Resistance to insulin occurs in chronic heart failure (CHF) and is related to prognosis. Studies of testosterone in non-(CHF) males suggest that physiological testosterone therapy improves insulin sensitivity. This was a single-blind placebo controlled crossover trial to determine the effect of testosterone replacement on insulin sensitivity in 13 men with moderate to severe CHF (ejection fraction 30.5+/-1.3). The primary outcome was the homeostatic model index (HOMA-IR) of fasting insulin sensitivity and secondary outcomes were body composition as measured by bioelectrical impedance and glucose tolerance to a standard 75 g oral glucose load. Analysis was performed on the delta values with the treatment effect of placebo compared with that of testosterone. At baseline HOMA-IR correlated with measures of body fat [% fat mass (rP=0.84, p=0.0001) and body mass index (rP=0.79, p=0.01)] but not with CHF severity. Testosterone reduced HOMA-IR (-1.9+/-0.8, p=0.03) indicating improved fasting insulin sensitivity. Testosterone also increased total mass (+1.5+/-0.5 kg, p=0.008) and decreased body fat (-0.8+/-0.3%, p=0.02). Testosterone improves fasting insulin sensitivity in men with CHF and may also increase lean body mass, these data suggest a favourable effect of testosterone on an important metabolic component of CHF.