The European Journal of Heart Failure 6 (2004) Philipp Bahrmann*, Uta M. Hengst, Babara M. Richartz, Hans R. Figulla

Transcription

1 The European Journal of Heart Failure 6 (2004) Pentoxifylline in ischemic, hypertensive and idiopathic-dilated cardiomyopathy: effects on left-ventricular function, inflammatory cytokines and symptoms Philipp Bahrmann*, Uta M. Hengst, Babara M. Richartz, Hans R. Figulla Abstract Clinic of Internal Medicine I, Friedrich-Schiller-University, Erlanger Allee 101, Jena, Germany Received 1 November 2002; received in revised form 16 May 2003; accepted 15 September 2003 Introduction: Tumor necrosis factor (TNF)-a and interleukin-6 (IL-6) are significantly elevated in patients with congestive heart failure (CHF). Pentoxifylline, a xanthin-derived agent, is known to inhibit the production of TNF-a and IL-6. Recent studies have shown that pentoxifylline produces an increase in ejection fraction, a decrease in left-ventricular chamber size and an improvement in clinical status in patients with idiopathic-dilated cardiomyopathy. Therefore, we studied the effects of pentoxifylline in ischemic, hypertensive and idiopathic-dilated cardiomyopathy. Methods: Primary endpoint was left-ventricular ejection fraction (LVEF) assessed by contrast 2D echocardiography. Secondary endpoints were concentrations of TNF-a, IL-6, brain natriuretic peptide, maximal oxygen uptake (VO ) assessed by cardiopulmonary exercise testing and Minnesota Living with Heart Failure Questionnaire score or New York Heart Association scale. Results: Forty-seven patients (31.9% ischemic, 21.3% hypertensive, 10.6% ischemic and hypertensive, 36.2% idiopathic-dilated cardiomyopathy) were randomly assigned to pentoxifylline 600 mg BID (ns23) or placebo (ns24) if they had a compensated CHF with a LVEF less than or equal to 40% and had taken their standard treatment consisting of angiotensin-converting enzyme inhibitors, diuretics and b-blockers for at least 3 months. Baseline demographic and clinical characteristics of each group were similar. Forty-one patients completed the study protocol and were analysed for primary and secondary endpoints. After 6 months of treatment, LVEF was unchanged in the pentoxifylline group compared with placebo (29"7 to33"10% vs. 27"9 to34"9%, respectively, PsNS). Also the secondary endpoints did not significantly change during follow-up. Conclusion: Additional treatment with pentoxifylline is neutral with regard to left-ventricular function, inflammatory cytokines and symptoms in patients with ischemic, hypertensive and idiopathic-dilated cardiomyopathy European Society of Cardiology. Published by Elsevier B.V. All rights reserved. Keywords: 1. Introduction Cardiomyopathy; Heart failure treatment; Ejection fraction; Cytokines; Pentoxifylline Because neuroendocrine pathophysiological changes, especially of the renin angiotensin aldosterone system and sympathetic nervous system, play an important role in the progression of congestive heart failure (CHF), angiotensin-converting enzyme (ACE) inhibitors and b- adrenergic blockers are routinely used for therapy. Despite optimal inhibition of the neuroendocrine systems *Corresponding author. Tel.: q ; fax: q address: philipp.bahrmann@med.unijena.de (P. Bahrmann). by ACE inhibitors and b-adrenergic blockers, leftventricular function may deteriorate over time w1x. Other mechanisms contributing to cardiac dysfunction may be involved such as activation of various cytokines. Numerous experimental and clinical studies have shown that higher levels of cytokines are found in the circulation and in the myocardium of patients with CHF than in controls. Pro-inflammatory cytokines like tumor necrosis factor (TNF)-a and interleukin-6 (IL-6) might be involved in the progression of CHF, inducing cardiac dysfunction, apoptosis, ventricular remodeling and dilatation w2 9x. In consequence, studies have been undertaken with TNF antagonists. In a short-term pilot study, the soluble TNF receptor etanercept produced dose /04/$ European Society of Cardiology. Published by Elsevier B.V. All rights reserved. doi: /j.ejheart

