Inflammatory cytokines have been related to the pathogenesis and progression of heart failure.1 Cytokines can modulate cardiovascular function through different mechanisms, such as depressing contractile function, promoting left ventricular remodelling, uncoupling myocardial β adrenergic receptors, and inducing apoptosis. Loss of myocytes caused by apoptosis or programmed cell death occurs in patients with heart failure and may contribute to progressive myocardial dysfunction.2 Fas is an apoptosis signalling surface receptor known to trigger programmed cell death in a variety of cell types,3 and increased plasma concentrations of soluble Fas receptors have been reported in patients with heart failure. Pentoxifylline is a xanthine derived agent known to inhibit the production of tumour necrosis factor α (TNFα). It was also found to inhibit apoptosis in different human cell types in vitro and in vivo.4 However, the effects of pentoxifylline on apoptosis signalling receptors in patients with heart failure has not been investigated. We have previously reported beneficial effects of pentoxifylline in patients with idiopathic dilated cardiomyopathy.5 Treatment with pentoxifylline resulted in a significant improvement in functional class and left ventricular function, and was associated with a reduction in TNFα plasma concentrations. The aims of the present study were to evaluate: (1) the effects of pentoxifylline on Fas/APO-1 plasma concentrations in patients with idiopathic dilated cardiomyopathy; (2) the possible correlation between the changes in the cytokine or Fas concentrations and the changes in left ventricular ejection fraction following six months of treatment with pentoxifylline; and (3) to establish predictors of outcome.

In a single centre, prospective, double blind, randomised, placebo controlled trial, we included 49 patients with idiopathic dilated cardiomyopathy in New York Heart Association functional class II or III and left ventricular ejection fraction < 40% by radionuclide angiography. All patients received treatment with digoxin, diuretics, and angiotensin converting enzyme inhibitors. After baseline examinations were performed, patients were randomised to pentoxifylline 400 mg three times daily (n = 25) or a matching placebo (n = 24). Mean (SD) age of the study population was 52 (11) years and baseline mean left ventricular ejection fraction 23 (9)%. There were no significant baseline differences between groups. Eight patients died during the study period (six in the placebo group). Treatment with pentoxifylline was associated with a significant improvement in functional class, and left ventricular ejection fraction (21 (8)% to 35 (16)%, p = 0.00007). No significant changes in left ventricular dimensions or function were observed in the placebo group. Twelve of the patients treated with pentoxifylline showed an improvement of the ejection fraction > 10 absolute units, and were arbitrarily defined as improvers, whereas only two patients in the placebo group showed a similar degree of improvement (p = 0.008 between groups).

TNFα and Fas plasma concentrations were significantly higher in the study group compared to 20 healthy volunteers (7.4 (5) pg/mlv 1.4 (1) pg/ml, p < 0.0001; and 6.7 (5) U/ml v 0.8 (0.2) U/ml, p < 0.0001, respectively). Baseline concentrations of TNFα and Fas were similar in the placebo and treatment arms. Patients treated with pentoxifylline had a significant reduction in the TNFα (6.6 (5) pg/ml to 1.9 (1) pg/ml, p = 0.0005) and Fas plasma concentrations (8.0 (7) U/ml to 5.5 (6) U/ml, p = 0.004) after six months, with no significant changes in the placebo group.

Despite a similar ejection fraction at baseline, non-improvers had larger left ventricular end diastolic (p = 0.06) and end systolic diameters (p = 0.05), and a higher E/A ratio (p = 0.03) compared to improvers. Baseline TNFα plasma concentration was significantly higher in the 12 patients that were defined as improvers in the pentoxifylline group compared to non-improvers (7.26 (3.2) pg/mlv 3.66 (2.3) pg/ml respectively, p = 0.01). Similarly, there was also a trend towards higher baseline Fas concentrations among improvers (p = 0.09). No other significant differences in the baseline characteristics were noted between these two groups. Following six months of treatment with pentoxifylline, both improvers and non-improvers showed a significant reduction in TNFα concentrations, although there was a trend towards a more significant decline among the improvers (reduction of 5.2 (4) pg/mlv 2.0 (3) pg/ml in improvers and non-improvers respectively, p = 0.06). There was a significant correlation between baseline Fas/APO-1 plasma concentrations, and changes in ejection fraction after treatment (r = 0.47, p = 0.03), and a significant negative correlation between the changes in Fas plasma concentrations, and the changes in ejection fraction at six months (r = −0.49, p = 0.03). No significant correlation was observed between baseline TNFα concentrations or changes in TNFα concentration, and changes in ejection fraction following treatment. Of all baseline variables analysed, only Fas/APO-1 concentration and E/A ratio emerged as independent predictors of improvement in the multivariate analysis (p = 0.02).

The reduction in Fas concentrations observed in this study suggests a possible inhibition of apoptosis by pentoxifylline. The finding that baseline concentrations of TNFα and Fas/APO-1 concentrations were higher in patients with a more favourable response to treatment is important to help us to understand the mechanism of action of pentoxifylline in patients with heart failure. The lack of statistical correlation between the changes in cytokine concentrations and the changes in left ventricular function might be explained by the fact that in the pentoxifylline group TNFα dropped even in patients that did not improve the ejection fraction. Finally, although current evidence supports the proposal that apoptosis occurs in heart failure, it is still unclear to what extent it contributes to the progression of myocardial dysfunction. Hirota and colleagues,6 using a gene knockout mice model, recently showed that apoptosis plays a critical role in the transition between compensatory cardiac hypertrophy and heart failure during aortic pressure overload. Therefore, the use of treatments that can block apoptotic pathways could be a useful strategy in the treatment of patients with heart failure. The results of our study suggest that attenuation of apoptosis together with a reduction in TNFα concentrations may be the mechanism of improvement of left ventricular function in patients with idiopathic dilated cardiomyopathy treated with pentoxifylline.