferritin levels were determined three to five times each year. For comparison, 35 healthy volunteers, who were matched with the patient group for age, sex, body surface area, and smoking habits, were also examined. The institutional review board has approved the research protocol and informed consent has been obtained from all subjects. 2.2. Biomarker measurements Venous blood samples were obtained after 12 h of fasting. Measurements included a complete blood count, HDL-, LDL-, and total cholesterol, as well as concentrations of hepatic transaminases, fibrinogen, and D-dimer. C-reactive protein (CRP) concentrations were obtained by a highsensitivity assay (N Latex CRP II, DADE Behring). The levels of soluble vascular cell adhesion molecule (sVCAM1), soluble intercellular adhesion molecule type-1 (sICAM), soluble E-selectin, high-sensitivity TNF-␣, TNF-␣ type I and type II receptor, granulocyte-macrophage colony stimulating factor (GM-CSF) and soluble Fas (sFas) were measured in duplicate by the quantitative sandwich enzyme-linked immunoassay technique (ELISA, R&D Systems, Minneapolis, MN, USA). All samples were stored in −80 ◦ C until assayed. 450 G. Hahalis et al. / Atherosclerosis 198 (2008) 448–457 2.3. Ultrasound studies Doppler echocardiographic examinations included Mmode, two-dimensional and Doppler studies. B-mode and Doppler ultrasonography was performed with a 5–12 MHz linear-array transducer. Patients were refrained from food intake and smoking for at least 6 h before ultrasonographic examinations. All vascular studies were performed at 2:00 p.m., after taking a medical history and measuring resting pulse and blood pressure, with each patient remaining supine in a quiet, air-conditioned room (temperature between 20 and 23 ◦ C), as previously described [15]. After obtaining a baseline scan, a pneumatic cuff was placed around the forearm and inflated to a pressure of 250 mmHg for 4.5 min. Cuff deflation resulted in reactive hyperemic blood flow, which is the stimulus for FMD and a surrogate measure of hyperemic shear stress. A scan of the brachial artery was performed within 45–90 s thereafter. A period of 10 min was then allowed for recovery of the vessel. Sublingual glyceryl trinitrate was subsequently administered (400 ␮g) to provoke flow-independent dilatation, and 3–4 min later the scan was taken. Post-ischemic FMD and FID were determined as the percentage of diameter change from the baseline values. The FMD/FID ratio was also used in order to assess the relative contribution of FMD to FID, i.e. to the maximally obtainable arterial smooth muscle cell (SMC) relaxation. Carotid artery examinations were performed with the patient’s head turned 45◦ from the side being scanned. The left and right common carotid arteries were examined in the anterolateral, posterolateral, and mediolateral directions. Reference point for measurement of the IMT was the beginning of the dilatation of the carotid bulb, with loss of the parallel configuration of the near and far walls of the common carotid artery. On the screen displaying the frozen magnified image, the sonographer recorded the distance between the boundaries of lumen-intima and media-adventitia interfaces at the far wall of the common carotid arteries. The mean IMT of the six measurements in each patient was then calculated. Plaque was defined as a focal structure encroaching into the arterial lumen at least 0.5 mm or 50% of the surrounding IMT. In the presence of plaque, IMT was measured at the nearest point free of plaque. An intraobserver variability analysis in a series of 35 consecutive studies showed a mean percentage error of <4% and <5% for FMD and IMT, respectively. block (forward stepwise approach; p to enter <0.05, p to remove >0.10). Significance was determined at the p < 0.05 level. All statistical analyses were performed using SPSS Version 13 (SPSS Inc.). 3. Results 3.1. Baseline characteristics The two groups had similar baseline characteristics, including blood pressure and plasma glucose levels. Thalassemic patients exhibited moderately elevated plasma ferritin levels and were estimated to be, on average, at an intermediate cardiovascular risk [2]. Liver biopsy was performed in 12 patients. Histologically, hepatitis and cirrhosis were found in four and one of these patients, respectively. Moreover, severe hepatic siderosis could be shown in 2 out of the 12 patients. Nineteen (55%) of the patients were on optimal desferrioxamine therapy, whereas history of splenectomy, endocrinopathy and active or healed hepatitis was evident in a significant minority of them. As compared with the non-optimally treated subgroup of thalassemic patients, those on optimal desferrioxamine therapy exhibited significantly lower ferritin levels (1011 ± 629 ng/ml vs. 3340 ± 1661 ng/ml, p < 0.001). Furthermore, patients with ␤-TM demonstrated lower-than-normal plasma cholesterol levels, including non-HDL cholesterol as well as increased concentrations of both plasma triglycerides and liver transaminases (Table 1). 