2 196 P. Bahrmann et al. / The European Journal of Heart Failure 6 (2004) dependent increases in ejection fraction, decreases in left-ventricular chamber size and improvement in clinical status w10,11x. However, large-scale trials (RENAIS- SANCE, RECOVER and RENEWAL) with etanercept in CHF were stopped early because of the low likelihood that the drug would show favourable effects. One possible explanation may be that the membrane-bound myocardial rather than the circulating TNF-a has an unfavourable effect on cardiac function w12x. But the membrane-bound myocardial TNF-a is not in range of the soluble TNF receptor etanercept. Pentoxifylline, a xanthin-derived agent, is known to decrease the level of inflammatory mediators through transcriptional blockade of inflammatory gene expression w9x. Consequently, the production of mediators like TNF-a, IL-6 and FasyApo-1 is reduced w13 15x. Recent studies have shown that pentoxifylline produces an increase in ejection fraction, a decrease in left-ventricular chamber size and an improvement in clinical status in patients with idiopathic-dilated and ischemic cardiomyopathy w16 20x. Therefore, we studied the effects of pentoxifylline on left-ventricular function, inflammatory cytokines and symptoms in ischemic, hypertensive and idiopathic-dilated cardiomyopathy. 2. Methods 2.1. Patients Forty-seven patients with compensated CHF were included in a single centre, prospective, double-blind, randomised, placebo-controlled trial between January 1999 and May The Institutional Ethics Committee approved the study protocol and all patients gave written, informed consent. The investigation conformed with the principles outlined in the Declaration of Helsinki. Inclusion criteria were (1) age between 18 and 70 years, (2) stable New York Heart Association (NYHA) class II or III CHF due to ischemic and hypertensive cardiomyopathy or idiopathic-dilated cardiomyopathy, (3) leftventricular ejection fraction (LVEF) less or equal than 40% assessed by contrast 2D echocardiography and (4) sinus rhythm. Exclusion criteria were (1) chronic obstructive pulmonary disease, (2) significant valvular heart disease, (3) disorders other than cardiomyopathy that could increase TNF-a and IL-6 concentrations (i.e. rheumatoid arthritis, sepsis, vasculitis, autoimmune disorders), (4) pregnancy, (5) severe liver disease (defined as enzymes more than twice the normal upper limit), (6) acute myocardial infarction, (7) haemorrhage and (8) any clinical condition judged by the investigators to prevent inclusion in the study. The diagnosis of cardiomyopathy was based on the patient s medical history, echocardiography, coronary angiography and ventriculography. All patients had taken their standard treatment consisting of ACE inhibitors, diuretics and b-blockers for at least 3 months prior to inclusion. All patients were randomly assigned to pentoxifylline 600 mg BID (ns23) or placebo (ns24). Clinical examination, contrast 2D echocardiography, cardiopulmonary exercise testing and TNF-a, IL-6 and brain natriuretic peptide (BNP) plasma level measurements were made at the beginning and after 6 months. After 3 months, a meeting was scheduled for clinical evaluation of compliance and adverse events. Primary endpoint was LVEF assessed by contrast 2D echocardiography. Secondary endpoints were concentrations of TNF-a, IL-6, BNP, VOassessed by cardiopulmonary exercise testing and Minnesota Living with Heart Failure Questionnaire (LHFQ) scores or NYHA class. We estimated the patient compliance by counting the number of returned pills. In our study, an average of 92% of pills were taken by patients with ischemic and hypertensive cardiomyopathy Chemokine and hormone measurement Blood samples for the measurement of TNF-a, IL-6 and BNP were collected at baseline and after 6 months. Following venepuncture of the antecubital vein, plasma was collected on ice into serum separator tubes and within 30 min of collection centrifuged at approximately 1000=g for 10 min. Samples were aliquoted and stored at y80 8C. For the quantitative determination of human TNF-a and IL-6 concentrations in plasma, we used Quantikine HSimmunoassays (R&D Systems, Minneapolis, USA). For the quantitative determination of BNP concentrations in plasma we used the Triage point-ofcare assay system (Biosite Diagnostics, San Diego, USA). A few drops of blood were added to a small disposable test device and the device was inserted into the Triage Meter, a hand-held battery powered fluorometer Functional class and exercise test For the assessment of LHFQ scores in patients with ischemic and hypertensive cardiomyopathy, we used a standardised questionnaire w21x. NYHA class in patients with idiopathic-dilated cardiomyopathy was recorded by a physician who was unaware of the treatment assigned. All patients performed incremental exercise tests on a cycle ergometer. In this protocol, measurements of gas exchange were made at rest, while the work rate was increased by 25 W every 2 min until the patient was limited by symptoms or was unable to continue safely and during a post-exercise period of 4 min. Simultaneously, the attending physician checked blood pressure and ECG recordings. VO2 maxwas determined according to the principles of exercise testing and interpretation by Wasserman w22x.

