Background: Several studies have reported a direct association between elevated plasma levels of inflammatory cytokines and worse functional class (New York Heart Association [NYHA]) and cardiac function, measured as left ventricular ejection fraction (LVEF). Thalidomide has recently shown to improve LVEF in chronic heart failure patients, accompanied by a marked decrease in plasma levels of tumor necrosis factor alpha (TNF-α). Methods: In a randomized prospective open label study of men and women with heart failure (HF) due to ischemic and non-ischemic cardiomyopathy who had systolic dysfunction (LVEF <40%) and NHYA classification, functional classes II and III were assigned to control (without thalidomide, 60 patients) or thalidomide group (20 patients). The initial dose of thalidomide was 100 mg once a day, and it was increased to 100 mg twice a day after a period of 10 days, if the prior dosage was well-tolerated. Demographic characteristics, etiology of HF, prior myocardial infarction, co-morbidities associated were registered and laboratory routine test, TNF-α serum levels, and echocardiogram were obtained at the beginning and after 6 months of follow-up. Results: Clinical status (NYHA) at the end of the follow-up period, improved moderately in both groups. TNF-α levels were initially of 5.88 ± 0.9 and 6.49 ± 1.82 vs. 6.32 ± 1.6 and 7.94 ± 3.8 pg/ml during follow-up, for thalidomide and control groups, respectively. There were non-significant differences in echocardiography variables. Conclusion: In conclusion, although there is a large amount of information supporting a direct relationship between TNF-α and worsening of symptoms and prognosis in patients with HF and recently, the beneficial effect on thalidomide treatment has been suggested, these preliminary observations should be confirmed in a larger prospective study, specially trying to clarify the action mechanisms.