3.2. Biomarkers Furthermore, fibrinogen levels were lower and D-dimer concentrations were increased among thalassemic patients, while values of sVCAM and sICAM-1 were significantly elevated in patients compared to controls indicating endothelial activation. The proinflammatory cytokines IL-6, TNF-␣ as well as the apoptotic marker sFas were also higher-thannormal in the patient group, whereas the TNF-␣ RII, but not the non-inducible TNF-␣ RI, was increased among thalassemic patients. Notably, CRP levels were non-significantly higher in the patient group, while plasma GM-CSF levels were detectable in only eight patients and one control subject (Table 2). 2.4. Statistics 3.3. Ultrasonography All continuous variables are presented as mean ± S.D. and compared between groups by the unpaired Student’s t-test with Welch correction. Bivariate relationships between outcome variables (IMT, FMD and FID) and predictor variables (clinical parameters, biochemical data and serum markers) were evaluated with Pearson correlation coefficients. We used multiple regression models to identify independent multivariate predictors of FMD, FID and IMT by entering all variables with significant correlation coefficients into the beginning On echocardiography, patients exhibited similar left ventricular ejection fraction as compared with the healthy control group (66 ± 5% vs. 67 ± 6%, p = ns), but greater cardiac dimensions, higher stroke volume, and increased left ventricular mass, as previously reported [2] (left ventricular diameter, end-diastolic: 5.1 ± 0.6 cm vs. 4.7 ± 0.6 cm; endsystolic: 3.4 ± 0.5 cm vs. 3.0 ± 0.4 cm, left atrial diameter: 3.8 ± 0.5 cm vs. 3.1 ± 0.4 cm, stroke volume: 77 ± 15 ml vs. 451 G. Hahalis et al. / Atherosclerosis 198 (2008) 448–457 Table 1 General characteristics of the study groups Age (years) Male sex, n (%) Body surface area (m2 ) Systolc blood pressure (mmHg) Diastolic blood pressure (mmHg) Heart rate (beats/min) Smoking, n (%) Hb, pre-transfusional (mg/dl) Serum ferritin over the last year (ng/ml) Serum ferritin over the last 5 years (ng/ml) Splenectomy, n (%) Hepatitis C, active or healed, n (%) History of hypothyroidism, n (%) History of hypoparathyroidism, n (%) History of hypogonadism, n (%) Glucose (mg/dl) Total cholesterol (mg/dl) LDL-cholesterol (mg/dl) HDL-cholesterol (mg/dl) Total/HDL-cholesterol ratio Non-HDL cholesterol (mg/dl) Triglycerides (mg/dl) Aspartate aminotransferase (U/l) Alanine aminotransferase (U/l) a ␤-TM (n = 35) Control (n = 35) p-Value 27 ± 7 49 (17) 1.66 ± 0.18 106 ± 13 68 ± 7 74 ± 10 40 (14) 10.3 ± 0.9 1975 ± 1635 2270 ± 1640 6 (17) 12 (34) 15 (42) 6 (17) 14 (40) 99 ± 13 120 ± 34 61 ± 27 32 ± 10 4.0 ± 1.6 88 ± 32 155 ± 61 41 ± 22 54 ± 39 29 ± 7 49 (17) 1.71 ± 0.19 111 ± 12 67 ± 8 74 ± 10 40 (14) 13.6 ± 1.4 122 ± 65a NA 0 0 0 0 0 94 ± 12 196 ± 41 130 ± 40 51 ± 14 4.2 ± 1.5 146 ± 42 82 ± 39 19 ± 6 21 ± 11 NS NS NS 0.1 NS NS NS <0.001 <0.001 NA <0.001 <0.001 <0.001 <0.001 <0.001 NS <0.001 <0.001 <0.001 NS <0.001 <0.001 <0.001 <0.001 Single measurement for each control subject; NA: non-applicable. 66 ± 21 ml, left ventricular mass: 122 ± 41 g vs. 78 ± 32 g, all p < 0.05). Furthermore, Doppler values of mitral inflow were compatible with increased preload in the patient group, without evidence of left ventricular diastolic dysfunction [higher early trasmitral inflow blood velocity E and similar earlyto-late (E/A) inflow velocity ratios as well as E-deceleration time among thalassemic patients in comparison with healthy controls (data not shown)]. The vascular ultrasonographic results are presented in Table 3. At comparable baseline brachial artery diameters and hyperemic blood flow between the two groups, both FMD and FID were blunted among thalassemic patients in comparison with healthy controls (Fig. 1). In the patient group, a lower-than-normal ratio of FMD/FID was also evident, whereas FMD appeared more pronounced among those patients who were optimally chelated (Fig. 2). Furthermore, Fig. 1. FMD and FID in the study groups (bars represent mean ± S.D.). a higher carotid IMT was found among thalassemic patients as compared with healthy subjects (Fig. 3). 