Reviewed research

Authors Budoff MJ, Ellenberg SS, Lewis CE, et al.

Review Date May 2017

Citation JAMA 2017;317(7):708-716



Both observational and intervention studies of the effects of testosterone on clinical cardiovascular outcomes have had conflicting results and the purported benefits or risks of testosterone treatment in older men are yet to be demonstrated conclusively. Although there have been several clinical trials of testosterone treatment in older men, none were designed to measure cardiovascular endpoints specifically. The Testosterone Trials is a group of seven trials designed to determine the efficacy of testosterone treatment in men aged 65 or older with age-related low testosterone, one of which was a Cardiovascular Trial.



The Cardiovascular Trial was designed to test the hypothesis that testosterone treatment of older men with low testosterone slows progression of non-calcified coronary artery plaque volume, assessed by coronary computed tomographic angiography (CCTA) as an indicator of coronary atherosclerosis.



The Cardiovascular Trial was a double-blinded, placebo-controlled trial at nine academic medical centers in the United States. Participants were recruited by mass mailings with respondents screened by telephone and two clinic visits. Inclusion criteria were men aged 65 years or older, serum testosterone levels that averaged lower than 275 ng/dL over two morning samples, and evidence of sexual dysfunction, physical dysfunction, and/or reduced vitality (the latter were inclusion criteria for the main trials). Exclusion criteria: high risk of prostate cancer; myocardial infarction or stroke within the previous three months and a systolic blood pressure higher than 160 mm Hg or diastolic blood pressure higher than 100 mm Hg.

Of the 357 men assessed for the Cardiovascular Trial, 170 men were enrolled between June 2010 and June 2014. Eighty-two men were assigned to receive a placebo gel and 88 received testosterone gel, with the dose adjusted to maintain the testosterone level in the normal range for young men, for 12 months.

Testosterone concentration was measured at baseline and 3, 6, 9 and 12 months. CCTA was performed at baseline and 12 months.

The primary outcome measure was non-calcified coronary artery plaque volume, as determined by CCTA. Previous studies have demonstrated an association between this measure and myocardial ischaemia and subsequent cardiovascular events. Secondary outcomes included total coronary artery plaque volume and coronary artery calcium score (range of 0 to >400 Agatston units, with higher values indicating more severe atherosclerosis).

The association of testosterone with outcomes related to coronary artery plaque and calcium scores was assessed using a multivariable linear model that adjusted for baseline plaque volume and all balancing variables: study site, indicator variables of participation in each primary efficacy trial, baseline testosterone concentration (≤200 ng/dL or >200 ng/dL), age (≤75 years or >75 years), use of antidepressants, and use of PDE5 inhibitors as covariables. The adjusted mean difference was calculated for the change in plaque volume from baseline to 12 months for the testosterone group compared with placebo.



Of 170 men who were enrolled, 138 (65 placebo and 73 testosterone treatment) completed the study and were included in the primary analysis. Mean (SD) age was 71.2 (5.7) years, and 81% were white.

At baseline, 70 men (50.7%) had a coronary artery calcification score higher than 300 Agatston units, reflecting severe atherosclerosis. Compared to placebo, testosterone treatment was associated with a significantly greater increase in non-calcified plaque volume from baseline to 12 months (median values of 204 mm3 baseline to 232 mm3 at 12 months vs 317 mm3 to 325 mm3, respectively; estimated difference, 41 mm3; 95% CI, 14 to 67 mm3; P = 0.003). For the secondary outcomes, the median total plaque volume increased from baseline to 12 months from 272 mm3 to 318 mm3 in the testosterone group vs 499 mm3 to 541 mm3 in the placebo group (estimated difference, 47 mm3; 95% CI, 13 to 80 mm3; P = 0.006), and the median coronary artery calcification score changed from 255 to 244 Agatston units in the testosterone group vs 494 to 503 Agatston units in the placebo group (estimated difference, -27 Agatston units; 95% CI, -80 to 26 Agatston units).

No major adverse cardiovascular events occurred in either group.



Among older men with symptomatic hypogonadism, treatment with testosterone gel for 12 months was associated with a significantly greater increase in non-calcified coronary artery plaque volume, as measured by CCTA, and total plaque volume compared to placebo. However, testosterone treatment was not associated with coronary artery calcium score. Larger studies are needed to understand the clinical implications of this finding.


Points to Note
  1. There are conflicting findings from previous studies that have assessed cardiovascular outcomes for older men having testosterone treatment.
  2. This placebo-controlled trial showed that testosterone treatment in older men with symptomatic hypogonadism is associated with a significantly greater increase in non-calcified plaque volume, a maker of coronary atherosclerosis, over 12 months compared with placebo.
  3. It should be noted that the participants in this study had high atherosclerotic risk as many were obese, hypertensive, current/former smokers or had type 2 diabetes.
  4. Given that the primary outcome is a surrogate measure, the clinical significance with respect to hard cardiovascular outcomes remains to be elucidated.
  5. The trial was not long enough or large enough to draw conclusions about the risk of testosterone treatment on major adverse cardiovascular events.
  6. These findings and those of the other Testosterone Trials do not change the current understanding that low testosterone levels due to obesity and other comorbidities of ageing are better addressed by lifestyle measures to treat the comorbidities, with consequent positive effects on testosterone levels.


Website: https://www.ncbi.nlm.nih.gov/pubmed/28241355

Reviewed by
Updated on