3 P. Bahrmann et al. / The European Journal of Heart Failure 6 (2004) Echocardiographic studies All echocardiographic studies were performed and analysed by one person blinded to treatment assigned. Patients were studied supine in partial left lateral position using one echocardiographic machine (Power- Vision 8000, Toshiba, Tokyo, Japan). For all contrast studies, 10 ml of saline with 0.2 ml of microbubbles (Levovist, Schering, Berlin, Germany) was injected rapidly through a 20G cannula in the antecubital vein. Two-dimensional imaging was performed with gains and dynamic range optimized for microbubbles detection. The images were acquired in a digital format and stored. The measurements were manually traced on the screen and the built-in contour detection algorithm assessed LVEF. It was repeated 3 times, and the average was used for further calculations Statistical analysis Treatment effects between the groups were compared by the Mann Whitney U-test. Changes in the quantitative parameters at baseline compared to the results at 6 months within each treatment group were assessed using the Wilcoxon-matched pairs test. All probability values were calculated two sided. Data were expressed as mean"standard deviation. All calculations were done with SPSS for Windows (Version , SPSS, Chicago, USA). 3. Results 3.1. Baseline characteristics In total, 47 patients (mean age 57"12 years, 3 women, 44 men) were randomly allocated to pentoxifylline 600 mg BID (ns23) or placebo (ns24). The cause of CHF in our study was in 31.9% ischemic, in 21.3% hypertensive, in 10.6% ischemic and hypertensive cardiomyopathy and in 36.2% idiopathic-dilated cardiomyopathy. Baseline characteristics of the study population were similar (Table 1). Forty-one patients (20 placebo, 21 pentoxifylline) completed the study protocol and were analysed for primary and secondary outcome (Table 2). Of the six patients who did not complete the study protocol, one patient died of gastrointestinal cancer (placebo), three patients complained about adverse events like bleeding (placebo) or side-effects like nausea or insomnia (placebo, pentoxifylline) and two patients failed to complete the course of study (placebo, pentoxifylline). Of 41 patients, 26 had ischemic and hypertensive cardiomyopathy and 15 idiopathic-dilated cardiomyopathy. Results for patients with ischemic and hypertensive cardiomyopathy are listed in Table 3 and for patients with idiopathic-dilated cardiomyopathy in Table 4. Table 1 Baseline characteristics of the study population Placebo Pentoxifylline P (ns24) (ns23) Age (years) 58"12 55"12 NS BMI (kgym ) 28"5 28"3 NS Gender (%) Male 24 (100) 20 (87) NS Female 3 (13) Functional class LHFQ score 36"15 26"15 NS NYHA scale 2.6" "0.3 NS Cause of HF (%) Ischemic 8 (33) 7 (30) NS Hypertensive 4 (17) 6 (26) NS Ischemic and hypertensive 3 (12) 2 (9) NS Idiopathic-dilated 9 (38) 8 (35) NS Treatment (a) AT antagonists (%) 2 (9) Lorsatan (mg) 1 (100) Irbesatan (mg) 1 (150) (b) ACE-inhibitors (%) 24 (100) 22 (96) NS Enalapril (mg) 3 (15"0) Cilazapril (mg) 1 (5) Ramipril (mg) 10 (9"1) 12 (9"3) NS Benazepril (mg) 3 (23"6) 3 (18"3) NS Perindopril (mg) 7 (5"2) 5 (4"2) NS Lisinopril (mg) 1 (20) 1 (20) NS (c) Diuretics (%) 24 (100) 23 (100) NS Furosemid (mg) 3 (60"20) 2 (40"0) NS Torasemid (mg) 9 (11"4) 8 (11"4) NS Xipamid (mg) 12 (31"12) 10 (32"10) NS Piretanid (mg) 1 (6) Hydrochlorothiazid (mg) 5 (24"3) 6 (19"7) NS Spironolactone (mg) 11 (25"0) 9 (28"8) NS (d) Digitalis (%) 12 (50) 12 (52) NS Digitoxin (mg) 12 (0.07"0.01) 10 (0.07"0.01) NS Digoxin (mg) 2 (0.23"0.04) (e)b-blockers (%) 24 (100) 21 (91) NS Carvedilol (mg) 6 (46"21) 4 (50"20) NS