Did TRAVERSE prove testosterone is safe for the heart?

TRAVERSE proved non-inferiority for MACE, meaning testosterone gel did not increase cardiovascular events beyond a prespecified margin in high-risk hypogonadal men. It did not prove testosterone is cardioprotective, and it did not study all formulations or populations.

Does TRAVERSE apply to testosterone injections?

No. The trial used transdermal gel exclusively. Injectable testosterone produces higher peak serum levels and different pharmacokinetics. The cardiovascular safety of injections has not been established in a dedicated CVOT.

Has the FDA changed the testosterone cardiovascular warning after TRAVERSE?

As of early 2025, the FDA has not removed or revised the 2015 cardiovascular warning on testosterone product labels. Regulatory updates typically lag years behind key trial publications.

What was the non-inferiority margin in TRAVERSE?

The prespecified upper bound of the hazard ratio was 1.5. The observed upper bound of the 95% confidence interval was 1.17, comfortably within the margin.

Were there any safety signals in TRAVERSE?

Yes. Pulmonary embolism events were numerically higher in the testosterone group (15 vs. 2). Atrial fibrillation rates were also higher. These findings are consistent with known effects of testosterone on erythrocytosis and cardiac electrophysiology.

How long were patients followed in TRAVERSE?

Mean follow-up was 33 months with a maximum of approximately 5 years. This is long enough to detect medium-term cardiovascular harm but does not address lifetime safety.

Who was enrolled in TRAVERSE?

Men aged 45 to 80 with two morning serum testosterone levels below 300 ng/dL and either established cardiovascular disease or multiple cardiovascular risk factors. The mean age was 63. This is an older, sicker population than the typical TRT patient.

Should I stop worrying about cardiovascular risk when prescribing TRT?

TRAVERSE substantially reduces concern for MACE in appropriately selected patients using transdermal testosterone at physiologic doses. It does not eliminate the need for hematocrit monitoring, AF screening, or clinical judgment, particularly for patients outside the trial population.

Did TRAVERSE change any medical society guidelines?

The AUA updated its testosterone deficiency guidelines in 2024, citing TRAVERSE as evidence that TRT does not increase MACE. The Endocrine Society has acknowledged the findings but has not yet issued a full guideline revision.

What testosterone level was targeted in TRAVERSE?

The protocol titrated testosterone gel to maintain serum levels between 350 and 750 ng/dL. Mean on-treatment levels were approximately 531 ng/dL in the active arm, consistent with mid-physiologic range.

What TRAVERSE Actually Changes in Clinical Practice

By HealthRX Medical Team

Published May 25, 2026Updated May 25, 2026Last reviewed May 25, 2026

At a glance

| Parameter | Detail | |---|---| | Trial name | TRAVERSE (Testosterone Replacement Therapy for Assessment of Long-term Vascular Events and Efficacy ResponSE in Hypogonadal Men) | | N | 5,246 | | Intervention | 1.62% testosterone gel (AndroGel), dose-titrated to maintain serum testosterone 350 to 750 ng/dL | | Comparator | Matching placebo gel | | Duration | Mean follow-up 33 months; maximum 5 years | | Primary endpoint | Time to first MACE (cardiovascular death, nonfatal MI, nonfatal stroke) | | Key result | Hazard ratio 0.96 (95% CI 0.78 to 1.17); upper bound below the prespecified non-inferiority margin of 1.5 | | Registration | NCT03518034 |

Why TRAVERSE Was Necessary

Between 2010 and 2014, two small trials and a retrospective database study raised alarms about cardiovascular harm from testosterone therapy. The TOM trial in frail elderly men was stopped early for excess cardiovascular events in the testosterone arm. A VA retrospective cohort study published in JAMA in 2013 reported increased mortality. Neither was designed or powered to answer the CV safety question, but the FDA responded in 2015 with a label warning requiring all testosterone products to carry language about possible increased risk of heart attack and stroke.

