- Email: [email protected]
Increased endothelial monocyte adhesiveness is related to clinical outcomes in chronic heart failure
International Journal of Cardiology 121 (2007) 276 – 283 www.elsevier.com/locate/ijcard
Increased endothelial monocyte adhesiveness is related to clinical outcomes in chronic heart failure Wei-Hsian Yin a,c,d , Jaw-Wen Chen b,d,⁎, Mason Shing Young a , Shing-Jong Lin b,c,d a
Division of Cardiology, Department of Internal Medicine, Cheng-Hsin Rehabilitation Medical Center, Taipei-Veterans General Hospital, Taiwan b Division of Cardiology, Taipei-Veterans General Hospital, Taiwan c Institute of Clinical Medicine, National Yang-Ming University, School of Medicine, Taipei, Taiwan d Cardiovascular Research Centre, National Yang-Ming University, School of Medicine, Taipei, Taiwan Received 30 May 2006; accepted 2 November 2006 Available online 22 December 2006
Abstract Background: Vascular inflammation and endothelial dysfunction are evident in patients with chronic heart failure (CHF). We hypothesized that circulating peripheral blood mononuclear cells (PBMCs) may be activated and the resultant increased endothelial monocyte adhesion may be functionally and pathophysiologically relevant in CHF. In the present study, we investigated the clinical significance of the activity of PBMCs in patients with CHF. Methods: PBMCs were isolated from 34 CHF patients, from 10 healthy volunteers (normal control group) and from 17 patients admitted for investigation of suspected coronary artery disease (disease control group). In each patient, the adhesiveness of PBMCs to cultured human aortic endothelial cells (HAECs) with or without tumor necrosis factor-alpha (TNF-α) stimulation was determined. Major adverse cardiac events (death, heart transplantation or hospitalization with worsening heart failure) were determined in the 34 CHF patients during a median follow-up period of 182 days. Results: Compared with those from both control groups and from mild CHF patients, PBMCs isolated from severe CHF patients adhered more to the HAECs. The endothelial adhesiveness of PBMCs correlated positively with the circulating levels of CAMs and can supply prognostic information in CHF patients. The difference between event-free curves based on the median levels of endothelial-PBMC adhesion was significant (log rank test, p = 0.0139). Conclusions: Endothelial adhesiveness of PBMCs is increased and correlated to clinical outcomes, and may be pathophysiologically relevant to the progression of CHF. © 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Chronic heart failure; Endothelial monocyte adhesiveness; Cell adhesion molecules; Prognosis
1. Introduction Recent studies suggest that immune activation and inflammation may play a role in the pathophysiology of chronic heart failure (CHF), which might have implications for the management of heart failure [1–4]. The main components of the immune system that are thought to be relevant to ⁎ Corresponding author. Division of Cardiology, Taipei-Veterans General Hospital, No. 201, Sec. 2, Shih-Pai Rd, Pei-Tou, Taipei 112, Taiwan. Tel.: +886 2 28712121x3793; fax: +886 2 28711601. E-mail address: [email protected] (J.-W. Chen). 0167-5273/$ - see front matter © 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijcard.2006.11.012
the pathophysiology of CHF include cytokines, cell adhesion molecules (CAMs) and nitric oxide [4]. It was reported by us and others that the activation of CAMs in endothelial cells and the circulating levels of soluble CAMs increase with the severity of CHF and are related to clinical outcomes in patients with CHF [5–10]. However, there is still debate as to whether CAMs are involved in the pathophysiology of CHF or merely markers of inflammatory response. In the development and progression of cardiovascular inflammation, a key step is the adherence of circulating mononuclear cells such as monocytes to the vascular endothelium through binding of highly regulated CAMs
W.-H. Yin et al. / International Journal of Cardiology 121 (2007) 276–283
expressed on the surface of monocytes or endothelial cells or both, followed by the subsequent extravasation of these mononuclear cells into the involved tissue [11,12]. Recent evidence further demonstrates that as compared to that in normal subjects, the expressions of CAMs such as endothelial-leukocyte adhesion molecule-1 (ELAM-1), intercellular cell adhesion molecule-1 (ICAM-1), vascular cell adhesion molecule-1 (VCAM-1) and integrin molecules can be enhanced within the myocardial microvasculature of patients with severe CHF [13], suggesting that vascular inflammation might be involved in the propagation and maintenance of cardiac deterioration in CHF even without ischemia in origin. Furthermore, circulating mononuclear cells such as monocytes could be activated and may play a significant role in the immunologic imbalance observed in advanced CHF [14]. In the present study, we tested the hypothesis that isolated peripheral blood mononuclear cells (PBMCs) from patients with CHF are activated and their adherence to cultured human aortic endothelial cells (HAECs) may be enhanced in vitro, and that the enhanced endothelial adhesiveness of PBMCs can supply prognostic information in these CHF patients. The novel findings of this study may provide a rationale for the clinical significance of mononuclear cell activation and vascular inflammation in the progression of CHF with both ischemic and non-ischemic origins. 2. Methods 2.1. Study population A total of 34 inpatients (28 men and 6 women, aged 63± 15 years) with chronic but relatively stable heart failure were enrolled from cardiology services at Cheng-Hsin Rehabilitation Medical Center. Men and women at least 18 years old with New York Heart Association (NYHA) functional class II to ambulatory class IV symptoms of CHF and a left ventricular ejection fraction (LVEF) of 40% or less by echocardiography within 7 days were eligible for enrolment. The 34 CHF patients were admitted for elective cardiac catheterization for evaluation of systolic dysfunction with unexplained causes despite non-invasive testing. The etiology of CHF was determined as ischemic when coronary angiography revealed N 70% luminal diameter narrowing in at least 1 major epicardial coronary artery. In those patients with CHF without coronary artery disease for whom the endomyocardial biopsy revealed findings compatible with dilated cardiomyopathy, the cause of CHF was determined to be dilated cardiomyopathy. The time intervals between blood sampling and admission were 1 ± 1 days. All necessary forms of conventional therapy were maintained throughout the study by the primary staff cardiologists. None of the selected patients were in frank heart failure, but all had symptoms of weakness and dyspnea at levels of activity varying from mild to a sedentary state. Patients were not allowed to participate if they had any
277
of the following occurring within two weeks before entry into the study: change in NYHA functional class, change in anti-failure medications, or administration of any intravenous medication for heart failure. Patients were excluded when they had hemodynamically significant obstructive valvular heart disease, cor pulmonale, restrictive or hypertrophic cardiomyopathy, myocarditis, constrictive pericarditis or congenital heart disease. Patients were excluded if there was severe co-morbidity, renal failure, or if there was clinical or laboratory evidence of systemic infection or an inflammatory illness such as sepsis, malignancy, arthritis or connective tissue disease. Those patients with non-steroid anti-inflammatory drugs, antioxidants, L-arginine, or lipid-lowering agents such as statins and fibrates were also excluded. Because it would have been ethically unfeasible to stop aspirin in patients with ischemic heart disease, the use of aspirin was determined by the attending physician. Two groups provided blood samples to serve as control subjects: the first comprised 10 healthy volunteers (normal control group); the second included 17 patients admitted for coronary angiography for investigation of suspected coronary artery disease and/or chest pain (disease control group). The following inclusion criteria were used for enrolment of the normal control group: negative family history of cardiovascular disease, nonsmoking status, and absence of hypercholesterolemia, hypertriglyceridemia and diabetes mellitus. On the basis of data from epidemiological and family studies, a cohort fulfilling these criteria is expected to have a very low prevalence of asymptomatic cardiovascular disease. Therefore, coronary angiography was not performed to rule out the presence of asymptomatic coronary artery disease because it would have been ethically unfeasible in these subjects. Baseline clinical evaluation, including a full medical history, physical examination, basic laboratory study and echocardiographic study, was performed on the 34 patients with CHF and on both control groups. No subject in either control group had any clinical signs or symptoms of heart failure and their LVEFs were all N 50% by echocardiography. All subjects gave informed consent, and the Human Investigation Committee of Cheng-Hsin Rehabilitation Medical Center approved the study protocol. 2.2. Cardiac catheterization and hemodynamic measurements At the time of enrolment, all 34 CHF patients and the 17 patients of the disease control group underwent cardiac catheterization or coronary arteriography for clinically indicated purposes. Left-sided cardiac catheterization, including left ventriculography and coronary arteriography, was performed in all 34 CHF patients and 17 disease control patients. Right-sided cardiac catheterization with or without endomyocardial biopsy was performed in 29 patients. Right heart pressure was measured with a 7Fr Swan-Ganz catheter, and cardiac output was measured by the thermodilution technique.
