Experimental and clinical studies have shown that the administration of recombinant human growth hormone can improve deteriorated left ventricular function and hemodynamics in patients with heart failure. Herein, we compared the effects of growth hormone versus placebo upon resting left ventricular ejection fraction, exercise capacity and neurohormonal status in patients with advanced heart failure. Nineteen patients with advanced cardiac heart failure (ejection fraction <30%) were studied at baseline and after 8 weeks of treatment with growth hormone (0.03 U/kg per day) or placebo. Primary end points were resting left ventricular ejection fraction, peak oxygen consumption and neurohormonal status, including plasma norepinephrine levels and insulin like growth factor-1 and its binding protein-3. Results are presented as median and interquartile ranges. Patients receiving growth hormone had a significant increase in insulin growth factor-1 plasma levels (median difference growth hormone=83 ng/ml [57–170] versus placebo=−6 ng/ml [−23–6], P<0.05) and its binding protein-3. However, no significant increase in left ventricular ejection fraction after growth hormone treatment (ejection fraction pre=16% [13–18] and post=17% [14–27]) was noticed when compared to placebo (ejection fraction pre=20% [15–24] and post=20% [15–26]). Also, no significant effect of growth hormone treatment was seen on peak oxygen consumption or norepinephrine plasma levels. Although the administration of growth hormone to patients with advanced cardiac heart failure was associated with a significant increase in insulin growth factor-1, there were no significant changes in ejection fraction, exercise capacity and/or neurohormonal status.