Bisoprolol (mg) 16 (6"2) 12 (6"3) NS Metoprolol (mg) 1 (100) 4 (107"60) NS Atenolol (mg) 1 (50) Sotalol (mg) 1 (400) (f) CSE-inhibitors (%) 11 (46) 13 (54) NS Simvastatin (mg) 3 (17"6) Fluvastatin (mg) 2 (20"0) 4 (30"12) NS Lovastatin (mg) 3 (30"17) 1 (40) NS Atorvastatin (mg) 3 (17"6) 2 (30"14) NS Pravastatin (mg) 3 (33"12) 3 (30"17) NS Abbreviations: BMI, body mass index; LHFQ, Minnesota Living with Heart Failure Questionnaire; NYHA, New York Heart Association; HF, heart failure. Values are the means"standard deviation; NS, not significant LV dimension and function In both groups, end-diastolic and end-systolic diameters did not change during follow-up (Table 2). In both groups, LVEF significantly improved after 6 months of treatment compared to baseline (P-0.05). But differ-

4 198 P. Bahrmann et al. / The European Journal of Heart Failure 6 (2004) Table 2 Results at baseline and at 6 months for all patients who completed the study period Placebo (ns20) Pentoxifylline (ns21) Baseline 6 Months P Baseline 6 Months P LHFQ score 38"15 38"17 NS26"15 25"17 NS NYHA scale 2.6" "0.9 NS2.6" "0.5 NS Systolic BP (mmhg) 118"13 117"14 NS125"12 123"17 NS Diastolic BP (mmhg) 73"10 69"9 NS 77"9 75"10 NS Heart rate (bpm) 74"10 73"12 NS73"11 72"11 NS LV EDD (mm) 70"7 68"8 NS 68"8 64"7 NS LV ESD (mm) 57"11 57"7 NS 55"10 56"6 NS LVEF (%) 27"9 34" "7 33" Exercise time (min) 10"4 11"5 NS 11"5 11"3 NS VO (mlykgymin) 15"6 18" "5 19" TNF-a (pgyml) 9"5 10"7 NS 11"7 10"8 NS IL-6 (pgyml) 7"6 6"4 NS 12"30 5"2 NS BNP (pgyml) 292" " " "307 NS Abbreviations: LHFQ, Minnesota Living with Heart Failure Questionnaire score; BP, blood pressure; EDD, end-diastolic diameter; ESD, endsystolic diameter; EF, ejection fraction; VO, maximal oxygen uptake. Values are the means"standard deviation; NS, not significant. Table 3 Results at baseline and at 6 months for the patients with ischemic and hypertensive cardiomyopathy who completed the study period Placebo (ns12) Pentoxifylline (ns14) Baseline 6 Months P Baseline 6 Months P LHFQ score 38"15 38"17 NS26"15 25"17 NS Systolic BP (mmhg) 116"15 113"17 NS119"16 120"19 NS Diastolic BP (mmhg) 77"12 72"12 NS74"10 74"9 NS Heart rate (bpm) 77"11 75"13 NS72"12 74"11 NS LV EDD (mm) 71"8 68"7 NS 67"7 63"7 NS LV ESD (mm) 57"12 56"3 NS 50"7 58"4 NS LVEF (%) 30"9 33"11 NS30"6 34"10 NS Exercise time (min) 9"4 10"4 NS 9"3 11" VO (mlykgymin) 15"6 17"4 NS 15"3 19" TNF-a (pgyml) 10"6 11"8 NS 12"6 11"8 NS IL-6 (pgyml) 8"6 8"5 NS 16"36 5"2 NS BNP (pgyml) 367" "223 NS175" "248 NS Abbreviations: LHFQ, Minnesota Living with Heart Failure Questionnaire score; BP, blood pressure; EDD, end-diastolic diameter; ESD, endsystolic diameter; EF, ejection fraction; VO, maximal oxygen uptake. Values are the means"standard deviation; NS, not significant. Table 4 Results at baseline and at 6 months for the patients with idiopathic-dilated cardiomyopathy who completed the study period Placebo (ns8) Pentoxifylline (ns7) Baseline 6 Months P Baseline 6 Months P NYHA scale 2.6" "0.9 NS2.6" "0.5 NS Systolic BP (mmhg) 111"12 118"13 NS124"10 116"15 NS Diastolic BP (mmhg) 70"10 67"8 NS 79"10 76"8 NS Heart rate (bpm) 71"9 71"11 NS74"9 68"10 NS LV EDD (mm) 69"6 69"10 NS70"9 67"7 NS LV ESD (mm) 58"10 58"8 NS 62"9 55"7 NS LVEF (%) 23"6 35" "8 32" Exercise time (min) 12"4 12"6 NS 15"6 9"2 NS VO (mlykgymin) 16"7 20"7 NS 18"7 18"8 NS TNF-a (pgyml) 7"3 8"5 NS 8"7 8"8 NS IL-6 (pgyml) 5"4 4"3 NS 3"2 3"2 NS BNP (pgyml) 180"218 61"60 NS355" "414 NS Abbreviations: BP, blood pressure; EDD, end-diastolic diameter; ESD, end-systolic diameter; EF, ejection fraction; VO uptake. Values are the means"standard deviation; NS, not significant., maximal oxygen