278
W.-H. Yin et al. / International Journal of Cardiology 121 (2007) 276–283
2.3. Culture of HAECs HAECs were isolated from the aortic portion of a single healthy donor, were provided as cryopreserved tertiary cultures by Cascade Biologics (Oregon, USA) and were grown in culture flasks in endothelial cell growth medium 200 (Cascade Biologics, USA) supplemented with fetal bovine serum (FBS, 2%), human epidermal growth factor (10 ng/mL), human fibroblast growth factor (3 ng/mL), heparin (10 μg/mL), penicillin (100 U/mL), streptomycin (100 pg/mL), and Fungizone (1.25 μg/mL). Cells were grown in a 5% CO2 humidified atmosphere at 37 °C and the growth medium was changed every other day until confluence. In this experiment, HAECs from second to eighth passage were used. Cells were identified positive for the von Willebrand factor and CD31 and negative for α-actin by immunohistochemistry. 2.4. Blood sampling and isolation of human PBMCs At the time of blood sampling, venous blood was drawn via central vein for those 29 patients who underwent right heart catheterization. For the other CHF patients and the control subjects, blood samples were collected from an indwelling catheter into vacuum tubes at bedside (peripheral vein) after ≧30 min of bed rest, with the patients in a supine position. PBMCs were isolated and extracted by density ultracentrifugation. In brief, 30 mL peripheral venous blood was drawn into a VACUTAINER® CPT™ Tube with 0.1 M sodium citrate (Becton Dickinson & Co., N.J., U.S.A.) at room temperature at bedside after 30 min of bed rest, with the study subjects in a supine position. The tubes were inverted 10 times gently. Then the blood was centrifuged and washed with EDTA/Hank's balanced salt solution (HBSS) solution (Sigma, St. Louis, Missouri, USA) as the operation instructions describe. The isolated PBMCs were then resuspended with serum-free RPMI1640 medium (M.A. Bioproducts, Walkersville, MD) for labeling. When the labeling procedure was completed, the RPMI1640 medium was removed by centrifugation and the cells were washed with EDTA/HBSS containing 2 mM Ca2+, 2 mM Mg2+, and 20 mM HEPES. The viability of the isolated PBMCs was assessed using the trypan blue exclusion. There was at least 95% viability in all samples used in these experiments. The isolation procedures were carried out at room temperature within 2 h of blood collection. After density ultracentrifugation, the separated plasma was frozen to −20 °C and stored at that temperature until analysis for circulating levels of CAMs. 2.5. Endothelial-PBMC adhesion assay For cell binding assay, HAECs (5 × 105 cells) were distributed into 6-well plates and were allowed to reach confluence. Before assay, the HAECs were incubated with or without 2 ng/mL TNF-α for 6 h at 37 °C. Isolated PBMCs from CHF patients and control subjects grown in RPMI 1640 medium containing 5% FBS were labeled for 45 min at
37 °C in darkness with 2′, 7′-bis (2-carboxyethyl)-5(6)carboxy-fluorescein acetoxymethyl ester (BCECF/AM, 10 μM, Boehringer-Mannheim, USA) in serum-free RPMI 1640 media, then the surplus was washed with HBSS to remove free dye and resuspended in RPMI 1640 containing 2% FBS. Labeled isolated PBMCs (3 × 105 cells) were added to each HAEC-containing well and incubation continued for 1 h at 37 °C. Non-adherent cells were removed by three gentle washes with HBSS, and the remaining cells were lysed with DMSO. Fluorescence was measured with an automated fluorometer (Multilabel Counter Victor2, Wallac, USA) at emission 530 nm and absorption 435 nm. 2.6. Measurement of circulating levels of CAMs In 32 of the CHF patients and the 10 normal subjects, assays for circulating VCAM-1, ICAM-1 and E-selectin were done concurrently to minimize any effects of repeated freeze–thaw cycles. The levels of VCAM-1, ICAM-1 and E-selectin were measured by means of enzyme-linked immunosorbent assay by commercial kits (R&D Systems, Inc., Minneapolis, MN for ICAM-1 and E-selectin and Biosource International, Camarillo, CA for VCAM-1). The intra-assay and interassay coefficients for all factors in our laboratory were approximately 5% and 10% respectively. 2.7. Clinical follow-up All CHF patients were observed through regular outpatient visits. Clinical information about major adverse cardiac events (death, requirement for heart transplantation or hospitalization with a primary diagnosis of worsening heart failure) during a median follow-up period of 182 days was provided by the cardiologists in charge without knowledge of the experimental results. 2.8. Data analysis All values are expressed as mean ± SEM. The patients with CHF were further divided into two groups based on their functional class: the mild CHF group (NYHA class II) and the severe CHF group (NYHA class III/IV). Univariate comparisons of clinical data, circulating levels of CAMs, and endothelial monocyte adhesiveness between two groups were made with the Student's t test for quantitative data and with the Fisher exact test for qualitative data. Comparisons between multiple groups were determined by means of a one-way analysis of variance (ANOVA) test. Linear regression analysis was used to determine the correlation between the endothelial monocyte adhesion and various circulating CAM levels. Kaplan–Meier analysis of cumulative event-free rates in CHF patients, stratified into two groups on the basis of median endothelial monocyte adhesion (expressed as % of control condition) were performed. The percentage of cell binding of the control condition (%) was determined by the fluorescent intensity of the attached cells to TNF-α-stimulated HAECs
W.-H. Yin et al. / International Journal of Cardiology 121 (2007) 276–283 Table 1 Baseline clinical characteristics of the study population Normal control (n = 10)
Disease control (n = 17)
Mild CHF (n = 15)
Severe CHF (n = 19)
45 ± 3 7(70) 65 ± 3 –
65 ± 3⁎ 12(71) 60 ± 2 13(76)
62 ± 4⁎ 12(80) 31 ± 2⁎# 3(20)#
64 ± 3⁎ 16(84) 24 ± 3⁎#§ 6(32)#
– –
12(70) 5(29)
9(60) 7(47)
12(63) 8(42)
– –
4(24) 3(18)
6(40) 2(13)
5(26) 2(10)
125 ± 4 68 ± 2 –
135 ± 5 71 ± 3 –
117 ± 5⁎#§ 95 ± 4⁎#§ 1889 ± 211
LVFP (mmHg)
–
18 ± 2
SVR (dyn s cm− 5) PVR (dyn s cm− 5) Medications, n (%) Diuretic Digitalis ACEI/ARB therapy Beta-blockers Nitrates Aspirin Monocyte adhesion (fluorescence) MACE rate (%) MACE, n (%) Death Heart transplantation Readmission for CHF
–
–
–
–
133 ± 5 85 ± 4⁎# 1883 ± 237 (n = 10) 26 ± 4# (n = 10) 1963 ± 250 (n = 10) 155 ± 26 (n = 10)
– – –
6(35) 1(6) 12(71)
13(87)# 8(53)# 10(67)
19(100)# 11(58)# 11(58)
– – – 2880 ± 987
9(53) 16(94) 12(70) 3809 ± 404
8(53) 10(67) 10(67) 3170 ± 535
10(53) 13(68) 12(63) 6893 ± 954⁎#§
–
–
40
53
– –
– –
0(0) 1(7)
4(21) 2(11)
–
–
5(33)
4(21)
Age (years) Male, n (%) LVEF (%) Ischemic heart disease (%) Hypertension (%) Diabetes mellitus (%) Smoking (%) Alcohol use (%) Hemodynamics SBP (mmHg) Heart rates (bpm) CI (L/min/m2)
30 ± 3# 2622 ± 291
There were more men than women in this sample. The average age of the CHF patients was 63 ± 15 years (48 to 80 years). The normal control group was significantly younger than the disease control group and the CHF population. However, the average ages were similar in the disease control group and the patients with mild and severe CHF. Fifteen patients were in NYHA class II (mild CHF group) and 19 were in class III/IV (severe CHF group). The cause of CHF was ischemic cardiomyopathy in 9 and dilated cardiomyopathy in 25. All of the 25 patients with dilated cardiomyopathy were proven by cardiac catheterization and endomyocardial biopsy. The incidence of ischemic heart disease was significantly higher in the disease control group than in the normal control group and the mild and severe CHF groups. However, the incidence of ischemic heart disease was similar in patients with mild and severe CHF. No Table 2 Patient characteristics and hemodynamic data in the patients with chronic heart failure who had major adverse cardiac events during follow-up and those who were event-free (n = 34)
228 ± 94
ACEI/ARB= angiotensin converting enzyme inhibitors or angiotensin II receptor blockers; CI = cardiac index; CHF = chronic heart failure; LVFP= left ventricular filling pressure; LVEF = left ventricular ejection fraction; MACE= major adverse cardiac event; PVR = pulmonary vascular resistance; SVR = systemic vascular resistance; SBP = systolic blood pressure. ⁎p b 0.01 vs. normal subjects, #p b 0.01 vs. disease controls, §p b 0. 05 vs. patients with mild CHF.
divided by fluorescent intensity of the control condition (PBMC binding to HAECs without TNF-α stimulation),×100. The differences in event-free curves were tested with a log rank test. A p value of less than 0.05 was considered statistically significant. 3. Results 3.1. Demographic and clinical features of the study population The demographic and clinical characteristics of the study population are shown in Table 1.
279
Age (years) Male, n (%) LVEF (%) Functional class Hemodynamics SBP (mmHg) Heart rates (bpm) CI (L/min/m2) LVFP (mmHg) SVR (dyn s cm− 5) PVR (dyn s cm− 5) Medication, n (%) Diuretic Digitalis ACEI/ARB therapy Beta-blockers Nitrates VCAM-1 (ng/ml) ICAM-1 (ng/ml) E-selectin (ng/ml) Endothelial adhesiveness of PBMC (fluorescence) Endothelial adhesiveness of PBMC (% of control) Follow-up period (days)
MACE (−) (n = 18)
MACE (+) (n = 16)
p value
59 ± 4 14 (78%) 29 ± 3 2.6 ± 0.2
68 ± 3 14 (88%) 26 ± 3 3.1 ± 0.2
NS NS NS NS
131 ± 5 84 ± 3 2160 ± 285 (n = 15) 24 ± 3 (n = 15) 1674 ± 314 (n = 15) 126 ± 34 (n = 15)
117 ± 6 97 ± 5 1750 ± 183 (n = 14) 29 ± 3 (n = 14) 2457 ± 225 (n = 14) 209 ± 52 (n = 14)
0.04 0.03 NS
16 (89) 10 (56) 12 (67) 12(67) 10(56) 442.0 ± 46.1 (n = 17) 180.8 ± 27.4 (n = 17) 33.3 ± 5.2 (n = 17) 3814 ± 715
16 (100) 9 (56) 9 (56) 6(38) 13(81) 546.8 ± 65.5 (n = 15) 276.1 ± 22.8 (n = 15) 64.3 ± 9.7 (n = 15) 6868 ± 1026
NS NS NS NS NS NS
0.018
239 ± 32
448 ± 43
b0.001
192 ± 28
80 ± 24
0.005
0.01 NS NS
0.022 0.049
ACEI/ARB = angiotensin converting enzyme inhibitors or angiotensin II receptor blockers; CI = cardiac index; ICAM-1 = intercellular adhesion molecule-1; LVFP = left ventricular filling pressure; LVEF = left ventricular ejection fraction; MACE = major adverse cardiac event; PBMC = peripheral blood mononuclear cells; PVR = pulmonary vascular resistance; SVR = systemic vascular resistance; SBP = systolic blood pressure; VCAM-1 = vascular cellular adhesion molecule-1.