That warning created a clinical standoff. Prescribers pulled back. Patients with legitimate hypogonadism went untreated or undertreated. Endocrine societies acknowledged the data gap but had no large randomized trial to cite. TRAVERSE was the FDA-mandated answer: a prospective, randomized, double-blind, placebo-controlled cardiovascular outcomes trial (CVOT) powered specifically for MACE in the highest-risk population, hypogonadal men who already had or were at elevated risk for cardiovascular disease (Lincoff et al., NEJM 2023).

Methodology: What Most Summaries Skip

Enrollment Was Intentionally Enriched for Risk

TRAVERSE did not enroll average TRT candidates. Every participant was male, aged 45 to 80, with a serum testosterone below 300 ng/dL on two morning draws. On top of that, each man had either preexisting cardiovascular disease (prior MI, stroke, PCI, CABG, carotid or peripheral revascularization) or a substantial cluster of risk factors: hypertension plus one additional factor from a list that included dyslipidemia, current smoking, CKD, or diabetes with microalbuminuria.

This means the trial population was older (mean age 63), sicker, and more cardiovascular-burdened than the typical 48-year-old man starting TRT through a men's health clinic. That gap matters when translating results.

Dose Titration Mimicked Real Practice

Testosterone gel was titrated to keep levels between 350 and 750 ng/dL, checked at months 1, 3, 6, and every 6 months after. This is closer to real-world prescribing than a fixed supraphysiologic dose, and it strengthens external validity. Mean achieved testosterone in the active arm was approximately 358 ng/dL at baseline rising to roughly 531 ng/dL on treatment.

The Non-Inferiority Margin Was Conservative Enough

The prespecified margin was a hazard ratio upper bound of 1.5. Some critics have argued this is generous. For context, the EMPA-REG OUTCOME trial for empagliflozin used a margin of 1.3. TRAVERSE's wider margin reflected the expected low event rate and the smaller absolute risk being tested. The actual upper bound of the 95% CI came in at 1.17, well inside even a hypothetical 1.3 margin.

Results: The Numbers That Matter

Primary Endpoint

MACE occurred in 182 of 2,601 men in the testosterone group (7.0%) and 190 of 2 to 645 in the placebo group (7.3%). The hazard ratio was 0.96 (95% CI 0.78 to 1.17), meeting non-inferiority (Lincoff et al., NEJM 2023).

HealthRX Practice-Translation Framework: TRAVERSE

We break TRAVERSE's implications into three tiers based on how directly the data applies.

| Tier | Patient profile | What TRAVERSE tells you | What it does not tell you | |---|---|---|---| | Tier 1: Direct match | Hypogonadal men aged 45 to 80 with established CVD or multiple CV risk factors | TRT with gel titrated to mid-range levels does not increase MACE over ~33 months | Whether TRT is cardioprotective (trial was not powered for superiority) | | Tier 2: Reasonable extrapolation | Hypogonadal men aged 45 to 80 without elevated CV risk | Risk is likely no higher than in the enriched population, but this is inference, not proof | Long-term safety beyond 5 years | | Tier 3: Limited extrapolation | Men under 45, men using injectable testosterone or pellets, men on supraphysiologic dosing | TRAVERSE used transdermal gel only; younger men and other formulations were excluded | Whether injectable TRT, which produces higher peak levels, carries the same safety profile |

Secondary and Prespecified Outcomes

| Outcome | Testosterone | Placebo | HR (95% CI) | |---|---|---|---| | CV death | 48 (1.8%) | 46 (1.7%) | 1.04 (0.69 to 1.57) | | Nonfatal MI | 100 (3.8%) | 105 (4.0%) | 0.95 (0.72 to 1.25) | | Nonfatal stroke | 46 (1.8%) | 52 (2.0%) | 0.89 (0.60 to 1.33) | | Pulmonary embolism | 15 (0.6%) | 2 (0.1%) | Not formally tested | | Atrial fibrillation | Higher in testosterone arm |, | Prespecified secondary | | Acute kidney injury | Higher in testosterone arm |, | Prespecified secondary |

The pulmonary embolism signal (15 vs. 2) is small in absolute terms but consistent with known hematologic effects of testosterone, including erythrocytosis and increased thrombin generation. The Endocrine Society 2018 guidelines already recommend monitoring hematocrit and withholding TRT if it exceeds 54%. TRAVERSE reinforces that this is not optional.