3.3.1. Determinants of impaired brachial artery dilatation and carotid artery IMT Table 2 Clotting factors, circulating adhesion molecules, and proinflammatory/proapoptotic markers Fibrinogen (mg/dl) D-dimers (mg/dl) C-reactive protein, high sensitivity (mg/dl) Soluble VCAM (ng/ml) Soluble IC AM-1 (pg/ml) E-Selectin (ng/ml) IL-6 (pg/ml) TNF-␣, high sensitivity (pg/ml) TNF-␣ type I receptor (pg/ml) TNF-␣ type II receptor (pg/ml) Soluble Fas (ng/ml) ␤-TM (n = 35) Control (n = 35) p-Value 239 0.41 0.30 881 988 43.8 3.3 3.5 1245 3443 4.2 303 0.25 0.19 451 177 49.4 1.3 1.8 1285 1518 2.3 <0.001 <0.001 0.1 <0.001 <0.001 NS <0.001 <0.001 NS <0.001 <0.001 ± ± ± ± ± ± ± ± ± ± ± 50 0.13 0.29 296 214 26.4 1.5 1.5 368 935 0.8 ± ± ± ± ± ± ± ± ± ± ± 60 0.05 0.16 45 23 21.9 0.9 0.7 581 344 1.0 VCAM denotes vascular cell adhesion molecule; ICAM, intercellular adhesion molecule; IL, interleukin; and TNF, tumor necrosis factor. 452 G. Hahalis et al. / Atherosclerosis 198 (2008) 448–457 Table 3 Ultrasound data IMT (mm) Baseline brachial artery diameter (mm) Baseline flow (ml/min) FMD brachial artery diameter (mm) FMD flow (ml/min) FMD (%) FID brachial artery diameter (mm) FID flow (ml/min) FID (%) FMD/FID ratio Table 4 Predictors of IMT ␤-TM (n = 35) Control (n = 35) p-Value 0.51 ± 0.07 3.71 ± 0.64 0.46 ± 0.07 3.92 ± 0.58 <0.01 NS 82 ± 56 3.95 ± 0.68 60 ± 43 4.33 ± 0.63 0.068 <0.05 363 ± 194 6.5 ± 2.7 4.27 ± 0.67 426 ± 185 10.6 ± 3.9 4.67 ± 0.65 NS <0.001 <0.05 85 ± 53 14.9 ± 4.9 0.45 ± 0.21 63 ± 44 18.2 ± 5.6 0.59 ± 0.16 0.067 0.01 <0.05 IMT denotes intima-media thickness of the carotid artery wall; FMD, flow-mediated dilatation of the brachial artery; FID, flow-independent (nitroglycerin-induced) dilatation of the brachial artery. Variable Univariate p-Value Multivariate (r) (β) p-Value 0.57 0.57 0.53 0.52 0.50 <0.001 <0.001 0.001 0.001 0.001 0.41 0.41 – – – <0.01 <0.01 – – – Patients with ␤-thalassemia major TNF-␣ 0.66 Age 0.37 Diastolic blood pressure 0.39 Total cholesterol 0.39 <0.001 <0.05 <0.05 <0.05 0.43 0.40 – – <0.05 <0.05 – – Healthy subjects Age Smoking Total/HDL-cholesterol Total cholesterol LDL-cholesterol hemoglobin level (r = −0.63, p < 0.001), and baseline arterial diameter (r = −0.37, p < 0.05). In multivariate models only hemoglobin level (β = −0.50, p < 0.001) was independent predictor of FID. Among patients, FID was significantly related to baseline arterial diameter (r = −0.52, p = 0.001) and mean serum ferritin level of the last 5 years (r = 0.46, p < 0.01), both of which conferred independent contributions in multivariate models. Moreover, we found a correlation between FMD and FID only among control subjects (r = 0.62, p < 0.001) but not among patients with ␤-TM. 3.4. Determinants of the carotid IMT Fig. 2. Effect of chelation therapy adherence on FMD. 3.3.1.1. Flow-mediated dilatation. In univariate analysis, adherence to iron chelation therapy was the sole predictor of FMD among thalassemic patients, whereas no significant predictor of FMD was identified among healthy controls. 3.3.1.2. Flow-independent dilatation. Healthy control subjects demonstrated a significant correlation of FID to We found no univariate correlation between serum ferritin levels or echocardiographic indices and IMT. Simple regression analysis demonstrated an inverse correlation of IMT with both FMD and FID among thalassemic patients but not among healthy control subjects (Fig. 4). Compared with patients on optimal chelation treatment, those on insufficient desferrioxamine therapy exhibited similar IMT (0.50 ± 0.07 vs. 0.53 ± 0.07, p = ns). Healthy subjects demonstrated a positive correlation of IMT with age, smoking, total/HDLcholesterol ratio, as well as total and LDL-cholesterol levels. In multivariate models, age and smoking independently predicted IMT. Among thalassemic patients, IMT was positively correlated with TNF-␣ levels, age, diastolic blood pressure and baseline arterial diameter. By multivariate analysis, TNF␣ concentrations and age were independently associated with IMT in the patient group (Table 4 and Fig. 5). 4. Discussion Fig. 3. IMT among thalassemic and control subjects. We found global vasomotor dysfunction of the brachial artery and increased carotid IMT in patients with ␤-TM as compared with healthy subjects. Both FMD and FID were inversely correlated with IMT among patients but not among healthy subjects. Abnormal FMD of the brachial artery denotes reduced bioavailability of NO and was more pronounced among non-optimally chelated patients. Impairment of FID suggests a defective SMC response to exogenous G. Hahalis et al. / Atherosclerosis 198 (2008) 448–457 453 Fig. 4. Correlation of FMD and FID with IMT in thalassemic subjects. NO donors and was associated with serum ferritin levels over the last 5 years, whereas the abnormal ratio of FMD/FID indicates that an intrinsic NO deficiency does exist in thalassemia, in addition to SMC dysfunction. Early carotid thickening in thalassemic patients is a novel finding. It was demonstrated on a background of a rather antiatherogenic lipid and rheological pattern, enhanced thrombin generation and fibrinolysis, endothelial activation, as well as proinflammatory and proapoptotic milieu. 