280
W.-H. Yin et al. / International Journal of Cardiology 121 (2007) 276–283
significant differences in sex, incidence of hypertension and diabetes mellitus, smoking status or alcohol use were detected among the disease control subjects and patients with mild and severe chronic CHF. On blood sampling, all 34 patients were clinically stable, while receiving continuous oral medications for CHF. Nineteen patients used digoxin and 32 used diuretics; 21 patients were also treated with angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, 18 with beta-blockers and 23 with nitrates. Patients with CHF used more diuretics and digitalis than did disease control subjects. However, the use of aspirin, angiotensin-converting enzyme inhibitor or angiotensin receptor blocker, nitrate and beta-blocker therapy were similar in disease control subjects and CHF patients. 3.2. Left ventricular ejection fraction and hemodynamic measurements The mean LVEF was significantly lower and the left ventricular filling pressure was significantly higher in the CHF patients than in the control subjects. The mean systolic blood pressure was significantly lower and the heart rate was significantly higher in the severe CHF group than in the mild CHF group and the control subjects. Furthermore, the mean LVEF was significantly lower in the severe CHF group than in the mild CHF group. The CHF patients were then divided into those who had major adverse cardiac events during follow-up and those who were event-free (Table 2). No significant differences in age, sex, mean LVEF, functional status, or uses of medications for heart failure were detected between the two groups. However, the hemodynamic parameters, such as systolic blood pressure, heart rates, and left ventricular filling pressure, were worse in the group with major adverse cardiac events than in the event-free group.
3.3. Endothelial-PBMC adhesion in CHF patients and control subjects We chose TNF-α to stimulate HAECs, and under our experimental conditions, TNF-α (2 ng/mL) induced a 2- to 4-fold increase in the binding of PBMCs to HAECs both in patients with CHF and in control subjects. When equal numbers of labeled isolated PBMCs were added to HAEC-containing wells, PBMCs isolated from severe CHF patients adhered more to cultured HAECs than those from normal and disease control subjects and from mild CHF patients, whether or not the HAECs had been preincubated with TNF-α. Fig. 1 shows the relation between the stage and degree of CHF and the PBMC adhesion, expressed as fluorescence detected by fluorometer. In the present study, we enrolled 17 patients admitted for investigation of suspected coronary artery disease and/or chest pain as disease control subjects. The incidence of ischemic heart disease was significantly higher in this group than in the mild and severe CHF groups. Because the patients with ischemic heart disease may have higher endothelial monocyte adhesiveness, the question of whether ischemic heart disease accounts for the enhanced monocyte adhesion in this study and affects the clinical outcomes is a major concern. To answer this question, we divided the CHF patients into two groups based on their underlying causes of CHF (ischemic vs. non-ischemic) and compared the levels of endothelial monocyte adhesion between the two groups by the Wilcoxon rank-sum test. We found that the endothelial monocyte adhesiveness were similar in the patients with ischemic and non-ischemic heart failure whether or not the HAECs had been preincubated with TNF-α (1576 ± 323 vs. 2046 ± 363, p = 0.468 at baseline and 6666 ± 1205 vs. 4742 ± 773, p = 0.203 after TNF-α stimulation, respectively).
Fig. 1. Comparison between the clinical severity of heart failure and endothelial adhesiveness of peripheral blood mononuclear cells (PBMCs), expressed as fluorescence detected by fluorometer. The binding of PBMCs to either unstimulated or TNF-α stimulated human aortic endothelial cells (HAECs) was significantly increased in patients with severe chronic heart failure (CHF). TNF-α: TNF-α stimulated HAECs. ⁎p b 0.001 vs. normal subjects, ⁎⁎p b 0.001 vs. disease controls, ⁎⁎⁎p b 0.001 vs. patients with mild CHF.
W.-H. Yin et al. / International Journal of Cardiology 121 (2007) 276–283
281
3.4. Correlation between endothelial-PBMC adhesion and serum levels of CAMs The circulating levels of all the inflammatory markers were significantly higher in the patients with CHF than in the 10 normal subjects (data not shown). Significant positive correlations, with the correlation coefficients between PBMCs adhesion and circulating levels of VCAM-1, ICAM-1 and E-selectin of 0.665 (p b 0.0001), 0.398 (p = 0.0242) and 0.418 (p = 0.0173) respectively, were noted (Fig. 2). 3.5. Endothelial-PBMC adhesion and prognosis Although the heart failure condition of the study population was relatively stable instead of being acutely decompensated, the event rate during follow-up was relatively high (16 of 34; 47%). The median follow-up period for the whole group was 182 days (128–288 days, 25th–75th percentiles); the median follow-up periods for the group with major adverse cardiac events versus the eventfree group were 80 days and 192 days respectively (Table 2). Four of the 34 patients died of cardiac causes (2 died of sudden death without premonition of the progression of symptoms, presumed to be due to arrhythmia, and 2 of intractable end-stage heart failure) during the follow-up period; all of them were in the severe CHF group. Three patients underwent heart transplantation (one in the mild CHF group and 2 of them in the severe CHF group) and 9 patients were readmitted for worsening heart failure (5 of them in the mild CHF group and 4 of them in the severe CHF group) (Table 1).
Fig. 2. The correlations between endothelial adhesiveness of peripheral blood mononuclear cells (PBMCs) and serum levels of adhesion molecules. Endothelial adhesiveness of PBMCs was significantly correlated to circulating levels of VCAM-1, ICAM-1 and E-selectin.
As medications such as non-steroid anti-inflammatory drugs, antioxidants, lipid-lowering agents such as statins, fibrates and so on may inhibit endothelial monocyte adhesion, patients with those medications were excluded from this study. We did not exclude patients taking aspirin, but the use of aspirin was similar in the disease control group and the CHF population.
Fig. 3. Kaplan–Meier event probability for patients with heart failure stratified in 2 groups on the basis of median endothelial adhesiveness of peripheral blood mononuclear cells (PBMCs) (expressed as % of control). Patients with more increased (N272.141%) endothelial adhesiveness of PBMCs had worse clinical outcomes than those with less increased (b272.141%) endothelial adhesiveness of PBMCs (p = 0.0139).