5 P. Bahrmann et al. / The European Journal of Heart Failure 6 (2004) Fig. 1. Ejection fraction (EF, %) in the placebo group and the pentoxifylline group at baseline and at the end of the study. NS, non-significant. *P-value is within the group between baseline and 6 months; **P-value is the difference between the groups during the study. ences in LVEF between both groups were not significant (Fig. 1). Subgroup analysis for patients with ischemic and hypertensive cardiomyopathy or idiopathic-dilated cardiomyopathy showed similar results (Tables 3 and 4) TNF-a, IL-6 and BNP levels In both groups, plasma levels of TNF-a and IL-6 remained unchanged after 6 months of treatment compared to baseline. In the placebo group, the BNP plasma levels significantly decreased from baseline to 6 months (P-0.05) while in the pentoxifylline group, they showed a trend toward a no significant decrease, indicating an improvement (Table 2). But if subgroups with ischemic and hypertensive cardiomyopathy or idiopathic-dilated cardiomyopathy were independently assessed, BNP did not significantly change (Tables 3 and 4) LHFQ scores, hemodynamics and exercise tolerance LHFQ scores and NYHA scale did not significantly change in both groups (Table 2). No significant difference was observed in systolic, diastolic blood pressure, heart rate and exercise time, but VOsignificantly increased in both groups after 6 months (Table 2). But only the pentoxifylline-treated subgroup with ischemic and hypertensive cardiomyopathy showed significant increases in exercise time (Ps0.006) and VO 0.022) compared with placebo (Table 3). 4. Discussion 4.1. Pentoxifylline and left-ventricular function (Ps The degree of improvement in LVEF for the pentoxifylline-treated group with idiopathic-dilated cardiomyopathy in our study was comparable to Skudicky s results w18x. But in contrast to our results, he did not find a significant increment in LVEF for the placebo-treated patients. Sliwa also showed an improvement in LVEF for patients with idiopathic-dilated cardiomyopathy but not for patients with peripartum cardiomyopathy under treatment with pentoxifylline w16,19,23x. The mechanism of the underlying principle is yet not clearly understood. Barnett and Touchon showed that the clinical improvement of pentoxifylline is related to its haemorheological properties w20x. Sliwa hypothesized that the inhibition of phosphodiesterase activity by pentoxifylline led to the improvement in LVEF via increase in intracellular cyclic adenosine monophosphate and decrease in TNF-a w16x. Skudicky also suspected that an inhibition of apoptosis by pentoxifylline is responsible for its beneficial effects on LVEF w17,18x Pentoxifylline and inflammatory cytokines In our study, baseline TNF-a plasma concentrations were 4 5 times higher compared to a previous study

6 200 P. Bahrmann et al. / The European Journal of Heart Failure 6 (2004) done by Skudicky et al. Nevertheless, baseline LV enddiastolic diameter and standard treatment consisting of ACE inhibitors, diuretics and b-blockers for at least 3 months were similar in both studies w18x. ACE inhibition and b-adrenergic blockade are associated with a dose-dependent decrease in cytokine levels w24,25x. Therefore, altered doses of ACE inhibitors and b-blockers may be taken into account for the different TNF-a plasma levels since the prescribed amount of pentoxifylline was similar in both studies. However, elevation of cytokine levels in CHF is inconstant. Even in severe CHF, normal values can be present w26x. Also, plasma levels of TNF-a and IL-6 may underestimate cardiac tissue levels. Cardiac intracellular cytokine