4.1. Vasomotor dysfunction The endothelium controls vascular integrity, vasomotor tone and inflammatory process. Dysfunctional endothelium facilitates subendothelial migration of leucocytes and atherogenesis by virtue of redox-sensitive induction of inflammatory genes and attending expression of chemokines and adhesion molecules [9]. Our results of increased levels of circulating adhesion molecules denote profound activation of monocytes and endothelial cells and confirm previous studies [5,19,20]. Endothelial dysfunction in patients with ␤TM is a conceivable finding in the context of high oxidative stress, hemolytic release of free hemoglobin and frequent blood transfusions [5,10,18]. It has been shown, that the severely disturbed balance between oxidative stress and antioxidant potential in ␤-TM parallels the extent of iron overload [18]. Furthermore, abnormal circulating erythrocytes in these patients demonstrate increased endothelial adhesiveness and are capable of initiating complex mechanisms of clot formation and endothelial perturbation [5]. Kyriakou et al. showed an association of adhesion molecules with both ferritin levels and TNF-␣ [20]. Other investigations on ␤-TM reported endothelial dysfunction but normal FID [19,44] and are thus partially at variance with the current findings. Of note, one study included underweight patients with near normal hemoglobin and cholesterol levels [44], while in the other report and in accordance with the current findings, FMD was associated with the extent of iron burden [19]. Impaired FID in ␤-TM is another novel finding, which probably signifies a more advanced vascular dysfunction and adversely affects prognosis [45]. In thalassemia, longstanding stressors may alter the integrity of the arterial wall, including distal signaling defects, SMC dysfunction [46] or other, disease-specific abnormalities. The positive association between FID and serum ferritin levels over the last 5 years is difficult to explain. Impairment of vascular SMC function could theoretically result from increased superoxide concentrations and/or hemosiderin deposits in SMCs in a milieu of amplified oxidative stress and iron overload [21,47]. Furthermore, elastic tissue abnormalities of the arterial wall in ␤-TM may modify SMC responsiveness and FID via conformational changes of elastin polymers [48]. Whether chronically increased iron overload and SMC function are causally related, suggesting favorable adaptive mechanisms in the arterial wall over longer time periods, remains unclear. 4.2. Carotid wall intima-media thickening Fig. 5. Correlation of TNF-␣ plasma levels with IMT in thalassemic patients. Apart from the established association between IMT and cardiovascular risk factors, the impact of inflammation on carotid disease is being increasingly investigated [28–33]. For example, increased IMT as well as circulating inflammatory markers were found to be associated with carotid artery disease [29], whereas histologically determined carotid plaque inflammation, especially infiltration with macrophages, was linked with plaque instability and time since stroke [30]. The reported immunomodulation in thalassemia is confirmed in our study and has clinical relevance in view of the potential risk for premature atherosclerosis [31–33,49]. In the current study, high plasma concentrations of proinflamma- 454 G. Hahalis et al. / Atherosclerosis 198 (2008) 448–457 tory cytokines and sFas among thalassemic patients suggest activation of proinflammatory as well as apoptotic mechanisms in patients with ␤-TM. No prior study on thalassemia has examined IMT; therefore, its possible associations with circulating proinflammatory markers are lacking. We have previously shown that stimulated mononuclear cells of patients with ␤-TM are capable of elaborating large amounts of TNF-␣ which is considered as a serum marker of immune activation [38]. The immunological abnormalities in ␤-TM are complex, well known and often profound [34–37]. Moreover, their underlying pathogenetic mechanisms, whether primary or acquired are incompletely understood and the clinical consequences unclear. Indirect evidence suggests that iron loading, not necessarily of transfusional origin, seems to be involved in this process. Thus, immunological defects tend to normalize following a successful chelation regimen, whereas clinically, the incidence of pericarditis has declined in the era of desferrioxamine therapy [2,37]. Whatever the source of inflammation may be, the repeatedly reported activation of vascular endothelium in ␤-TM clearly point toward an augmented inflammatory process