282
W.-H. Yin et al. / International Journal of Cardiology 121 (2007) 276–283
We divided the 34 patients with CHF into two groups based on those who had major adverse cardiac events during follow-up versus those who were event-free (Table 2). The patients who had adverse events had significantly higher ICAM-1 and E-selectin levels and endothelial adhesiveness of PBMCs (p = 0.022, p = 0.049, and p b 0.0001, respectively) than the patients who were event-free. We then performed Kaplan–Meier analyses of event-free survival in the CHF patients on the basis of median endothelial monocyte adhesion (expressed as % of control). The difference between event-free curves based on the median levels of endothelial-PBMC adhesion was significant (log rank test, p = 0.0139) (Fig. 3). 4. Discussion 4.1. Comparison with previous studies of monocyte adhesion in CHF It has been shown that CAMs are upregulated in failing myocardium [10] as well as in the myocardial microvasculature [13] and that circulating CAM levels are increased in CHF patients [5–10]. We and others have also reported that the circulating levels of soluble CAMs increase with the severity of CHF and are related to clinical outcomes in patients with CHF [5,6,9,10]. The present study represents an addition to the growing literature of evidence of inflammation in patients with CHF. The main finding is that PBMCs from severe CHF patients demonstrate a greater adhesiveness to cultured HAECs than do cells from controls and mild CHF patients. The results are consistent with the hypothesis that the endothelial monocyte adhesiveness, an index of cell activity, is increased and appears to have predictive value for combined endpoint of death, transplant and heart failure readmission. It is suggested that vascular inflammation could be present and the presence of more sticky effector cells may contribute to the clinical progression of CHF with both ischemic and non-ischemic origins. To the best of our knowledge, these findings have not been reported before. 4.2. Potential role of monocyte adhesion in the pathophysiology of CHF Patients with advanced CHF have elevated circulating levels of proinflammatory cytokines and CAMs, irrespective of the cause of heart failure [1–10]. Endothelial cell activation by cytokines or other inflammatory mediators results in the expression of CAMs such as VCAM-1, ICAM1 and E-selectin, which in turn facilitates the attachment of blood leukocytes to endothelial cells [4,11–13]. The adherent cells are capable of producing greater amounts of enzymes (such as elastase) and toxic oxygen metabolites than nonadherent cells are, resulting in injury to the endothelium and transmigration of leukocytes [15]. The activated endothelial cells may alter intercellular gap
formation owing to disruption of cellular junctions [16]. Both damaged vascular integrity and impaired endothelial barrier function may increase endothelial permeability and transport of proteins and liquid via the paracellular pathway, resulting in endothelial dysfunction and tissue edema [17–19]. Furthermore, the leukocytes residing in the intima may produce a number of cytokines and growth factors, which may result in endothelial dysfunction, impairment of coronary vascular reserve and myocardial tissue perfusion, apoptosis of cardiomyocytes and endothelial cells, inhibition of myocyte contractility and myocardial fibrosis — a multitude of changes leading to progressive cardiac dysfunction and worsening heart failure [4]. Therefore, CAMs induced by inflammatory cytokines and the subsequent enhanced endothelial adhesion of PBMCs may play a direct role in the progression of CHF [13]. 4.3. Clinical implications Congestive heart failure is characterized by impaired cardiac performance and neurohormonal imbalance. Recent studies have also shown that vascular inflammation, endothelial activation and dysfunction are evident in CHF, as a result of increased oxidative stress status, reduced blood flow and decreased shear stress stimuli [1–4,20–22]. Even though the ultimate role of inflammation and endothelial dysfunction in the pathophysiology of the syndrome of heart failure is not yet fully understood, abnormalities of endothelial function may contribute to impaired myocardial perfusion due to myocardial microvascular inflammation [13] and the increased peripheral vasomotor tone both at rest and during exercise [17–19,23,24]. Both may be directly relevant to the pathophysiology of dyspnea, fatigue, exercise intolerance, cachexia and interstitial edema in CHF. Therefore, for therapeutic purposes, it could be one of the targets of anti-failure therapy to improve endothelial function and to inhibit vascular inflammation by preventing the expression of inflammatory molecules such as TNF-α and CAMs. This hypothesis is consistent with the fact that, while significantly improving long-term prognosis, treatment with angiotensin-converting enzyme inhibitors, angiotensin receptor blocker and carvedilol have been shown to decrease the plasma levels of TNF-α and CAMs and reduce endothelial inflammation in CHF patients [25–27]. 4.4. Study limitations One of the limitations of the study was the relatively small number of CHF patients included in each group. Even though significant correlation of the endothelial adhesiveness of PBMCs to prognosis was observed, the outcome data are rather limited and should be considered hypothesisgenerating rather than conclusive. Furthermore, the study did not investigate the molecular mechanisms of the enhanced endothelial adhesiveness of PBMCs in patients with CHF. Nor does it test the possible inhibitory effects of anti-failure
W.-H. Yin et al. / International Journal of Cardiology 121 (2007) 276–283
drugs on adhesiveness of PBMCs to endothelial cells in vitro. However, we have recently demonstrated that carvedilol, a medication currently widely used for clinical CHF, can directly, through its antioxidant property, inhibit the in vitro expression of VCAM-1 and the adhesiveness of human PBMCs to cultured endothelial cells stimulated by TNF-α [27]. While the cardiac protection of carvedilol in CHF is largely attributed to its antioxidant capability [28], it would be interesting to investigate further the potential effect of antioxidants and their related anti-inflammatory mechanisms on both myocardial microvascular and peripheral vascular protection regarding the progression of CHF. 5. Conclusions The adhesiveness of human PBMCs to endothelial cells is increased with the severity of CHF and correlated to clinical outcomes in patients with CHF. This is consistent with the hypothesis that increased expression of CAMs and enhanced PBMC adhesion may be functionally and pathophysiologically relevant. Acknowledgments The present study was supported in part by Cheng-Hsin Rehabilitation Medical Center grant 93–06 (to W.H.Y. and S.J.L.) and in part by Taipei-Veterans General Hospital grant VGH-91–263 (to J.W.C.). We also thank Ms. Chin-Feng Cheng for her excellent assistance in preparing and typing the manuscript. References [1] Seta Y, Shan K, Bozdurt B, Oral H, Mann DL. Basic mechanisms in heart failure: the cytokine hypothesis. J Card Fail 1996;2:243–9. [2] Mann DL. Activation of inflammatory mediators in heart failure. In: Mann DL, editor. Ed. Heart Failure: A Companion to Braunwald's Heart Disease. 1st ed. Philadelphia: Saunders; 2004. p. 159–80. [3] Hasper D, Hummel M, Kleber FX, Reindl I, Volk HD. Systemic inflammation in patients with heart failure. Eur Heart J 1998;19:761–5. [4] Sharma R, Coats AJS, Anker SD. The role of inflammatory mediators in chronic heart failure: cytokines, nitric oxide, and endothelin-1. Int J Cardiol 2000;72:175–86. [5] Yin WH, Chen JW, Jen HL, et al. The prognostic value of circulating soluble cell adhesion molecules in patients with chronic congestive heart failure. Eur J Heart Fail 2003;5:507–16. [6] Tsutamoto T, Hisanaga T, Fukai D, et al. Prognostic value of plasma soluble intercellular adhesion molecule-1 and endothelin-1 concentration in patients with chronic congestive heart failure. Am J Cardiol 1995;76:803–8. [7] Andreassen AK, Nordoy I, Simonsen S, et al. Levels of circulating adhesion molecules in congestive heart failure and after heart transplantation. Am J Cardiol 1998;81:604–8. [8] Devaux B, Scholz D, Hirche A, Klovekorn WP, Schaper J. Upregulation of cell adhesion molecules and the presence of low grade inflammation in human chronic heart failure. Eur Heart J 1997;18:470–9.
283
[9] Tousoulis D, Homaei H, Ahmed N, et al. Increased plasma adhesion molecules in patients with heart failure who suffered from ischemic and dilated cardiomyopathy. Am Heart J 2001;141:277–80. [10] Yin WH, Jen HL, Chiang MC, Huang WP, Feng AN, Young MS. Circulating soluble tumor necrosis factor-α and cell adhesion molecules in patients with acute cardiogenic pulmonary edema. J Emerg Crit Care Med 2004;15:9–19. [11] McGill SN, Ahmed NA, Christou NV. Endothelial cells: role in infection and inflammation. World J Surg 1998;22:171–8. [12] Carlos TM, Harlan JM. Leukocyte-endothelial adhesion molecules. Blood 1994;84:2068–101. [13] Wilhelmi MH, Leyh RG, Wilhelmi M, Haverich A. Upregulation of endothelial adhesion molecules in hearts with congestive and ischemic cardiomyopathy: immunohistochemical evaluation of inflammatory endothelial cell activation. Eur J Cardio-thorac Surg 2005;27:122–7. [14] Conraads VM, Bosmans JM, Schuerwegh AJ, et al. Intracellular monocyte cytokine production and CD 14 expression are up-regulated in severe vs mild chronic heart failure. J Heart Lung Transplant 2005;24:854–9. [15] Nathan CF. Neutrophil activation on biological surfaces. Massive secretion of hydrogen peroxide in response to products of macrophages and lymphocytes. J Clin Invest 1987;80:1550–60. [16] Johnson A, Phillips P, Hocking D, Tsan MF, Ferro T. Protein kinase inhibitor prevents pulmonary edema in response to H2O2. Am J Physiol 1989;256:H1012–22. [17] Geppert A, Zorn G, Heinz G, Huber K, Siostrzonek P. Soluble selectins in the pulmonary and systemic circulation in acute cardiogenic and noncardiogenic pulmonary failure. Intensive Care Med 2001;27:521–7. [18] Cotter G, Kaluski E, Moshkovitz Y, Milovanov O, Krakover R, Vered Z. Pulmonary edema: new insight on pathogenesis and treatment. Curr Opin Cardiol 2001;16:159–63. [19] Partridge CA, Horvath CJ, Del Vecchio PJ, Phillips PG, Malik AB. Influence of extracellular matrix in tumor necrosis factor-induced increase in endothelial permeability. Am J Physiol 1992;263:L627–33. [20] Drexler H, Hayoz D, Munzel T, et al. Endothelial function in chronic congestive heart failure. Am J Cardiol 1992;69:1596–601. [21] Tentolouris C, Tousoulis D, Antoniades C, et al. Endothelial function and proinflammatory cytokines in patients with ischemic heart disease and dilated cardiomyopathy. Int J Cardiol 2004;94:301–5. [22] Ferrari R, Bachetti T, Agnoletti L, Comini L, Curello S. Endothelial function and dysfunction in heart failure. Eur Heart J 1998;19(suppl G): G412–47. [23] Katz SD. The role of endothelium-derived vasoactive substances in the pathophysiology of exercise intolerance in patients with congestive heart failure. Prog Cardiovasc Dis 1995;38:23–50. [24] Berry C, Clark AL. Catabolism in chronic heart failure. Eur Heart J 2000;21:521–32. [25] Tsutamoto T, Wada A, Maeda K, et al. Angiotensin II type I receptor antagonist decreases plasma levels of tumor necrosis factor alpha, interleukin-6 and soluble adhesion molecules in patients with chronic heart failure. J Am Coll Cardiol 2000;35:714–21. [26] Drexler H, Kurz S, Jeserich M, Munzel T, Hornig B. Effect of chronic angiotensin-converting enzyme inhibition on endothelial function in patients with chronic heart failure. Am J Cardiol 1995;76:13E–8E. [27] Chen JW, Lin FY, Chen YH, Wu TC, Chen YL, Lin SJ. Carvedilol inhibits TNF-α-induced endothelial transcription factor activation, adhesion molecule expression and adhesiveness to human mononuclear cells. Arterioscler Thromb Vasc Biol 2004;24:2075–81. [28] Nakamura K, Kusano K, Nakamura Y, et al. Carvedilol decreases elevated oxidative stress in human failing myocardium. Circulation 2002;105:2867–71.
