Physiological testosterone therapy increases exercise capacity and reduces symptom scores in men with chronic heart failure (CHF). Tumour necrosis factor-alpha (TNF-alpha) exerts a significant pathologic activity in CHF, and physiologic testosterone replacement therapy is associated with reduced serum levels of TNF-alpha in hypogonadal men with concomitant coronary artery disease. It is unknown whether testosterone exerts a similar immunomodulatory action in men with CHF. Testosterone therapy administered in three placebo-controlled studies, for either 6 hours (two 30-mg buccal tablets, n=12) or 3 months (fortnightly 100 mg intra muscular injection, n=20; or daily 5 mg transdermally, n=62). The effects of testosterone were also assessed on lipopolysaccharide (LPS)-induced TNF- production in whole blood obtained from 27 men with CHF. Incubation with testosterone (10 nM, 1 M, and 100 M) resulted in a reduction in LPS-induced TNF- production from 12.6 +/- 1.3 to 11.2 +/- 1.1 (P = 0.053), 10.3 +/- 1.1 (P = 0.0046), and 9.2 +/- 1.1 (P = 0.000066) ng/ml, respectively. However in men with CHF, serum levels of TNF- were similar before and after treatment with testosterone or placebo, irrespective of the length of study or route of administration. The clinically beneficial actions of testosterone in men with CHF are unlikely to be mediated by reducing TNF-alpha.