in these patients. In the current study, we found that only age and TNF-␣ were independent predictors of IMT in the patient group. The proatherogenic properties of TNF-␣ encompass the enhanced expression of leukocyte adhesion molecules on endothelial cells, SMCs, and macrophages, as well as of IL-6 on SMCs [49]. Furthermore, TNF-␣ promotes iron sequestration and oxidative stress in human endothelial cells as well as SMC migration, T-cell activation, collagen degradation and possibly SMC apoptosis in human atheromas [50,51]. Desferrioxamine treatment reduces TNF-␣ bioavailability via steady state inhibition as well as enhanced inactivation through TNF-␣ RII binding and limits the in vivo activation of the transcription factor NF-␬B. It appears therefore plausible that IMT might be casually linked rather than merely associated with TNF-␣ in the context of incomplete iron removal [39] and high oxidative stress, i.e. conditions which seem to amplify TNF-␣-induced expression of several atherogenic genes [50]. In a previous clinical study, an inconsistent association between IMT and plasma TNF␣ levels was evident [33]. Other investigators demonstrated either an association of TNF␣ expression in circulating mononuclear cells with the clinical course of carotid disease [28] or that circulating TNF-␣ receptors, rather than TNF-␣ itself, independently determined carotid plaque thickness [32]. Our findings of a non-significant elevation of plasma CRP concentrations in the patient group is probably due to inadequate sample power and the young age of the study subjects, because older patients with ␤-thalassemia intermedia, i.e. a less severe phenotype demonstrate high CRP levels (own unpublished data). Whether sFas receptor levels, as a marker of apoptosis rate, are associated with atherosclerosis is debated [52]. Regardless of the source of apoptotic markers in thalassemia [43], we confirmed others [52] by showing that increased sFas levels in our patient group were not related to IMT. Lack of association between carotid IMT and examined molecules possibly imply that they are not heavily involved in carotid atherosclerosis among patients with ␤TM. Alternatively, they may be less linked with prevalent than with incident cardiovascular disease and/or rather depend on plasma-lowering mechanisms of chelation therapy as well as splenectomy status [31,37,39,40]. We found lower-than-normal plasma cholesterol concentrations in the patient group as previously shown [19]. The lipid profile in ␤-TM has been considered as atheroprotective and largely attributed to LDL removal by the hyperplastic bone marrow and to chronic monocyte/macrophage activation with attendant heightened cytokine secretion in the setting of chronic inflammation [19,53]. 4.3. Clinical relevance of the study findings Although FMD and IMT appear pathogenetically linked, they are not interchangeable in terms of short-/mid-term measurements changes and prognosis. In fact, the former surrogate marker of atherosclerosis, but not IMT, demonstrates a ‘dynamic’ temporal measurements pattern, that is often the result of successful treatment of atherogenic stimuli, including intensive chelation in ␤-TM [54]. Moreover, FMD likely carries independent prognostic potential, which is additional to that conferred by IMT alone [55], thus indicating that each of them measures different aspects and stages of early atherosclerosis. Does increased IMT among our patients indicate premature atherosclerosis? We suggest that this is the case for the following reasons: first, IMT was independently associated with age, which is pathophysiologically obvious and, by simple regression analysis, inversely correlated with markers of early atherosclerosis, i.e. FMD and FID; second, carotid plaques were encountered in three study patients but in none of the control subjects, a conceivable appearing finding in the context of increased IMT (Fig. 6) [28]; third, carotid plaques are commonly identified among older patients with ␤-thalassemia intermedia (own unpublished data); fourth, we have recently shown that IMT independently determines the risk of transient ischemic attacks [56]; and fifth, it has been demonstrated that even a limited blood transfusion program in the setting of acute coronary syndromes may have a major impact on the strong association between bleeding and cardiovascular complications [57]. Recent studies point toward both similarities and differences of symptomatic carotid versus coronary artery disease, including plaque morphology and differential impact of serum cholesterol levels [30]. Such differences may be operative in the observed distortion of the relationship between incident coronary artery disease and event rates in other vascular beds in ␤-TM [53,58]. Among our ␤-TM patients at our Institution, for example, three young adults suffered from ischemic strokes and two from venous thromboembolic episodes, but none of them developed coronary artery disease over an 18-year period. We also found that the coronary G. Hahalis et al. / Atherosclerosis 198 (2008) 448–457 455 Fig. 6. Ultrasonographic images of the common carotid artery of three thalassemic patients, (A through E), demonstrating plaques (arrows). In B and D, the color flow of the first two patients is also depicted. In F, the carotid intima-media of a healthy subject appears thin and smooth. arteries of such patients appear smooth angiographically [59]. Increased IMT suggests that carotid artery disease could be a possible cause of strokes in cardiac disease-free patients with ␤-TM. Alternatively, the coronary arteries may reflect a well-protected vascular bed by the thalassemic lipid profile, whereas the carotid disease represents a model of premature aging, which could rather be a marker than the cause of increased propensity for vascular complications. Co-existing genetic factors, which are evident in ∼40% of these patients, and/or pre-disposing conditions during life (e.g., divergence from optimal chelation treatment, splenectomy, surgery or immobilization) could out-of-proportion enhance the thrombotic risk, as compared with thalassemic patients without such risk factors [4]. 4.4. Limitations The current study is observational and therefore unable to prove causality of the documented associations. It is possible that circulating TNF-␣ in ␤-TM simply reflects a proinflammatory “vulnerable” environment and is not causally linked with IMT. Since only single biochemical measurements were undertaken, the cumulative temporal burden of exposure is uncertain. The consistency of data on ␤-TM suggest however a considerable activation of both endothelium and proinflammatory mechanisms over time. Our study was not specifically designed to elucidate the role of iron on vascular function and structure in ␤-TM. Lack of association between serum ferritin concentrations and both FMD and IMT does not preclude the notion that iron overload is implicated in atherogenesis, but rather suggests a subtle, underlying pathophysiology. Thus, serum ferritin has been correlated with pro-coagulant/pro-inflammatory environment in prior studies on ␤-TM [19,20,39]. Furthermore, iron burden may interact in a complex manner with multiple factors, including the properties of thalassemic patients’ own erythrocytes, chronic anemia, oxidative stress, enormously expanded bone marrow, splenectomy status as well as high output and hypermetabolic state, all of which may participate in thalassemic atherogenesis [2,4–18,20,40]. Last, measures of iron loading in a small sample of thalassemic patients are probably imprecise to establish an association between iron overload and subclinical atherosclerosis. In fact, ferritin, although of prognostic relevance, is influenced by concomitant liver dis- 456 G. Hahalis et al. / Atherosclerosis 198 (2008) 448–457 ease (which was evident in about one-third of thalassemic patients) and, as an acute phase reactant, by acute illnesses [2]. The alternative option of patients’ categorization into those on optimal vs. non-optimal chelation treatment may confer additional information (as shown in FMD), because this approach encompasses not only “biochemistry” but also clinical aspects and the overall clinician’s viewpoint. 5. Conclusions We demonstrated an impairment of arterial vasorelaxation and premature carotid atherosclerosis in ␤-TM. Low-level inflammation and a prothrombotic state may counteract atheroprotective mechanisms, accelerate vascular ageing and explain the relatively high incidence of vascular complications in these multi-transfused patients. References [1] Ladis V, Chouliaras G, Berdousi H, Kanavakis E, Katamis C. Longitudinal study of survival and causes of death in patients with thalassemia major in Greece. Ann NY Acad Sci 2005;1054:445–50. [2] Hahalis G, Alexopoulos D, Kremastinos DT, Zoumbos NC. Heart failure in ␤-thalassemia syndromes: a decade of progress. Am J Med 2005;118:957–67. [3] Borgna-Pignatti C, Cappellini MD, De Stefano P, et al. Survival and complications in thalassemia. Ann NY Acad Sci 2005;1054:40–7. [4] Taher A, Isma’eel H, Mehio G, et al. Prevalence of thromboembolic events among 8,860 patients with thalassaemia major and intermedia in the Mediterranean area and Iran. Thromb Haemost 2006;96:488–91. [5] Eldor A, Rachmilewitz EA. The hypercoagulable state in thalassemia. Blood 2002;99:36–43. [6] Pratico D, Pasin M, Barry OP, et al. Iron-dependent human platelet activation and hydroxyl radical formation: involvement of protein kinase