Increased endothelial monocyte adhesiveness is related to clinical outcomes in chronic heart failure Wei-Hsian Yin a,c,d , Jaw-Wen Chen b,d,⁎, Mason Shing Young a , Shing-Jong Lin b,c,d a
Division of Cardiology, Department of Internal Medicine, Cheng-Hsin Rehabilitation Medical Center, Taipei-Veterans General Hospital, Taiwan b Division of Cardiology, Taipei-Veterans General Hospital, Taiwan c Institute of Clinical Medicine, National Yang-Ming University, School of Medicine, Taipei, Taiwan d Cardiovascular Research Centre, National Yang-Ming University, School of Medicine, Taipei, Taiwan Received 30 May 2006; accepted 2 November 2006 Available online 22 December 2006
Abstract Background: Vascular inflammation and endothelial dysfunction are evident in patients with chronic heart failure (CHF). We hypothesized that circulating peripheral blood mononuclear cells (PBMCs) may be activated and the resultant increased endothelial monocyte adhesion may be functionally and pathophysiologically relevant in CHF. In the present study, we investigated the clinical significance of the activity of PBMCs in patients with CHF. Methods: PBMCs were isolated from 34 CHF patients, from 10 healthy volunteers (normal control group) and from 17 patients admitted for investigation of suspected coronary artery disease (disease control group). In each patient, the adhesiveness of PBMCs to cultured human aortic endothelial cells (HAECs) with or without tumor necrosis factor-alpha (TNF-α) stimulation was determined. Major adverse cardiac events (death, heart transplantation or hospitalization with worsening heart failure) were determined in the 34 CHF patients during a median follow-up period of 182 days. Results: Compared with those from both control groups and from mild CHF patients, PBMCs isolated from severe CHF patients adhered more to the HAECs. The endothelial adhesiveness of PBMCs correlated positively with the circulating levels of CAMs and can supply prognostic information in CHF patients. The difference between event-free curves based on the median levels of endothelial-PBMC adhesion was significant (log rank test, p = 0.0139). Conclusions: Endothelial adhesiveness of PBMCs is increased and correlated to clinical outcomes, and may be pathophysiologically relevant to the progression of CHF. © 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Chronic heart failure; Endothelial monocyte adhesiveness; Cell adhesion molecules; Prognosis
1. Introduction Recent studies suggest that immune activation and inflammation may play a role in the pathophysiology of chronic heart failure (CHF), which might have implications for the management of heart failure [1–4]. The main components of the immune system that are thought to be relevant to ⁎ Corresponding author. Division of Cardiology, Taipei-Veterans General Hospital, No. 201, Sec. 2, Shih-Pai Rd, Pei-Tou, Taipei 112, Taiwan. Tel.: +886 2 28712121x3793; fax: +886 2 28711601. E-mail address: [email protected] (J.-W. Chen). 0167-5273/$ - see front matter © 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijcard.2006.11.012
the pathophysiology of CHF include cytokines, cell adhesion molecules (CAMs) and nitric oxide [4]. It was reported by us and others that the activation of CAMs in endothelial cells and the circulating levels of soluble CAMs increase with the severity of CHF and are related to clinical outcomes in patients with CHF [5–10]. However, there is still debate as to whether CAMs are involved in the pathophysiology of CHF or merely markers of inflammatory response. In the development and progression of cardiovascular inflammation, a key step is the adherence of circulating mononuclear cells such as monocytes to the vascular endothelium through binding of highly regulated CAMs
W.-H. Yin et al. / International Journal of Cardiology 121 (2007) 276–283
expressed on the surface of monocytes or endothelial cells or both, followed by the subsequent extravasation of these mononuclear cells into the involved tissue [11,12]. Recent evidence further demonstrates that as compared to that in normal subjects, the expressions of CAMs such as endothelial-leukocyte adhesion molecule-1 (ELAM-1), intercellular cell adhesion molecule-1 (ICAM-1), vascular cell adhesion molecule-1 (VCAM-1) and integrin molecules can be enhanced within the myocardial microvasculature of patients with severe CHF [13], suggesting that vascular inflammation might be involved in the propagation and maintenance of cardiac deterioration in CHF even without ischemia in origin. Furthermore, circulating mononuclear cells such as monocytes could be activated and may play a significant role in the immunologic imbalance observed in advanced CHF [14]. In the present study, we tested the hypothesis that isolated peripheral blood mononuclear cells (PBMCs) from patients with CHF are activated and their adherence to cultured human aortic endothelial cells (HAECs) may be enhanced in vitro, and that the enhanced endothelial adhesiveness of PBMCs can supply prognostic information in these CHF patients. The novel findings of this study may provide a rationale for the clinical significance of mononuclear cell activation and vascular inflammation in the progression of CHF with both ischemic and non-ischemic origins. 2. Methods 2.1. Study population A total of 34 inpatients (28 men and 6 women, aged 63± 15 years) with chronic but relatively stable heart failure were enrolled from cardiology services at Cheng-Hsin Rehabilitation Medical Center. Men and women at least 18 years old with New York Heart Association (NYHA) functional class II to ambulatory class IV symptoms of CHF and a left ventricular ejection fraction (LVEF) of 40% or less by echocardiography within 7 days were eligible for enrolment. The 34 CHF patients were admitted for elective cardiac catheterization for evaluation of systolic dysfunction with unexplained causes despite non-invasive testing. The etiology of CHF was determined as ischemic when coronary angiography revealed N 70% luminal diameter narrowing in at least 1 major epicardial coronary artery. In those patients with CHF without coronary artery disease for whom the endomyocardial biopsy revealed findings compatible with dilated cardiomyopathy, the cause of CHF was determined to be dilated cardiomyopathy. The time intervals between blood sampling and admission were 1 ± 1 days. All necessary forms of conventional therapy were maintained throughout the study by the primary staff cardiologists. None of the selected patients were in frank heart failure, but all had symptoms of weakness and dyspnea at levels of activity varying from mild to a sedentary state. Patients were not allowed to participate if they had any
277
of the following occurring within two weeks before entry into the study: change in NYHA functional class, change in anti-failure medications, or administration of any intravenous medication for heart failure. Patients were excluded when they had hemodynamically significant obstructive valvular heart disease, cor pulmonale, restrictive or hypertrophic cardiomyopathy, myocarditis, constrictive pericarditis or congenital heart disease. Patients were excluded if there was severe co-morbidity, renal failure, or if there was clinical or laboratory evidence of systemic infection or an inflammatory illness such as sepsis, malignancy, arthritis or connective tissue disease. Those patients with non-steroid anti-inflammatory drugs, antioxidants, L-arginine, or lipid-lowering agents such as statins and fibrates were also excluded. Because it would have been ethically unfeasible to stop aspirin in patients with ischemic heart disease, the use of aspirin was determined by the attending physician. Two groups provided blood samples to serve as control subjects: the first comprised 10 healthy volunteers (normal control group); the second included 17 patients admitted for coronary angiography for investigation of suspected coronary artery disease and/or chest pain (disease control group). The following inclusion criteria were used for enrolment of the normal control group: negative family history of cardiovascular disease, nonsmoking status, and absence of hypercholesterolemia, hypertriglyceridemia and diabetes mellitus. On the basis of data from epidemiological and family studies, a cohort fulfilling these criteria is expected to have a very low prevalence of asymptomatic cardiovascular disease. Therefore, coronary angiography was not performed to rule out the presence of asymptomatic coronary artery disease because it would have been ethically unfeasible in these subjects. Baseline clinical evaluation, including a full medical history, physical examination, basic laboratory study and echocardiographic study, was performed on the 34 patients with CHF and on both control groups. No subject in either control group had any clinical signs or symptoms of heart failure and their LVEFs were all N 50% by echocardiography. All subjects gave informed consent, and the Human Investigation Committee of Cheng-Hsin Rehabilitation Medical Center approved the study protocol. 2.2. Cardiac catheterization and hemodynamic measurements At the time of enrolment, all 34 CHF patients and the 17 patients of the disease control group underwent cardiac catheterization or coronary arteriography for clinically indicated purposes. Left-sided cardiac catheterization, including left ventriculography and coronary arteriography, was performed in all 34 CHF patients and 17 disease control patients. Right-sided cardiac catheterization with or without endomyocardial biopsy was performed in 29 patients. Right heart pressure was measured with a 7Fr Swan-Ganz catheter, and cardiac output was measured by the thermodilution technique.
