Treatment of androgen deficiency with testosterone is undoubtedly beneficial for symptom relief and improving the quality of life in males with primary hypogonadism.
However, concerns about the heightened risk of adverse cardiovascular events may cause reluctance to use testosterone replacement therapy, even though the evidence for such an effect is limited.
The trouble with assessing the risk of adverse cardiovascular events is that they are rare enough to require large numbers of subjects in clinical trials for sufficient power to detect an effect of treatment.
Conventional meta-analyses of testosterone replacement therapy have, so far, failed to provide a definitive answer about safety[1].
Meta-analysis of individual participant data, in which original datasets are accessed to allow comparisons between individual participants in different studies, provides several benefits over conventional aggregate-level meta-analyses[2].
An individual participant data meta-analysis approach has now been applied to assess mortality and cardiovascular event outcomes of testosterone treatment of men with hypogonadism[3].
Placebo-controlled trials of at least three months of (any form of) testosterone treatment, for men aged 18 or older with testosterone levels of 12 nmol/L or lower, were included.
Thirty-five such studies were considered eligible for inclusion, and individual participant data were obtained for 17 of these controlled trials. This resulted in data from 1750 men who received testosterone and 1681 who received a placebo.
The duration of testosterone treatment in the included studies ranged from three months to three years, and the average duration of follow-up was nine and a half months.
Adverse cardiovascular and cerebrovascular events were no more common in testosterone-treated men (7.5%) than in the placebo group (7.2%: odds ratio 1.03, 95% confidence interval 0.77-1.38).
A benefit of using individual participant data was the ability to demonstrate that the effect of testosterone treatment on cardiovascular and cerebrovascular events was not influenced by age, smoking, diabetes, or by testosterone levels prior to treatment.
There were few participant deaths (0.4% for the testosterone group and 0.8% for the placebo), which limited the ability to draw meaningful conclusions about the effect of testosterone treatment on mortality (odds ratio 0.46, confidence interval 0.81-1.42).
Physiological consequences of testosterone treatment included lower total and HDL cholesterol, and higher haemoglobin concentration and haematocrit, compared to placebo.
Fasting glucose levels were lower in testosterone-treated men than in the placebo group (but not when participants with diabetes were excluded).
The results of this independent participant data meta-analysis are valuable for informing the management of men with hypogonadism.
The risk of adverse cardiovascular events was not increased by testosterone treatment for men with androgen deficiency, which is reassuring about the short-to-medium-term safety of the testosterone replacement therapy.
The studies included in the individual participant data meta-analysis provided sufficient periods of time for testosterone to affect physiology and influence cardiovascular disease outcomes, but there remains a need for information about long-term testosterone treatment and outcomes over years.
The ‘Testosterone Replacement Therapy for Assessment of Long-term Vascular Events and Efficacy ResponSE in Hypogonadal Men (TRAVERSE) Study’ is designed to provide information about cardiovascular and prostate safety up to approximately five years and should be completed soon[4].