C. Circulation 1999;99:3118–24. [7] Balla G, Jacob HS, Eaton JW, Belcher JD, Vercellotti GM. Hemin: a possible physiological mediator of low density lipoprotein oxidation and endothelial injury. Arterioscler Thromb 1991;11:1700–11. [8] Scaccabarozzi A, Arosio P, Weiss G, et al. Relationship between TNFalpha and iron metabolism in differentiating human monocytic THP-1 cells. Br J Haematol 2000;110:978–84. [9] Kunsch C, Medford RM. Oxidative stress as a regulator of gene expression in the vasculature. Circ Res 1999;85:753–66. [10] Schechter AN, Gladwin MT. Hemoglobin and the paracrine and endocrine functions of nitric oxide. N Engl J Med 2003;348: 1483–5. [11] Day SM, Duquaine D, Mundada LV, et al. Chronic iron administration increases vascular oxidative stress and accelerates arterial thrombosis. Circulation 2003;107:2601–6. [12] Rooyakkers TM, Stroes ES, Kooistra MP, et al. Ferric saccharate induces oxygen radical stress and endothelial dysfunction in vivo. Eur J Clin Invest 2002;32:S9–16. [13] Nitenberg A, Paycha F, Ledoux S, et al. Coronary artery responses to physiological stimuli are improved by deferoxamine but not by l-arginine in non-insulin-dependent diabetic patients with angiographically normal coronary arteries and no other risk factors. Circulation 1998;97:736–43. [14] Tuomainen TP, Diczfalusy U, Kaikkonen J, Nyyssonen K, Salonen JT. Serum ferritin concentration is associated with plasma levels of cholesterol oxidation products in man. Free Radic Biol Med 2003;35: 922–8. [15] Gaenzer H, Marschang P, Sturm W, et al. Association between increased iron stores and impaired endothelial function in patients with hereditary hemochromatosis. J Am Coll Cardiol 2002;40:2189–94. [16] Gillum RF. Body iron stores and atherosclerosis. Circulation 1997;96:3261–3. [17] Brizzi P, Isaja T, D’Agata A, et al. Oxidized LDL antibodies (OLAB) in patients with beta-thalassemia major. Atheroscler Thromb 2002;9:139–44. [18] Livrea MA, Tesoriere L, Pintaudi AM, et al. Oxidative stress and antioxidant status in beta-thalassemia major: iron overload and depletion of lipid-soluble antioxidants. Blood 1996;88:3608–14. [19] Aggeli C, Antoniades C, Cosma C, et al. Endothelial dysfunction and inflammatory process in transfusion-dependent patients with betathalassemia major. Int J Cardiol 2005;105:80–4. [20] Kyriakou DS, Alexandrakis MG, Kyriakou ES, et al. Activated peripheral blood and endothelial cells in thalassemia patients. Ann Hematol 2001;80:577–83. [21] Tsomi K, Karagiorga-Lagana M, Fragodimitri C, Karabatsos F, Katsiki V. Arterial elastorrhexis: manifestation of a generalized connective tissue disorder in beta-thalassemia major. Eur J Heamatol 1999;63:287–94. [22] Aessopos A, Farmakis D, Loukopoulos D. Elastic tissue abnormalities resembling pseudo-xanthoma elasticum in beta thalassemia and the sickling syndromes. Blood 2002;99:30–5. [23] Aessopos A, Farmakis D, Karagiorga M, Rombos I, Loucopoulos D. Pseudoxanthoma elasticum lesions and cardiac complications as contributing factors for strokes in ␤-thalassemia patients. Stroke 1997;28:2421–4. [24] Heitzer T, Schlinzig T, Krohn K, Meinertz T, Munzel T. Endothelial dysfunction, oxidative stress, and risk of cardiovascular events in patients with coronary artery disease. Circulation 2001;104:2673–8. [25] Mancini GB, Dahlof B, Diez J. Surrogate markers for cardiovascular disease. Circulation 2004;109:IV22–30. [26] Levenson J, Gariepy J, Salomon J, Denarie N, Simon A. Association of plasma viscosity and carotid thickening in a French working cohort. Am J Hypertens 2000;12:753–875. [27] Drueke T, Witko-Sarsat V, Massy Z, et al. Iron therapy, advanced oxidation protein products, and carotid artery intima-media thickness in end-stage renal disease. Circulation 2002;106:2212–7. [28] Profumo E, Esposito C, Buttari B, et al. Intracellular expression of TNF-␣ and IL-6 in peripheral mononuclear cells is associated with the outcome of contralateral disease after carotid endarteriectomy. Atherosclerosis 2006;(May 4) [Epub ahead of print]. [29] Magyar MT, Szikszai Z, Balla J, et al. Early-onset carotid atherosclerosis is associated with increased intima-media thickness and elevated serum levels of inflammatory markers. Stroke 2003;34:58–63. [30] Virmani R, Ladich ER, Burke AP, Kolodgie FD. Histopathology of carotid atherosclerotic disease. Neurosurgery 2006;59:S219–27. [31] Schillinger M, Exner M, Mlekusch W, et al. Inflammation and carotid artery—risk for atherosclerosis study (ICARAS). Circulation 2005;111:2203–9. [32] Elkind MS, Cheng J, Boden-Albala B, et al. Tumor necrosis factor receptor levels are associated with carotid atherosclerosis. Stroke 2002;33:31–7. [33] Skoog T, Dichtl W, Boquist S, et al. Plasma tumour necrosis factor-␣ and early carotid atherosclerosis