278
W.-H. Yin et al. / International Journal of Cardiology 121 (2007) 276–283
2.3. Culture of HAECs HAECs were isolated from the aortic portion of a single healthy donor, were provided as cryopreserved tertiary cultures by Cascade Biologics (Oregon, USA) and were grown in culture flasks in endothelial cell growth medium 200 (Cascade Biologics, USA) supplemented with fetal bovine serum (FBS, 2%), human epidermal growth factor (10 ng/mL), human fibroblast growth factor (3 ng/mL), heparin (10 μg/mL), penicillin (100 U/mL), streptomycin (100 pg/mL), and Fungizone (1.25 μg/mL). Cells were grown in a 5% CO2 humidified atmosphere at 37 °C and the growth medium was changed every other day until confluence. In this experiment, HAECs from second to eighth passage were used. Cells were identified positive for the von Willebrand factor and CD31 and negative for α-actin by immunohistochemistry. 2.4. Blood sampling and isolation of human PBMCs At the time of blood sampling, venous blood was drawn via central vein for those 29 patients who underwent right heart catheterization. For the other CHF patients and the control subjects, blood samples were collected from an indwelling catheter into vacuum tubes at bedside (peripheral vein) after ≧30 min of bed rest, with the patients in a supine position. PBMCs were isolated and extracted by density ultracentrifugation. In brief, 30 mL peripheral venous blood was drawn into a VACUTAINER® CPT™ Tube with 0.1 M sodium citrate (Becton Dickinson & Co., N.J., U.S.A.) at room temperature at bedside after 30 min of bed rest, with the study subjects in a supine position. The tubes were inverted 10 times gently. Then the blood was centrifuged and washed with EDTA/Hank's balanced salt solution (HBSS) solution (Sigma, St. Louis, Missouri, USA) as the operation instructions describe. The isolated PBMCs were then resuspended with serum-free RPMI1640 medium (M.A. Bioproducts, Walkersville, MD) for labeling. When the labeling procedure was completed, the RPMI1640 medium was removed by centrifugation and the cells were washed with EDTA/HBSS containing 2 mM Ca2+, 2 mM Mg2+, and 20 mM HEPES. The viability of the isolated PBMCs was assessed using the trypan blue exclusion. There was at least 95% viability in all samples used in these experiments. The isolation procedures were carried out at room temperature within 2 h of blood collection. After density ultracentrifugation, the separated plasma was frozen to −20 °C and stored at that temperature until analysis for circulating levels of CAMs. 2.5. Endothelial-PBMC adhesion assay For cell binding assay, HAECs (5 × 105 cells) were distributed into 6-well plates and were allowed to reach confluence. Before assay, the HAECs were incubated with or without 2 ng/mL TNF-α for 6 h at 37 °C. Isolated PBMCs from CHF patients and control subjects grown in RPMI 1640 medium containing 5% FBS were labeled for 45 min at
37 °C in darkness with 2′, 7′-bis (2-carboxyethyl)-5(6)carboxy-fluorescein acetoxymethyl ester (BCECF/AM, 10 μM, Boehringer-Mannheim, USA) in serum-free RPMI 1640 media, then the surplus was washed with HBSS to remove free dye and resuspended in RPMI 1640 containing 2% FBS. Labeled isolated PBMCs (3 × 105 cells) were added to each HAEC-containing well and incubation continued for 1 h at 37 °C. Non-adherent cells were removed by three gentle washes with HBSS, and the remaining cells were lysed with DMSO. Fluorescence was measured with an automated fluorometer (Multilabel Counter Victor2, Wallac, USA) at emission 530 nm and absorption 435 nm. 2.6. Measurement of circulating levels of CAMs In 32 of the CHF patients and the 10 normal subjects, assays for circulating VCAM-1, ICAM-1 and E-selectin were done concurrently to minimize any effects of repeated freeze–thaw cycles. The levels of VCAM-1, ICAM-1 and E-selectin were measured by means of enzyme-linked immunosorbent assay by commercial kits (R&D Systems, Inc., Minneapolis, MN for ICAM-1 and E-selectin and Biosource International, Camarillo, CA for VCAM-1). The intra-assay and interassay coefficients for all factors in our laboratory were approximately 5% and 10% respectively. 2.7. Clinical follow-up All CHF patients were observed through regular outpatient visits. Clinical information about major adverse cardiac events (death, requirement for heart transplantation or hospitalization with a primary diagnosis of worsening heart failure) during a median follow-up period of 182 days was provided by the cardiologists in charge without knowledge of the experimental results. 2.8. Data analysis All values are expressed as mean ± SEM. The patients with CHF were further divided into two groups based on their functional class: the mild CHF group (NYHA class II) and the severe CHF group (NYHA class III/IV). Univariate comparisons of clinical data, circulating levels of CAMs, and endothelial monocyte adhesiveness between two groups were made with the Student's t test for quantitative data and with the Fisher exact test for qualitative data. Comparisons between multiple groups were determined by means of a one-way analysis of variance (ANOVA) test. Linear regression analysis was used to determine the correlation between the endothelial monocyte adhesion and various circulating CAM levels. Kaplan–Meier analysis of cumulative event-free rates in CHF patients, stratified into two groups on the basis of median endothelial monocyte adhesion (expressed as % of control condition) were performed. The percentage of cell binding of the control condition (%) was determined by the fluorescent intensity of the attached cells to TNF-α-stimulated HAECs
W.-H. Yin et al. / International Journal of Cardiology 121 (2007) 276–283 Table 1 Baseline clinical characteristics of the study population Normal control (n = 10)
Disease control (n = 17)
Mild CHF (n = 15)
Severe CHF (n = 19)
45 ± 3 7(70) 65 ± 3 –
65 ± 3⁎ 12(71) 60 ± 2 13(76)
62 ± 4⁎ 12(80) 31 ± 2⁎# 3(20)#
64 ± 3⁎ 16(84) 24 ± 3⁎#§ 6(32)#
– –
12(70) 5(29)
9(60) 7(47)
12(63) 8(42)
– –
4(24) 3(18)
6(40) 2(13)
5(26) 2(10)
125 ± 4 68 ± 2 –
135 ± 5 71 ± 3 –
117 ± 5⁎#§ 95 ± 4⁎#§ 1889 ± 211
LVFP (mmHg)
–
18 ± 2
SVR (dyn s cm− 5) PVR (dyn s cm− 5) Medications, n (%) Diuretic Digitalis ACEI/ARB therapy Beta-blockers Nitrates Aspirin Monocyte adhesion (fluorescence) MACE rate (%) MACE, n (%) Death Heart transplantation Readmission for CHF
–
–
–
–
133 ± 5 85 ± 4⁎# 1883 ± 237 (n = 10) 26 ± 4# (n = 10) 1963 ± 250 (n = 10) 155 ± 26 (n = 10)
– – –
6(35) 1(6) 12(71)
13(87)# 8(53)# 10(67)
19(100)# 11(58)# 11(58)
– – – 2880 ± 987
9(53) 16(94) 12(70) 3809 ± 404
8(53) 10(67) 10(67) 3170 ± 535
10(53) 13(68) 12(63) 6893 ± 954⁎#§
–
–
40
53
– –
– –
0(0) 1(7)
4(21) 2(11)
–
–
5(33)
4(21)
Age (years) Male, n (%) LVEF (%) Ischemic heart disease (%) Hypertension (%) Diabetes mellitus (%) Smoking (%) Alcohol use (%) Hemodynamics SBP (mmHg) Heart rates (bpm) CI (L/min/m2)
30 ± 3# 2622 ± 291
There were more men than women in this sample. The average age of the CHF patients was 63 ± 15 years (48 to 80 years). The normal control group was significantly younger than the disease control group and the CHF population. However, the average ages were similar in the disease control group and the patients with mild and severe CHF. Fifteen patients were in NYHA class II (mild CHF group) and 19 were in class III/IV (severe CHF group). The cause of CHF was ischemic cardiomyopathy in 9 and dilated cardiomyopathy in 25. All of the 25 patients with dilated cardiomyopathy were proven by cardiac catheterization and endomyocardial biopsy. The incidence of ischemic heart disease was significantly higher in the disease control group than in the normal control group and the mild and severe CHF groups. However, the incidence of ischemic heart disease was similar in patients with mild and severe CHF. No Table 2 Patient characteristics and hemodynamic data in the patients with chronic heart failure who had major adverse cardiac events during follow-up and those who were event-free (n = 34)
228 ± 94
ACEI/ARB= angiotensin converting enzyme inhibitors or angiotensin II receptor blockers; CI = cardiac index; CHF = chronic heart failure; LVFP= left ventricular filling pressure; LVEF = left ventricular ejection fraction; MACE= major adverse cardiac event; PVR = pulmonary vascular resistance; SVR = systemic vascular resistance; SBP = systolic blood pressure. ⁎p b 0.01 vs. normal subjects, #p b 0.01 vs. disease controls, §p b 0. 05 vs. patients with mild CHF.
divided by fluorescent intensity of the control condition (PBMC binding to HAECs without TNF-α stimulation),×100. The differences in event-free curves were tested with a log rank test. A p value of less than 0.05 was considered statistically significant. 3. Results 3.1. Demographic and clinical features of the study population The demographic and clinical characteristics of the study population are shown in Table 1.
279
Age (years) Male, n (%) LVEF (%) Functional class Hemodynamics SBP (mmHg) Heart rates (bpm) CI (L/min/m2) LVFP (mmHg) SVR (dyn s cm− 5) PVR (dyn s cm− 5) Medication, n (%) Diuretic Digitalis ACEI/ARB therapy Beta-blockers Nitrates VCAM-1 (ng/ml) ICAM-1 (ng/ml) E-selectin (ng/ml) Endothelial adhesiveness of PBMC (fluorescence) Endothelial adhesiveness of PBMC (% of control) Follow-up period (days)
MACE (−) (n = 18)
MACE (+) (n = 16)
p value
59 ± 4 14 (78%) 29 ± 3 2.6 ± 0.2
68 ± 3 14 (88%) 26 ± 3 3.1 ± 0.2
NS NS NS NS
131 ± 5 84 ± 3 2160 ± 285 (n = 15) 24 ± 3 (n = 15) 1674 ± 314 (n = 15) 126 ± 34 (n = 15)
117 ± 6 97 ± 5 1750 ± 183 (n = 14) 29 ± 3 (n = 14) 2457 ± 225 (n = 14) 209 ± 52 (n = 14)
0.04 0.03 NS
16 (89) 10 (56) 12 (67) 12(67) 10(56) 442.0 ± 46.1 (n = 17) 180.8 ± 27.4 (n = 17) 33.3 ± 5.2 (n = 17) 3814 ± 715
16 (100) 9 (56) 9 (56) 6(38) 13(81) 546.8 ± 65.5 (n = 15) 276.1 ± 22.8 (n = 15) 64.3 ± 9.7 (n = 15) 6868 ± 1026
NS NS NS NS NS NS
0.018
239 ± 32
448 ± 43
b0.001
192 ± 28
80 ± 24
0.005
0.01 NS NS
0.022 0.049
ACEI/ARB = angiotensin converting enzyme inhibitors or angiotensin II receptor blockers; CI = cardiac index; ICAM-1 = intercellular adhesion molecule-1; LVFP = left ventricular filling pressure; LVEF = left ventricular ejection fraction; MACE = major adverse cardiac event; PBMC = peripheral blood mononuclear cells; PVR = pulmonary vascular resistance; SVR = systemic vascular resistance; SBP = systolic blood pressure; VCAM-1 = vascular cellular adhesion molecule-1.
