The TRAVERSE Trial in Plain English: What 5,246 Men Just Taught Us About TRT and Your Heart

For three decades, every conversation about testosterone replacement therapy started the same way: but is it safe for my heart? Doctors didn’t agree. The FDA didn’t either — in 2015 it slapped a cardiovascular warning on every testosterone product in the U.S. and left it there for a decade.

In 2023, the largest placebo-controlled testosterone trial ever run finally answered the question. In February 2025, the FDA agreed with the trial and removed the warning. The story is good news, mostly — but it isn’t only good news, and the honest version is more useful than the headline.

Here is what TRAVERSE actually found, what it didn’t find, and who its results don’t apply to.

TRAVERSE (Testosterone Replacement Therapy for Assessment of Long-term Vascular Events and Efficacy Response in Hypogonadal Men) was published in the New England Journal of Medicine on July 13, 2023, with lead author Dr. A. Michael Lincoff of the Cleveland Clinic.

The trial enrolled 5,246 men, aged 45 to 80, all with two morning testosterone levels under 300 ng/dL plus symptoms of hypogonadism — meaning they qualified for treatment under standard guidelines. Every man also had established cardiovascular disease or a high cardiovascular risk profile. This was not a study of young, healthy gym-goers. It was a study of exactly the men a cautious clinician would worry about.

Half received a daily 1.62% testosterone gel, dosed to a target of 350–750 ng/dL. Half received a matching placebo gel. They were followed for an average of just under two years on treatment, with longer follow-up for safety endpoints — specifically designed to answer the cardiovascular question the FDA had flagged a decade earlier.

The primary endpoint was MACE — major adverse cardiovascular events, a composite of cardiovascular death, non-fatal heart attack, and non-fatal stroke. The result:

• Testosterone group: 7.0% had a MACE event.
• Placebo group: 7.3% had a MACE event.

Statistically, that meets the formal definition of non-inferior — testosterone did not increase the risk. After 30 years of debate and a black-box warning, the trial designed to settle it produced a flat line. For men who fit the trial’s profile and are treated to a physiologic range with monitoring, the cardiovascular fear that drove the 2015 warning was not supported by the data.

That finding is what the FDA acted on. On February 28, 2025, the agency directed sponsors of all testosterone products to remove the cardiovascular language from the boxed warning. The change applied class-wide — gels, patches, injectables, the new oral undecanoate products, the whole shelf.

The label change was not a clean bill of health. Three things stayed.

Blood pressure language was added, not removed. The FDA had also been reviewing postmarket ambulatory blood pressure monitoring (ABPM) studies submitted by sponsors. Those studies confirmed that testosterone products raise blood pressure modestly, across the class. The 2025 label keeps that warning and adds the specific ABPM data — so blood pressure monitoring is now in the label, not the cardiovascular composite warning.

The “limitation of use” for age-related low testosterone stayed. This is the line that says testosterone is approved for men with hypogonadism due to specific medical conditions (genetic, pituitary, testicular) — not for men whose levels have drifted lower with age and lifestyle. That distinction matters and we’ll come back to it.

The trial’s own secondary safety signals are on the label. This is the part the TRT marketing leaves out, and it’s the part patients deserve to hear plainly.

In a trial of 5,246 men, “non-inferior on the primary outcome” doesn’t mean nothing went wrong. TRAVERSE found a handful of adverse events that occurred more often in the testosterone group than placebo. They were small in absolute terms, but they were real, and the new label reflects them.

• Atrial fibrillation: 3.5% on testosterone vs. 2.4% on placebo. A modest but statistically meaningful uptick in non-fatal arrhythmias. AFib isn’t a heart attack, but it isn’t nothing — it raises stroke risk over time and often requires its own treatment.
• Pulmonary embolism: 0.9% on testosterone vs. 0.5% on placebo. Roughly double the rate, though both numbers are small in absolute terms.
• Acute kidney injury: 2.3% on testosterone vs. 1.5% on placebo. A signal worth knowing about in a population that already had higher baseline kidney risk.

There were also more fractures in the testosterone group, which was unexpected and remains under analysis.