in healthy middle-aged men. Eur Heart J 2002;23:376–83. [34] Neri A, Callea V, D’Ascola DG, Martino B. Immune status in thalassemia major. Haematologica 1986;71:511–6. [35] Rougier F, Cornu E, Praloran V, Denizot Y. IL-6 and IL-8 production by human bone marrow stromal cells. Cytokine 1998;10:93–7. [36] Lombardi G, Matera R, Minervini MM, et al. Serum levels of cytokines and soluble antigens in polytransfused patients with beta-thalassemia major: relationship to immune status. Haematologica 1994;79:406–12. [37] Del Vecchio GC, Schettini F, Piacente L, et al. Effects of deferiprone on immune status and cytokine pattern in thalassaemia major. Acta Haematol 2002;108:144–9. G. Hahalis et al. / Atherosclerosis 198 (2008) 448–457 [38] Salsaa B, Zoumbos NC. A distinct pattern of cytokine production from blood mononuclear cells in multitransfused patients with betathalassaemia. Clin Exp Immunol 1997;107:589–92. [39] Bellocq A, Israel-Biet D, Cadranel J, et al. Reduction of tumour necrosis factor alpha expression and signalling in peripheral blood mononuclear cells from patients with thalassaemia or sickle cell anaemia upon treatment with desferrioxamine. Cytokine 1999;11:168–72. [40] Archararit N, Chuncharunee S, Pornvoranunt A, et al. Serum C-reactive protein level in postsplenectomized thalassemic patients. J Med Assoc Thai 2000;83:S63–9. [41] Spiess BD. Blood transfusion for cardiopulmonary bypass: the need to answer a basic question. J Cardiothorac Vasc Anesth 2002;16:535–8. [42] Tricot O, Mallat Z, Heymes C, et al. Relation between endothelial cell apoptosis and blood flow direction in human atherosclerotic plaques. Circulation 2000;101:2450–3. [43] Yuan J, Angelucci E, Lucarelli G, et al. Accelerated programmed cell death (apoptosis) in erythroid precursors of patients with severe betathalassemia (Cooley’s anemia). Blood 1993;82:374–7. [44] Cheung YF, Chan GC, Ha SY. Arterial stiffness and endothelial function in patients with beta-thalassemia major. Circulation 2002;106:2561–6. [45] Adams MR, Robinson J, McCredie R, et al. Smooth muscle dysfunction occurs independently of impaired endothelium-dependent dilatation in adults at risk of atherosclerosis. J Am Coll Cardiol 1998;32:123–7. [46] Nikolaidis LA, Mathier MA, Doverspike A, et al. Coronary blood flow responses are impaired independent of NO and endothelial function in conscious dogs with dilated cardiomyopathy. J Card Fail 2005;11:313–21. [47] Miller Jr FJ, Gutterman DD, Rios CD, et al. Superoxide production in vascular smooth muscle contributes to oxidative stress and impaired relaxation in atherosclerosis. Circ Res 1998;82:1298–305. [48] Herrera B, Desco MM, Eisenberg G, Garcıa-Barreno P, Del Canizo JF. Role of elastic fibers in cooling-induced relaxation. Cryobiology 2002;44:54–61. 457 [49] Young JL, Libby P, Schönbeck U. Cytokines in the pathogenesis of atherosclerosis. Thromb Haemost 2002;88:554–67. [50] Nanami M, Ookawara T, Otaki Y, et al. Tumor necrosis factor-␣induced iron sequestration and oxidative stress in human endothelial cells. Arterioscler Thromb Vasc Biol 2005;25:2495–501. [51] Boyle JJ, Weissberg PL, Bennett MR. Tumor necrosis factor-alpha promotes macrophage-induced vascular smooth muscle cell apoptosis by direct and autocrine mechanisms. Arterioscler Thromb Vasc Biol 2003;23:1553–8. [52] van der Meer IM, Oei HH, Hofman A, et al. Soluble Fas, a mediator of apoptosis, C-reactive protein, and coronary and extracoronary atherosclerosis. Atherosclerosis 2006;189:464–9. [53] Fridlender ZG, Rund D. Myocardial infarction in a patient with betathalassemia major: first report. Am J Hematol 2004;75:52–5. [54] Tanner MA, Galanello R, Dessi C, et al. A randomized, placebocontrolled, double-blind trial of the effect of combined therapy with deferoxamine and deferiprone on myocardial iron in thalassemia major using cardiovascular magnetic resonance. Circulation 2007;115:1876–84. [55] Chan SY, Mancini GB, Kuramoto L, et al. The prognostic importance of endothelial dysfunction and carotid atheroma burden in patients with coronary artery disease. J Am Coll Cardiol 2003;42:1037–43. [56] Kalogeropoulos A, Terzis G, Chrysanthopoulou A, et al. Risk for transient ischemic attacks is mainly determined by intima-media thickness and carotid plaque echogenicity. Atherosclerosis 2007;192:190–6. [57] Eikelboom JW, Mehta SR, Anand SS, et al. Adverse impact of bleeding on prognosis in patients with acute coronary syndromes. Circulation 2006;116:774–82. [58] Singh R, Venketasubramanian N. Reccurent cerebral infarction in betathalassemia major. Cerebrovasc Dis 2004;17:344–5. [59] Kremastinos DT, Tiniakos G, Theodorakis GN, Katritsis DG, Toutouzas PK. Myocarditis in ␤-thalassemia major: a cause of heart failure. Circulation 1995;91:66–71.