280
W.-H. Yin et al. / International Journal of Cardiology 121 (2007) 276–283
significant differences in sex, incidence of hypertension and diabetes mellitus, smoking status or alcohol use were detected among the disease control subjects and patients with mild and severe chronic CHF. On blood sampling, all 34 patients were clinically stable, while receiving continuous oral medications for CHF. Nineteen patients used digoxin and 32 used diuretics; 21 patients were also treated with angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, 18 with beta-blockers and 23 with nitrates. Patients with CHF used more diuretics and digitalis than did disease control subjects. However, the use of aspirin, angiotensin-converting enzyme inhibitor or angiotensin receptor blocker, nitrate and beta-blocker therapy were similar in disease control subjects and CHF patients. 3.2. Left ventricular ejection fraction and hemodynamic measurements The mean LVEF was significantly lower and the left ventricular filling pressure was significantly higher in the CHF patients than in the control subjects. The mean systolic blood pressure was significantly lower and the heart rate was significantly higher in the severe CHF group than in the mild CHF group and the control subjects. Furthermore, the mean LVEF was significantly lower in the severe CHF group than in the mild CHF group. The CHF patients were then divided into those who had major adverse cardiac events during follow-up and those who were event-free (Table 2). No significant differences in age, sex, mean LVEF, functional status, or uses of medications for heart failure were detected between the two groups. However, the hemodynamic parameters, such as systolic blood pressure, heart rates, and left ventricular filling pressure, were worse in the group with major adverse cardiac events than in the event-free group.
3.3. Endothelial-PBMC adhesion in CHF patients and control subjects We chose TNF-α to stimulate HAECs, and under our experimental conditions, TNF-α (2 ng/mL) induced a 2- to 4-fold increase in the binding of PBMCs to HAECs both in patients with CHF and in control subjects. When equal numbers of labeled isolated PBMCs were added to HAEC-containing wells, PBMCs isolated from severe CHF patients adhered more to cultured HAECs than those from normal and disease control subjects and from mild CHF patients, whether or not the HAECs had been preincubated with TNF-α. Fig. 1 shows the relation between the stage and degree of CHF and the PBMC adhesion, expressed as fluorescence detected by fluorometer. In the present study, we enrolled 17 patients admitted for investigation of suspected coronary artery disease and/or chest pain as disease control subjects. The incidence of ischemic heart disease was significantly higher in this group than in the mild and severe CHF groups. Because the patients with ischemic heart disease may have higher endothelial monocyte adhesiveness, the question of whether ischemic heart disease accounts for the enhanced monocyte adhesion in this study and affects the clinical outcomes is a major concern. To answer this question, we divided the CHF patients into two groups based on their underlying causes of CHF (ischemic vs. non-ischemic) and compared the levels of endothelial monocyte adhesion between the two groups by the Wilcoxon rank-sum test. We found that the endothelial monocyte adhesiveness were similar in the patients with ischemic and non-ischemic heart failure whether or not the HAECs had been preincubated with TNF-α (1576 ± 323 vs. 2046 ± 363, p = 0.468 at baseline and 6666 ± 1205 vs. 4742 ± 773, p = 0.203 after TNF-α stimulation, respectively).
Fig. 1. Comparison between the clinical severity of heart failure and endothelial adhesiveness of peripheral blood mononuclear cells (PBMCs), expressed as fluorescence detected by fluorometer. The binding of PBMCs to either unstimulated or TNF-α stimulated human aortic endothelial cells (HAECs) was significantly increased in patients with severe chronic heart failure (CHF). TNF-α: TNF-α stimulated HAECs. ⁎p b 0.001 vs. normal subjects, ⁎⁎p b 0.001 vs. disease controls, ⁎⁎⁎p b 0.001 vs. patients with mild CHF.
W.-H. Yin et al. / International Journal of Cardiology 121 (2007) 276–283
281
3.4. Correlation between endothelial-PBMC adhesion and serum levels of CAMs The circulating levels of all the inflammatory markers were significantly higher in the patients with CHF than in the 10 normal subjects (data not shown). Significant positive correlations, with the correlation coefficients between PBMCs adhesion and circulating levels of VCAM-1, ICAM-1 and E-selectin of 0.665 (p b 0.0001), 0.398 (p = 0.0242) and 0.418 (p = 0.0173) respectively, were noted (Fig. 2). 3.5. Endothelial-PBMC adhesion and prognosis Although the heart failure condition of the study population was relatively stable instead of being acutely decompensated, the event rate during follow-up was relatively high (16 of 34; 47%). The median follow-up period for the whole group was 182 days (128–288 days, 25th–75th percentiles); the median follow-up periods for the group with major adverse cardiac events versus the eventfree group were 80 days and 192 days respectively (Table 2). Four of the 34 patients died of cardiac causes (2 died of sudden death without premonition of the progression of symptoms, presumed to be due to arrhythmia, and 2 of intractable end-stage heart failure) during the follow-up period; all of them were in the severe CHF group. Three patients underwent heart transplantation (one in the mild CHF group and 2 of them in the severe CHF group) and 9 patients were readmitted for worsening heart failure (5 of them in the mild CHF group and 4 of them in the severe CHF group) (Table 1).
Fig. 2. The correlations between endothelial adhesiveness of peripheral blood mononuclear cells (PBMCs) and serum levels of adhesion molecules. Endothelial adhesiveness of PBMCs was significantly correlated to circulating levels of VCAM-1, ICAM-1 and E-selectin.
As medications such as non-steroid anti-inflammatory drugs, antioxidants, lipid-lowering agents such as statins, fibrates and so on may inhibit endothelial monocyte adhesion, patients with those medications were excluded from this study. We did not exclude patients taking aspirin, but the use of aspirin was similar in the disease control group and the CHF population.
Fig. 3. Kaplan–Meier event probability for patients with heart failure stratified in 2 groups on the basis of median endothelial adhesiveness of peripheral blood mononuclear cells (PBMCs) (expressed as % of control). Patients with more increased (N272.141%) endothelial adhesiveness of PBMCs had worse clinical outcomes than those with less increased (b272.141%) endothelial adhesiveness of PBMCs (p = 0.0139).
