The Reply

Abstract

We thank Traish et al for the opportunity to reiterate our message and to remind them about the position of the endocrine societies on testosterone replacement therapy in aging men. The Endocrine Society guidelines state that testosterone replacement therapy should be given to men with underlying disorders that cause hypogonadism. As such, we chose not to include men with known causes of hypogonadism in our study, as testosterone replacement therapy would likely be given to this population. Conversely, testosterone replacement therapy is currently not recommended for all men with age-related decrease in testosterone. The Testosterone Trials reported some benefits in aging men with low testosterone levels, but these benefits were considered modest by some authors, and for the Endocrine Society and the European Society of Endocrinology, the clinical benefits of testosterone replacement therapy in this population remains uncertain. Thus, although Traish et al appear to strongly support testosterone replacement therapy in aging men with functional hypogonadism, there is currently no clear consensus in the scientific community. We found an increased risk of cardiovascular events with current use of testosterone replacement therapy, compared with no use.1Loo SY, Azoulay L, Nie R, Dell'Aniello S, Yu OHY, Renoux C. Cardiovascular and cerebrovascular safety of testosterone replacement therapy in hypogonadism among aging men: a cohort study [e-pub ahead of print]. Am J Med. 2019 Apr 3. pii:S0002-9343(19)30281-5. doi:10.1016/j.amjmed.2019.03.022Google Scholar The risk was highest in the first 2 years of use, a finding that was statistically significant. It should be noted that the timing of this risk is very similar to what has been reported for the association between testosterone replacement therapy and venous thromboembolism. The point estimates were very stable in all sensitivity analyses, most around 1.21, as shown in Figure 2. Traish et al state that a critical confounder, inadequately recognized, is that testosterone deficiency is a risk for increased cardiovascular events. However, all men had low testosterone levels at cohort entry, so we selected a homogeneous population in this respect. Despite an increased cardiovascular risk, we found a strong protective effect on mortality with current use, whereas past testosterone replacement therapy was associated with a strong increased risk of death. This would mean that patients stopping testosterone replacement therapy just a few weeks prior have an immediate nearly twofold increased risk of dying. Such a pattern is indicative of reverse causality rather than a true causal effect, that is, high-risk patients stop using testosterone replacement therapy due to worsening health. To address this issue, we performed a sensitivity analysis using 5-alpha-reductase inhibitors, a negative control exposure with no known effects on cardiovascular risk or mortality. This analysis revealed no association with cardiovascular events, but a lower risk of all-cause mortality with current use and an increased risk with past use. This pattern, which mirrored the one observed with testosterone replacement therapy, provides further evidence that reverse causality might explain the decreased mortality risk observed with current use of testosterone replacement therapy. Finally, nowhere in the manuscript do we state that aging men should not be treated. However, based on the results of our study and from prior studies, we caution physicians who prescribe testosterone replacement therapy to aging men with low testosterone levels that there is a possible increased risk of cardiovascular events. Patients also should be aware of this potential risk. This message is in line with the warning issued by health agencies.