None of these wipes out the headline finding. They do mean that “no MACE signal” should not be confused with “no risk.” A clinician reading TRAVERSE responsibly takes away two messages at once: the cardiovascular fear was overstated and afib, clotting, and kidney function deserve attention in any man on TRT. Both are true.

You may have seen claims that TRAVERSE showed testosterone reduces diabetes by 20% or more. That figure circulates in TRT clinic marketing. The full story is more complicated.

In February 2024, a pre-planned substudy was published in JAMA Internal Medicine — same data, focused on TRAVERSE participants with prediabetes or diabetes at baseline. At every time point through 48 months, progression from prediabetes to diabetes was numerically lower in the testosterone group (7.8% vs. 10.7% at 12 months; 10.1% vs. 14.6% at 24 months). But the formal statistical test across all time points was not significant (omnibus P = 0.49). The authors concluded that testosterone “should not be used as a therapeutic intervention to prevent or treat diabetes in men with hypogonadism.”

The honest read: a numerical trend in the right direction, but the trial designed to test it did not find that trend statistically meaningful. A clinic that pitches testosterone as a diabetes prevention strategy is reading the data more aggressively than the authors did.

This is where most of the public conversation goes sideways. TRAVERSE answered a specific question about a specific population. The further you get from that population, the less the trial actually tells you.

• Men under 45 weren’t in the trial. TRAVERSE enrolled men 45 to 80. The cardiovascular and arrhythmia data don’t cleanly extrapolate to a 32-year-old who started TRT after a “hormone optimization” telehealth visit.
• Healthy men without cardiovascular risk weren’t the focus. Every participant had either established CVD or high cardiovascular risk and documented hypogonadism. Generalizing the safety findings to healthy young men is a stretch in one direction; generalizing to even sicker, older men is a stretch in the other.
• The trial used topical gel. It did not test weekly cypionate injections, subcutaneous pellets, or oral undecanoate. The blood-level patterns differ — injections produce supraphysiologic peaks; pellets can run high for months — and those patterns are mechanistically what drive the afib and polycythemia signals at the bedside. TRAVERSE’s safety profile is the gel safety profile.
• The trial enrolled men with documented hypogonadism, confirmed properly. Two morning total testosterone levels under 300 ng/dL plus symptoms. Many men prescribed TRT today do not meet that bar. The American Urological Association’s current guideline still calls for two separate morning draws, drawn at the same lab, ideally two to four weeks apart, before diagnosis. A single borderline lab from a telehealth platform is not the same threshold the trial used.

For a man who actually meets the AUA criteria — documented low testosterone on two morning draws, with symptoms, after reversible causes (sleep apnea, obesity, opioid use, untreated thyroid disease) have been considered — TRAVERSE is reassuring. It is the largest, longest, most rigorously placebo-controlled answer we have to the cardiovascular safety question, and it came back favorable on the outcome most patients care about most.

But “reassuring” is not the same as “no monitoring required.” A reasonable TRT protocol still includes baseline labs, hematocrit checks (testosterone thickens blood — polycythemia), PSA monitoring, blood pressure tracking, follow-up labs to confirm levels landed in the physiologic range rather than above it, and a fertility conversation for men under 45 because exogenous testosterone shuts down the body’s own production. None of that is in TRAVERSE — but all of it is in good clinical practice.

If you’ve been on the fence about TRT because of the old cardiovascular warning, that warning is gone for a reason. If you’ve been pitched TRT online based on a single borderline lab and a symptom checklist, you’re not in the population TRAVERSE studied, and the trial’s results don’t fully apply to you. Both reads are honest.

At Naya Wellness, our nurse practitioners screen carefully before recommending testosterone — two morning draws, full hormone panel including LH/FSH and SHBG, screening for reversible causes, baseline hematocrit and PSA, and a clear conversation about what TRAVERSE shows and where it stops. Carolyn Pampalone, MSN, FNP-BC has spent nearly two decades in hormone care and can help you sort out whether TRT fits your situation. You can talk it through at our Chatsworth or Santa Barbara clinics, or by telehealth across California.

Two California locations and statewide telehealth — click a card to view details, or call to schedule a consult.