282
W.-H. Yin et al. / International Journal of Cardiology 121 (2007) 276–283
We divided the 34 patients with CHF into two groups based on those who had major adverse cardiac events during follow-up versus those who were event-free (Table 2). The patients who had adverse events had significantly higher ICAM-1 and E-selectin levels and endothelial adhesiveness of PBMCs (p = 0.022, p = 0.049, and p b 0.0001, respectively) than the patients who were event-free. We then performed Kaplan–Meier analyses of event-free survival in the CHF patients on the basis of median endothelial monocyte adhesion (expressed as % of control). The difference between event-free curves based on the median levels of endothelial-PBMC adhesion was significant (log rank test, p = 0.0139) (Fig. 3). 4. Discussion 4.1. Comparison with previous studies of monocyte adhesion in CHF It has been shown that CAMs are upregulated in failing myocardium [10] as well as in the myocardial microvasculature [13] and that circulating CAM levels are increased in CHF patients [5–10]. We and others have also reported that the circulating levels of soluble CAMs increase with the severity of CHF and are related to clinical outcomes in patients with CHF [5,6,9,10]. The present study represents an addition to the growing literature of evidence of inflammation in patients with CHF. The main finding is that PBMCs from severe CHF patients demonstrate a greater adhesiveness to cultured HAECs than do cells from controls and mild CHF patients. The results are consistent with the hypothesis that the endothelial monocyte adhesiveness, an index of cell activity, is increased and appears to have predictive value for combined endpoint of death, transplant and heart failure readmission. It is suggested that vascular inflammation could be present and the presence of more sticky effector cells may contribute to the clinical progression of CHF with both ischemic and non-ischemic origins. To the best of our knowledge, these findings have not been reported before. 4.2. Potential role of monocyte adhesion in the pathophysiology of CHF Patients with advanced CHF have elevated circulating levels of proinflammatory cytokines and CAMs, irrespective of the cause of heart failure [1–10]. Endothelial cell activation by cytokines or other inflammatory mediators results in the expression of CAMs such as VCAM-1, ICAM1 and E-selectin, which in turn facilitates the attachment of blood leukocytes to endothelial cells [4,11–13]. The adherent cells are capable of producing greater amounts of enzymes (such as elastase) and toxic oxygen metabolites than nonadherent cells are, resulting in injury to the endothelium and transmigration of leukocytes [15]. The activated endothelial cells may alter intercellular gap
formation owing to disruption of cellular junctions [16]. Both damaged vascular integrity and impaired endothelial barrier function may increase endothelial permeability and transport of proteins and liquid via the paracellular pathway, resulting in endothelial dysfunction and tissue edema [17–19]. Furthermore, the leukocytes residing in the intima may produce a number of cytokines and growth factors, which may result in endothelial dysfunction, impairment of coronary vascular reserve and myocardial tissue perfusion, apoptosis of cardiomyocytes and endothelial cells, inhibition of myocyte contractility and myocardial fibrosis — a multitude of changes leading to progressive cardiac dysfunction and worsening heart failure [4]. Therefore, CAMs induced by inflammatory cytokines and the subsequent enhanced endothelial adhesion of PBMCs may play a direct role in the progression of CHF [13]. 4.3. Clinical implications Congestive heart failure is characterized by impaired cardiac performance and neurohormonal imbalance. Recent studies have also shown that vascular inflammation, endothelial activation and dysfunction are evident in CHF, as a result of increased oxidative stress status, reduced blood flow and decreased shear stress stimuli [1–4,20–22]. Even though the ultimate role of inflammation and endothelial dysfunction in the pathophysiology of the syndrome of heart failure is not yet fully understood, abnormalities of endothelial function may contribute to impaired myocardial perfusion due to myocardial microvascular inflammation [13] and the increased peripheral vasomotor tone both at rest and during exercise [17–19,23,24]. Both may be directly relevant to the pathophysiology of dyspnea, fatigue, exercise intolerance, cachexia and interstitial edema in CHF. Therefore, for therapeutic purposes, it could be one of the targets of anti-failure therapy to improve endothelial function and to inhibit vascular inflammation by preventing the expression of inflammatory molecules such as TNF-α and CAMs. This hypothesis is consistent with the fact that, while significantly improving long-term prognosis, treatment with angiotensin-converting enzyme inhibitors, angiotensin receptor blocker and carvedilol have been shown to decrease the plasma levels of TNF-α and CAMs and reduce endothelial inflammation in CHF patients [25–27]. 4.4. Study limitations One of the limitations of the study was the relatively small number of CHF patients included in each group. Even though significant correlation of the endothelial adhesiveness of PBMCs to prognosis was observed, the outcome data are rather limited and should be considered hypothesisgenerating rather than conclusive. Furthermore, the study did not investigate the molecular mechanisms of the enhanced endothelial adhesiveness of PBMCs in patients with CHF. Nor does it test the possible inhibitory effects of anti-failure
W.-H. Yin et al. / International Journal of Cardiology 121 (2007) 276–283
drugs on adhesiveness of PBMCs to endothelial cells in vitro. However, we have recently demonstrated that carvedilol, a medication currently widely used for clinical CHF, can directly, through its antioxidant property, inhibit the in vitro expression of VCAM-1 and the adhesiveness of human PBMCs to cultured endothelial cells stimulated by TNF-α [27]. While the cardiac protection of carvedilol in CHF is largely attributed to its antioxidant capability [28], it would be interesting to investigate further the potential effect of antioxidants and their related anti-inflammatory mechanisms on both myocardial microvascular and peripheral vascular protection regarding the progression of CHF. 5. Conclusions The adhesiveness of human PBMCs to endothelial cells is increased with the severity of CHF and correlated to clinical outcomes in patients with CHF. This is consistent with the hypothesis that increased expression of CAMs and enhanced PBMC adhesion may be functionally and pathophysiologically relevant. Acknowledgments The present study was supported in part by Cheng-Hsin Rehabilitation Medical Center grant 93–06 (to W.H.Y. and S.J.L.) and in part by Taipei-Veterans General Hospital grant VGH-91–263 (to J.W.C.). We also thank Ms. Chin-Feng Cheng for her excellent assistance in preparing and typing the manuscript. References [1] Seta Y, Shan K, Bozdurt B, Oral H, Mann DL. Basic mechanisms in heart failure: the cytokine hypothesis. J Card Fail 1996;2:243–9. [2] Mann DL. Activation of inflammatory mediators in heart failure. In: Mann DL, editor. Ed. Heart Failure: A Companion to Braunwald's Heart Disease. 1st ed. Philadelphia: Saunders; 2004. p. 159–80. [3] Hasper D, Hummel M, Kleber FX, Reindl I, Volk HD. Systemic inflammation in patients with heart failure. Eur Heart J 1998;19:761–5. [4] Sharma R, Coats AJS, Anker SD. The role of inflammatory mediators in chronic heart failure: cytokines, nitric oxide, and endothelin-1. Int J Cardiol 2000;72:175–86. [5] Yin WH, Chen JW, Jen HL, et al. The prognostic value of circulating soluble cell adhesion molecules in patients with chronic congestive heart failure. Eur J Heart Fail 2003;5:507–16. [6] Tsutamoto T, Hisanaga T, Fukai D, et al. Prognostic value of plasma soluble intercellular adhesion molecule-1 and endothelin-1 concentration in patients with chronic congestive heart failure. Am J Cardiol 1995;76:803–8. [7] Andreassen AK, Nordoy I, Simonsen S, et al. Levels of circulating adhesion molecules in congestive heart failure and after heart transplantation. Am J Cardiol 1998;81:604–8. [8] Devaux B, Scholz D, Hirche A, Klovekorn WP, Schaper J. Upregulation of cell adhesion molecules and the presence of low grade inflammation in human chronic heart failure. Eur Heart J 1997;18:470–9.
283
[9] Tousoulis D, Homaei H, Ahmed N, et al. Increased plasma adhesion molecules in patients with heart failure who suffered from ischemic and dilated cardiomyopathy. Am Heart J 2001;141:277–80. [10] Yin WH, Jen HL, Chiang MC, Huang WP, Feng AN, Young MS. Circulating soluble tumor necrosis factor-α and cell adhesion molecules in patients with acute cardiogenic pulmonary edema. J Emerg Crit Care Med 2004;15:9–19. [11] McGill SN, Ahmed NA, Christou NV. Endothelial cells: role in infection and inflammation. World J Surg 1998;22:171–8. [12] Carlos TM, Harlan JM. Leukocyte-endothelial adhesion molecules. Blood 1994;84:2068–101. [13] Wilhelmi MH, Leyh RG, Wilhelmi M, Haverich A. Upregulation of endothelial adhesion molecules in hearts with congestive and ischemic cardiomyopathy: immunohistochemical evaluation of inflammatory endothelial cell activation. Eur J Cardio-thorac Surg 2005;27:122–7. [14] Conraads VM, Bosmans JM, Schuerwegh AJ, et al. Intracellular monocyte cytokine production and CD 14 expression are up-regulated in severe vs mild chronic heart failure. J Heart Lung Transplant 2005;24:854–9. [15] Nathan CF. Neutrophil activation on biological surfaces. Massive secretion of hydrogen peroxide in response to products of macrophages and lymphocytes. J Clin Invest 1987;80:1550–60. [16] Johnson A, Phillips P, Hocking D, Tsan MF, Ferro T. Protein kinase inhibitor prevents pulmonary edema in response to H2O2. Am J Physiol 1989;256:H1012–22. [17] Geppert A, Zorn G, Heinz G, Huber K, Siostrzonek P. Soluble selectins in the pulmonary and systemic circulation in acute cardiogenic and noncardiogenic pulmonary failure. Intensive Care Med 2001;27:521–7. [18] Cotter G, Kaluski E, Moshkovitz Y, Milovanov O, Krakover R, Vered Z. Pulmonary edema: new insight on pathogenesis and treatment. Curr Opin Cardiol 2001;16:159–63. [19] Partridge CA, Horvath CJ, Del Vecchio PJ, Phillips PG, Malik AB. Influence of extracellular matrix in tumor necrosis factor-induced increase in endothelial permeability. Am J Physiol 1992;263:L627–33. [20] Drexler H, Hayoz D, Munzel T, et al. Endothelial function in chronic congestive heart failure. Am J Cardiol 1992;69:1596–601. [21] Tentolouris C, Tousoulis D, Antoniades C, et al. Endothelial function and proinflammatory cytokines in patients with ischemic heart disease and dilated cardiomyopathy. Int J Cardiol 2004;94:301–5. [22] Ferrari R, Bachetti T, Agnoletti L, Comini L, Curello S. Endothelial function and dysfunction in heart failure. Eur Heart J 1998;19(suppl G): G412–47. [23] Katz SD. The role of endothelium-derived vasoactive substances in the pathophysiology of exercise intolerance in patients with congestive heart failure. Prog Cardiovasc Dis 1995;38:23–50. [24] Berry C, Clark AL. Catabolism in chronic heart failure. Eur Heart J 2000;21:521–32. [25] Tsutamoto T, Wada A, Maeda K, et al. Angiotensin II type I receptor antagonist decreases plasma levels of tumor necrosis factor alpha, interleukin-6 and soluble adhesion molecules in patients with chronic heart failure. J Am Coll Cardiol 2000;35:714–21. [26] Drexler H, Kurz S, Jeserich M, Munzel T, Hornig B. Effect of chronic angiotensin-converting enzyme inhibition on endothelial function in patients with chronic heart failure. Am J Cardiol 1995;76:13E–8E. [27] Chen JW, Lin FY, Chen YH, Wu TC, Chen YL, Lin SJ. Carvedilol inhibits TNF-α-induced endothelial transcription factor activation, adhesion molecule expression and adhesiveness to human mononuclear cells. Arterioscler Thromb Vasc Biol 2004;24:2075–81. [28] Nakamura K, Kusano K, Nakamura Y, et al. Carvedilol decreases elevated oxidative stress in human failing myocardium. Circulation 2002;105:2867–71.