concentrations are not taken into account by our measurements w27x. In our study, TNF-a and IL-6 levels after 6 months of treatment were not significantly reduced in comparison to baseline. Also, Skudicky did not find a significant change in TNF-a plasma concentration under the same dose of pentoxifylline w18x. The latter concluded that an additive treatment with carvedilol might have suppressed TNF-a plasma concentrations in their study since b- adrenergic blockade decreases myocardial expression of TNF-a w25x. But Sliwa et al. revealed a significant lower TNF-a plasma concentration in the pentoxifylline-treated group with idiopathic-dilated cardiomyopathy than in the placebo group w16,19x. Recently, Sliwa confirmed that pentoxifylline reduces the levels of both apoptotic mediators, TNF-a and FasyApo-1, in patients with severe, decompensated heart failure due to idiopathic-dilated cardiomyopathy w19x. But in both studies, patients were not treated with b-blockers. Therefore, Skudickys results are in agreement with our study Pentoxifylline and symptoms Cardiovascular disability assessed by LHFQ scores or NYHA scale did not change in both groups. These findings are in contrast to the results of various other studies w16 18,20,23x. Several studies have demonstrated that circulating BNP concentrations increase with the severity of CHF based on NYHA classification w28,29x. In our study, additional treatment with pentoxifylline is neutral with regard to BNP levels and VO, another objective assessment for grading severity. However, in patients with ischemic and hypertensive cardiomyopathy, pentoxifylline had a beneficial influence on VO while placebo showed a trend towards an increase. In conclusion, the results of this randomised, placebocontrolled study show that addition of pentoxifylline to treatment with ACE inhibitors, diuretics and b-blockers is neutral with regard to left-ventricular function, inflammatory cytokines and symptoms in patients with ischemic, hypertensive and idiopathic-dilated cardiomyopathy. Acknowledgments We thank Mrs Kohler and Mrs Schmidt for excellent technical assistance. References w1x Cleland JGF. Heart failure: a medical hydra. Lancet 1998;352(Suppl):1 2. w2x Levine B, Kalman J, Mayer L, Fillit HM, Packer M. Elevated circulating levels of tumor necrosis factor in congestive heart failure. N Engl J Med 1990;323: w3x Mann DL, Young JB. Basic mechanisms in congestive heart failure. Recognizing the role of proinflammatory cytokines. Rev Chest 1994;105: w4x Torre-Amione G, Kapadia S, Benedict C, Oral H, Young B, Mann DL. Proinflammatory cytokine levels in patients with depressed left ventricular ejection fraction: a report from the studies of left ventricular dysfunctions (SOLVD). J Am Coll Cardiol 1996;27y5: w5x Torre-Amione G, Kapadia S, Lee J, et al. Tumor necrosis factor-alpha and tumor necrosis factors in failing human heart. Circulation 1996;93y4: w6x Seta Y, Shan K, Bozkurt B, Oral H, Mann DL. Basic mechanisms in heart failure: the cytokine hypothesis. J Cardiac Failure 1996;2: w7x Bolger AP, Anker SD. Tumor necrosis factor in chronic heart failure. Drugs 2000;60: w8x Dispersyn GD, Borgers M. Apoptosis in the heart: about programmed cell death and survival. News Physiol Sci 2001;16:41 7. w9x Mann DL. Inflammatory mediators and the failing heart: past, present, and the foreseeable future. Circ Res 2002;91: w10x Deswal A, Bozkurt B, Seta Y, et al. Safety and efficacy of a soluble P75 tumor necrosis factor receptor (Enbrel, etanercept) in patients with advanced heart failure. Circulation 1999;99: w11x Bozkurt B, Torre-Amione G, Warren MS, et al. Results of targeted anti-tumor necrosis factor therapy with etanercept (ENBREL) in patients with advanced heart failure. Circulation 2001;103: w12x Kelly RA, Smith TW. Cytokines and cardiac contractile function. Circulation 1997;95: w13x Belloc F, Jaloustre C, Dumain P, Lacombe F, Lenoble M, Boisseau MR. Effect of pentoxifylline on apoptosis of cultured cells. J Cardiovasc Pharmacol 1995;25(Suppl. 