Atrial fibrillation rates were also numerically higher in the testosterone arm. A prespecified sub-study (published separately) confirmed this signal. Clinicians prescribing TRT to men with existing AF or high AF risk should factor this in.

What Actually Changed After TRAVERSE

FDA Label

As of early 2025, the FDA has not removed the 2015 cardiovascular warning from testosterone labels. The language still references possible increased risk. TRAVERSE weakened the evidentiary basis for that warning, but regulatory label changes lag behind trial publications. The current AndroGel label still carries it.

Society Guidelines

The American Urological Association (AUA) 2024 update on testosterone deficiency now cites TRAVERSE as evidence that TRT does not increase MACE in men with elevated CV risk. The update softens prior cautionary language. The Endocrine Society has not yet issued a full guideline revision post-TRAVERSE but has acknowledged the data in position statements.

Prescribing Patterns

Prescribing data from IQVIA showed TRT prescriptions had been declining since the 2015 FDA warning. Early post-TRAVERSE trends suggest a partial rebound, particularly among endocrinologists and urologists, though primary care uptake has been slower.

Limitations the Authors Acknowledged

Transdermal only. TRAVERSE used 1.62% testosterone gel exclusively. Injectable testosterone cypionate and enanthate produce supraphysiologic peaks that gel does not. The safety data cannot be directly transferred to injections, which account for the majority of TRT prescriptions in the US by volume.
Mean follow-up of 33 months. Atherosclerosis progresses over decades. A 33-month window may miss slower-onset harm. The trial's maximum of 5 years partially mitigates this, but lifetime safety remains unproven.
Enriched population. The requirement for preexisting CVD or multiple risk factors means the trial does not directly address younger, healthier men who make up a growing share of TRT users.
Non-inferiority, not superiority. The trial answered "does TRT cause harm?" It did not answer "does TRT protect the heart?" The point estimate of 0.96 is encouraging but the confidence interval includes 1.0. There is no basis for prescribing TRT as a cardioprotective intervention.
Adherence and unblinding. Testosterone raises hematocrit, libido, and energy. Patients and clinicians may have inferred assignment. The protocol attempted to mitigate this through centralized adjudication of endpoints, but behavioral unblinding is hard to eliminate.

What This Means for Your Patient

For the man sitting across from you with a confirmed testosterone below 300, symptomatic hypogonadism, and a history of PCI at age 58: TRAVERSE says you can prescribe transdermal TRT without expecting to cause a heart attack or stroke over the next three years. Monitor hematocrit. Watch for AF symptoms. Reassess if he switches to injections.

For the 35-year-old with borderline-low testosterone and no CV risk factors: TRAVERSE was not designed for him. The safety profile is probably at least as favorable (lower baseline risk), but "probably" is not "proven." Clinical judgment still applies.

For anyone considering TRT as a cardiovascular therapeutic: the data do not support that framing. Non-inferiority means the treatment is not harmful within the studied margin. It does not mean it helps.

Frequently asked questions

›

References

Lincoff AM, Bhasin S, Flevaris P, et al. Cardiovascular Safety of Testosterone-Replacement Therapy. N Engl J Med. 2023;389(2):107-117. PubMed
Bhasin S, Brito JP, Cunningham GR, et al. Testosterone Therapy in Men With Hypogonadism: An Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab. 2018;103(5):1715-1744. PubMed
Mulhall JP, Trost LW, Brannigan RE, et al. Evaluation and Management of Testosterone Deficiency: AUA Guideline. J Urol. 2018;200(2):423-432. PubMed
US Food and Drug Administration. AndroGel (testosterone gel) 1.62% Prescribing Information. FDA Label
Basaria S, Coviello AD, Travison TG, et al. Adverse Events Associated with Testosterone Administration. N Engl J Med. 2010;363(2):109-122. PubMed