madman
Super Moderator
Cardiovascular Safety of Testosterone-Replacement Therapy
To the Editor: Lincoff et al. (July 13 issue)1 report the results of the Testosterone Replacement Therapy for Assessment of Long-term Vascular Events and Efficacy Response in Hypogonadal Men (TRAVERSE) trial, a noninferiority trial comparing the effect of testosterone-replacement therapy with that of placebo on major adverse cardiac events in men with hypogonadism and preexisting or high cardiovascular risk. Testosterone-replacement therapy was noninferior to placebo, but the testosterone group had a higher incidence of nonfatal arrhythmias, including atrial fibrillation. Atrial fibrillation is known to be an important risk factor for cardiovascular events, and the relationship between testosterone and atrial fibrillation remains unclear. Results from the Finland Cardiovascular Risk Study (FINRISK) suggested a link between low testosterone levels and an increased risk of atrial fibrillation among men,2 and another study showed a positive correlation between total plasma testosterone concentration and the incidence of atrial fibrillation among men.3 Further investigation of the potential nonlinear relationship and the dose–response association between testosterone levels and atrial fibrillation is warranted. In addition, analyzing different subgroups on the basis of age, cardiovascular risk types, and baseline testosterone levels would provide valuable evidence for personalized treatment and safety. Finally, the effects of the mode of administration of testosterone, exogenous testosterone doses, and testosterone metabolites on cardiovascular safety should not be overlooked.
Yuan Huang, M.D.
Jiangbin Hospital of Guangxi Zhuang Autonomous Region
Nanning, China
Bin Shen, M.D.
Yi‑fan Zhou, M.D.
People’s Hospital of Guangxi Zhuang Autonomous Region
Nanning, China
To the Editor: The current recommendation for testosterone-replacement therapy is to attain a testosterone level of 450 to 600 ng per deciliter.1 The TRAVERSE trial attempted to address whether testosterone-replacement therapy increased the risk of cardiovascular complications and used a testosterone level of 350 ng per deciliter as the lower boundary of the target range, which is approximately 23% lower than the recommended lower boundary. This aim was not achieved: more than 50% of men had testosterone levels below 350 ng per deciliter. Although this lower testosterone level was noninferior with respect to safety, concerns of cardiovascular risk are not dispelled by the results of the TRAVERSE trial. It is important to note that the efficacy data were not reported; hence, concluding that this dosing strategy offers benefits or should be implemented would be difficult. It is of concern that despite the fact that the currently recommended lower boundary for testosterone of 450 ng per deciliter was not reached, these lower levels were associated with an increased risk of pulmonary embolism, arrhythmias, atrial fibrillation, and kidney injury, and the confidence intervals do not clearly exclude a 50% increased risk of nonfatal myocardial infarction, nonfatal stoke, revascularization, heart failure, or venous thromboembolism. Finally, the data from the large number of testosterone levels collected could have clarified the testosterone–cardiovascular disease conundrum by providing a correlation between cardiovascular disease events and testosterone levels.
Saul Malozowski, M.D., Ph.D.
National Institute of Diabetes and Digestive and Kidney Diseases
Bethesda, MD
David M. Reboussin, Ph.D.
Wake Forest University School of Medicine
Winston-Salem, NC
To the Editor: Although the TRAVERSE trial provides welcome data regarding the cardiovascular safety of testosterone-replacement therapy in older men, it is unclear whether the patients were truly hypogonadal and would be offered testosterone treatment in clinical practice. The patients had obesity, and approximately two thirds had diabetes. “Diabesity” (i.e., concomitant diabetes and obesity) may reduce serum levels of sex hormone–binding globulin and total testosterone, whereas the free testosterone level remains normal (i.e., “pseudo-hypogonadism”)1 . How many patients had a reduced serum free testosterone level? Some hypogonadal symptoms that were chosen as eligibility criteria, such as fatigue, decreased energy, and low mood, are not specific and are common among men with normal testosterone levels.2 These symptoms may reflect sleep apnea and depression, conditions that are common among men with diabesity. The absence of hypogonadism and of a perceived benefit may explain the 61.4% of patients who discontinued testosterone. The incidence of discontinuation was lower in the Testosterone Trials, which stringently selected men with age-associated hypogonadism.3 Finally, maximizing the management of coexisting conditions, including strategies for weight loss and glycemic control and treatment of sleep apnea and depression, can safely reduce symptoms and improve health outcomes. Weight loss can also increase the serum testosterone level in increments similar to that achieved with testosterone-replacement therapy in the TRAVERSE trial.4
Mathis Grossmann, F.R.A.C.P., Ph.D.
Jeffrey D. Zajac, F.R.A.C.P., Ph.D.
University of Melbourne
Heidelberg, VIC, Australia
The authors reply: We agree with Huang et al. that the relationship between circulating testosterone levels and the risk of atrial fibrillation remains unclear. The FINRISK study1 showed a weak negative association between baseline testosterone levels and the risk of atrial fibrillation or stroke (a composite end point), whereas the Atherosclerosis Risk in Communities (ARIC) study2 showed a positive association between testosterone levels and incident atrial fibrillation. A meta-analysis did not show a significant difference between testosterone treatment and placebo with respect to the incidence of cardiac arrhythmias among men.3 We agree that more detailed post hoc analyses of the data, including subgroup analyses, may help generate exploratory hypotheses for future studies.
The narrow target range of testosterone levels mentioned by Malozowski and Reboussin is not recommended by the Endocrine Society guidelines, which suggest that clinicians aim to attain testosterone values in the mid-normal range.4 Testosterone levels during treatment vary substantially at different times of the day and on different days in the same person, which makes it difficult to maintain testosterone levels continuously within the narrow limits mentioned, even if average 24-hour concentrations are within the normal range. Testosterone levels in the TRAVERSE trial were measured at the nadir during therapy (24 hours after the last dose). Nadir levels are 30 to 40% lower than the peak levels 2 to 8 hours after gel application. Because the trial was intended to rule out a cardiovascular hazard, the efficacy data was reserved for subsequent manuscripts. We agree that the reported safety data, along with the efficacy data, will facilitate a more informed appraisal of the potential benefits and risks of testosterone-replacement therapy.
Grossman and Zajac express concern about whether the patients in our trial were truly hypogonadal. Patients were enrolled in the trial if they met the criteria established by the Endocrine Society guidelines for the diagnosis of hypogonadism.4 Patients were required to have one or more symptoms of hypogonadism and two fasting, morning testosterone levels of less than 300 ng per deciliter. We agree that some of the symptoms of hypogonadism, such as fatigue, can be nonspecific and overlap with other chronic diseases. Eligibility criteria included the presence of preexisting cardiovascular disease or three or more cardiovascular risk factors, which resulted in a high percentage of men with obesity, diabetes, or both — a prevalence that is similar to that among hypogonadal men receiving testosterone treatment in the United States.5 The results of the trial do not preclude the need for management of diabetes, obesity, sleep apnea, clinical depression, or other coexisting conditions before considering testosterone-replacement therapy.
Shalender Bhasin, M.B., B.S.
Brigham and Women’s Hospital
Boston, MA
A. Michael Lincoff, M.D.
Steven E. Nissen, M.D.
Cleveland Clinic
Cleveland, OH