2):S71 S74. w14x Waage A, Sorensen M, Stordal B. Differential effects of pentoxifylline on tumor necoris factor and interleukin-6 production. Lancet 1990;335:543. w15x Zabel P, Wolter DT, Schonharting MM, et al. Oxpentifylline in endotoxaemia. Lancet 1989;2: w16x Sliwa K, Skudicky D, Candy G, Wisenbaugh T, Sareli P. Randomised investigation of effects of pentoxifylline on leftventricular performance in idiopathic dilated cardiomyopathy. Lancet 1998;351: w17x Skudicky D, Sliwa K, Bergemann A, Candy G, Sareli P. Reduction in FasyApo-1 plasma concentrations correlates with improvement in left ventricular function in patients with idiopathic dilated cardiomyopathy treated with pentoxifylline. Heart 2000;84: w18x Skudicky D, Bergemann A, Sliwa K, Candy G, Sareli P. Benefical effects of pentoxifylline in patients with idiopathic

7 P. Bahrmann et al. / The European Journal of Heart Failure 6 (2004) dilated cardiomyopathy treated with angiotensin-converting enzyme inhibitors and carvedilol. Circulation 2001;103: w19x Sliwa K, Woodiwiss A, Candy G, et al. Effects of pentoxifylline on cytokine profiles and left ventricular performance in patients with decompensated congestive heart failure secondary to idiopathic dilated cardiomyopathy. Am J Cardiol 2002;90: w20x Barnett JC, Touchon RC. Therapy of ischemic cardiomyopathy with pentoxifylline. Angiology 1990;41: w21x Rector TS, Cohn JN. Assessment of patient outcome with the Minnesota Living with Heart Failure questionnaire: reliability and validity during randomized, double-blind, placebo-controlled trial of pimobendan. Pimobendan Multicenter Research Group. Am Heart J 1992;124: w22x Wasserman K, Hansen JE, Sue DY, Casaburi R, Whipp BJ. Principles of exercise testing and interpretation. 3rd ed.. Philadelphia: Lippincott, Williams and Wilkins, w23x Sliwa K, Skudicky D, Candy G, Bergemann A, Hopley M, Sareli P. The addition of pentoxifylline to conventional therapy improves outcome in patients with peripartum cardiomyopathy. Eur J Heart Failure 2002;4: w24x Gullestad L, Aukrust P, Ueland T, et al. Effect of high- versus low-dose angiotensin converting enzyme inhibition on cytokine levels in chronic heart failure. J Am Coll Cardiol 1999;34: w25x Prabhu SD, Chandrasekar B, Murray DR, Freeman GL. b- adrenergic blockade in developing heart failure. Circulation 2000;101: w26x Testa M, Yeh M, Lee P, et al. Circulating levels of cytokines and their endogenous modulators in patients with mild to severe congestive heart failure due to coronary artery disease or hypertension. J Am Coll Cardiol 1996;28: w27x Muller-Werdan U, Schumann H, Loppnow H, et al. Endotoxin and tumor necrosis factor alpha exert a similar proinflammatory effect in neonatal rat cardiomyoctes, but have different cardiodepressant profiles. J Mol Cell Cardiol 1998;30: w28x Clerico A, Jervasi G, Del Chicca MG, et al. Circulating levels of cardiac naturetic peptide (ANP and BNP) measured by highly sensitive and specific immunoradiometric assays in normal subjects and in patients with different degrees of heart failure. J Endocrinol Invest 1998;21: w29x Marantz PR, Tobin JN, Wassertheil-Smoller S, et al. The relationship between left-ventricular systolic function and congestive heart failure diagnosed by clinical criteria. Circulation 1998;77: