Short-term Testosterone Use Doesn’t Spur CV Events: Meta-analysis

Men with hypogonadism see no added risk of cardiovascular events when taking testosterone, at least over the relatively short term, a meta-analysis of placebo-controlled, randomized trials suggests.

But while there appears to be no signs that monthslong testosterone use is harmful, researchers of the new study caution that it’s still unknown what happens when the therapy lasts for several years.

Speaking with TCTMD, Channa N. Jayasena, PhD (Imperial College London, England), one of the paper’s two senior authors, said that the study is reassuring but comes with caveats. Helpfully, its results “quell some of the controversy [that’s happened] in the past about different meta-analyses finding different things,” he said, adding that there have been no trials powered to detect differences in CV events.

“We haven’t managed to find evidence that testosterone increases cardiovascular risk,” said Jayasena. “However, we clearly can’t exclude that in the long term or a higher-powered trial that could be found in the future.”

Healthcare professionals need to know what to tell patients in the here and now.

“Speaking as a clinician who regularly starts men on testosterone and likes to think of myself as balanced, I want to say that in the past the problem has been that people have been over- and underprescribing it. . . . What we want is more consistency,” he suggested. For “someone who clearly would benefit from it, then I would be very reassuring and say, ‘Look, this is really safe and the chance of anything bad happening is very small.’”

Complicating the picture is the fact that the cutoff for what defines hypogonadism varies across the world, Jayasena added.

The new paper, with Jemma Hudson, MSc, as first author and Miriam Brazzelli, PhD, as a senior author (both University of Aberdeen, Scotland), was published online yesterday in the Lancet Healthy Longevity. This meta-analysis joins a growing body of literature that undercuts the notion that testosterone therapy is cardioprotective. In 2015, the US Food and Drug Administration urged caution when it comes to testosterone, a move that was followed by more studies suggesting the potential for harm, even when use is indicated.

Erin D. Michos, MD (Johns Hopkins University School of Medicine, Baltimore, MD), who co-authored an accompanying editorial with Matthew J. Budoff, MD (University of California, Los Angeles and Lundquist Institute), told TCTMD she doesn’t find the current study “all that reassuring,” as it’s a “meta-analysis of a bunch of smaller studies that had short follow-up.”

Forthcoming results from the TRAVERSE trial hopefully will provide some insights. The RCT is testing 5 years of topical testosterone versus placebo in over 5,000 men with hypogonadism who have existing CVD or who are at high risk of CVD.

No Difference in Mortality or CV Events

For the meta-analysis, investigators searched the literature for randomized controlled trials that enrolled adult men with a screening testosterone of 12 nmol/L (350 ng/dL) or less and tested any testosterone treatment given for at least 3 months (no matter the formulation, dose frequency, or route of administration) versus placebo.

They identified 17 RCTs with patient-level data for 3,431 participants, who had a mean age of 65 years and took testosterone for a mean duration of 9.5 months. Nearly nine in 10 were white and a similar proportion were nonsmokers. Their mean body mass index was 30 kg/m². Diabetes and angina each were seen in slightly more than one-quarter of participants, and around 8% had had a previous MI.

Testosterone treatment did not affect blood pressure or glycemic markers, and its use was associated with “modest lowering” of total cholesterol, HDL cholesterol, and triglyceride levels.

Cardiovascular events, such as arrhythmia, coronary heart disease, heart failure, cerebrovascular events, and MI, occurred at similar rates with testosterone and placebo (7.5% vs 7.2%; P = 0.62). Neither patient age nor baseline testosterone, smoking, or diabetes were linked to cardiovascular risk. Deaths were rare, and only reported by 14 of the RCTs, but also were evenly distributed between the two groups (0.4% vs 0.8%; P = 0.13).

The researchers found similar results when combining data from the 17 trials that provided patient-level data with another 18 that did not.

Some of the nuances of their findings came as a surprise, Jayasena noted, such as the lack of difference in outcomes by age or by baseline testosterone level. “We guessed that it may be safer the lower testosterone you have to start with. So if you’ve got incredibly low testosterone, it’s almost like having an empty bath; then it’s probably safer, we thought, filling up the bath than if you’re just below the limit and have more of a risk of overflowing and therefore having bad effects,” he explained. “But we didn’t see that.”

Michos pointed out that it’s normal for men to see their testosterone decrease by about 2% per year as they age and stressed that not all need therapy in order to bring it back to a certain level. While “many people are using it to ‘treat the number,’” she said, “right now the only indication for testosterone therapy is if someone not only has a low level but has symptoms of hypogonadism, [such as] erectile dysfunction, loss of muscle mass, or decreased libido that’s bothering them. Then it’s really an individualized discussion based on how symptomatic they are and their underlying cardiovascular risk.”

In her own practice, Michos said she’s “very cautious about hormone therapy” but that it comes down to shared decision-making. For a symptomatic patient with low testosterone and low risk of CVD, she would counsel them that while there’s no evidence of short-term harm, “we really don’t have good data to date about the long-term safety.”