"This naturally occurring and cheap hormone seems to improve functional capacity and functional class compared with placebo in men with chronic heart failure," said lead investigator Dr Christopher Malkin (Royal Hallamshire Hospital, Sheffield, UK). "We're seeing effects within the normal physiological range of testosterone. We're not treating patients to super-physiological levels but to levels that you would probably see in elite athletes. With the data we have, it is still difficult to tease out the mechanism, although it is probably multifactorial, since we know that testosterone can act on numerous sites in heart-failure syndrome."

During his presentation, Malkin said that chronic heart failure is associated with maladaptive and prolonged neurohormonal and proinflammatory cytokine activation. This, he said, can cause a metabolic shift toward catabolism, as well as vasodilator incapacity and loss of skeletal muscle bulk and function. Androgens, while determinants of anabolic function and muscle strength, also possess anti-inflammatory and vasodilatory properties, he noted.

In this double-blind, placebo-controlled study, investigators randomized 76 men with documented heart failure, mean ejection fraction 32.5%, to a 5-mg transdermal testosterone patch or placebo. Patients were elderly, mean age 64 years, and a majority of them were in NYHA class 2 and 3 heart failure. The primary end point of the study was functional capacity as assessed by the incremental shuttle walk test at 12 months. Secondary end points included muscle strength and skeletal muscle bulk, assessed by grip strength and cross-sectional computed tomography, respectively, BNP and hematocrit levels, as well as pulse and blood pressure.

Overall, baseline testosterone levels were low, approximately 13 nmol/L. The endocrinologic definition of low androgen levels is less than 11 nmol/L, said Malkin, and nearly 25% of patients met this definition. The average testosterone level of a healthy male approximately 30 years of age exceeds 20 nmol/L, although this declines over time.

In total, while 76 patients started the study, only 42 completed treatment. Malkin said the testosterone-patch preparation, made with an adhesive and ethanol solvent in which the testosterone is embedded, causes unpleasant skin reactions, and this was the most-often cited reason for study dropouts. Among those who completed the trial, testosterone levels increased approximately 5.3 nmol/L among those treated with the supplemental androgen therapy, a lower-than-expected increase. Despite this, there was a significant improvement in functional capacity compared with placebo. Those treated with testosterone had a 38-m improvement in the shuttle walk test, corresponding to an 18% increase from baseline.

"It is important to note that 25% of patients in this study were androgen deficient, and the amount of testosterone that we gave them was very small," said Malkin. "Still, a number of those patients remained androgen deficient. If we'd improved the testosterone levels further, we might have seen better effects than we observed with these data."

In terms of secondary end points, Malkin said there were documented improvements in grip strength, but no change in cross-sectional skeletal muscle area. The left ventricular cavity increased by 1 cm, and left ventricular mass index, a measure of the mass of the heart, was reduced. On average, there was a 0.50 improvement in NYHA class, with 35% of patients' symptoms improving by at least one functional class. There was no change in BNP levels.

"I think the next thing to do would be to try a different preparation of testosterone," said Malkin. "Half of the patients didn't tolerate the skin patch well, and I think we need to confirm these effects with a more suitable preparation. There are now gels, which are much more tolerable, and actually, most patients seem to prefer the injections."

Exercise training was not part of the study, and as a result, patients were not advised to initiate strength-training programs. Malkin said his group is planning to conduct future studies that assess the effects of exercise and testosterone therapy on heart-failure patients and would like to test future preparations of testosterone to determine their efficacy.