A Warning That Changed Medical Practice
Ask a primary care physician whether testosterone therapy is safe for the heart and you will likely get a careful pause. On March 3, 2015, the FDA placed a cardiovascular warning on every testosterone product sold in the United States, citing "a possible increased risk of heart attacks and strokes." The agency simultaneously narrowed the approved indication, noting that benefits and safety had not been established for age-related testosterone decline. Physicians pulled back. Patients stopped asking. Within three years, testosterone prescriptions among established users had fallen 48%, and new prescriptions had plummeted 62%.
The warning was based on two observational studies published in rapid succession in late 2013 and early 2014. Neither was a randomized trial. Both had serious methodological problems that emerged quickly but influenced policy slowly. And when the FDA finally got the gold-standard evidence it demanded, the result contradicted the premise that had justified the warning in the first place.
The Studies That Triggered a Decade of Caution
The first study landed on November 6, 2013, when Rebecca Vigen and colleagues published a retrospective analysis in JAMA of 8,709 male veterans with low testosterone who had undergone coronary angiography. The paper reported that testosterone therapy was associated with a 29% increase in a composite of death, heart attack, and stroke (HR 1.29, 95% CI 1.04β1.58). Media coverage was alarming and immediate.
Less than three months later, William Finkle and colleagues published a study in PLOS ONE using insurance claims data showing an elevated risk of nonfatal myocardial infarction in the 90 days following a testosterone prescription. The FDA issued a safety communication on January 31, 2014, and convened an advisory committee that September. By March 2015, the label change was final.
The problems with the Vigen study surfaced rapidly. It underwent two published corrections after investigators discovered that nearly 10% of the supposedly all-male dataset was contaminated by women, that over 1,000 men had been erroneously excluded from one comparison group, and that the raw adverse event rate was roughly half as high in the testosterone group (10.1%) as in the untreated group (21.2%). The study's opposite conclusion relied on statistical adjustments for more than 50 variables using a methodology its own senior author had described as needing "further study" just one year earlier. Twenty-nine medical societies formally petitioned JAMA for retraction, citing "gross data mismanagement and contamination." JAMA declined. The Finkle study contained only prescription-fill records with no data on whether patients actually took the medication, no information on hypogonadal symptoms, and no controls for baseline cardiovascular risk.
The Trial the FDA Demanded
The FDA's 2015 guidance required testosterone manufacturers to conduct a cardiovascular safety trial. The result was TRAVERSE: the Testosterone Replacement Therapy for Assessment of Long-term Vascular Events and Efficacy Response in Hypogonadal Men. Conducted at 316 clinical sites across the United States between 2018 and 2022, TRAVERSE enrolled 5,246 men aged 45 to 80 who had either preexisting cardiovascular disease (54%) or multiple cardiovascular risk factors (46%), and who had two confirmed fasting testosterone levels below 300 ng/dL. Participants were randomized 1:1 to daily transdermal testosterone gel or matching placebo gel. The mean treatment duration was 22 months and mean follow-up was 33 months.
What the Trial Found
The primary safety endpoint was a composite of cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke. Among 5,198 patients who received at least one dose, a primary event occurred in 182 patients (7.0%) in the testosterone group and 190 patients (7.3%) in the placebo group: a hazard ratio of 0.96 (95% CI 0.78 to 1.17, P<0.001 for noninferiority). The point estimate actually favored testosterone, though not significantly. Cardiovascular death occurred in 3.4% of the testosterone group versus 4.0% of the placebo group (HR 0.84, 95% CI 0.63β1.12). Death from any cause was nearly identical at 5.5% versus 5.7%.
The finding was not isolated. The Androgen Society's 2024 position paper documented that TRAVERSE's conclusion aligned with two additional large randomized controlled trials, multiple smaller RCTs, several large observational studies, and 19 prior meta-analyses. "It has now been conclusively determined," the Society stated, "that testosterone therapy is not associated with increased risks of heart attack, stroke, or CV death."
The Human Cost of a Flawed Signal
Jacques Baillargeon and colleagues at the University of Texas Medical Branch tracked testosterone prescribing among nearly 10 million insured men from 2002 through 2016. At the 2013 peak, 3.20% of men aged 30 to 64 were using testosterone; by 2016, the rate had fallen to 1.67%. New prescriptions dropped from 1.26% to 0.48% per year. Scale those percentages to the approximately 66 million American men in that age range, and the decline represents roughly one million fewer men receiving treatment annually. Over the eight years between the FDA warning and the TRAVERSE publication, that amounts to millions of man-years of untreated hypogonadism: fatigue, muscle wasting, depressed mood, sexual dysfunction, and accelerating bone density loss in men who had a clinically treatable condition but were told the treatment might kill them.
The Strongest Counterargument
The most serious critique of TRAVERSE comes from within its own data. While the trial cleared testosterone of major cardiac events, it found significantly higher rates of atrial fibrillation in the testosterone group (3.5% vs. 2.4%, P=0.02), acute kidney injury (2.3% vs. 1.5%, P=0.04), and a non-significant trend toward more pulmonary embolism (0.9% vs. 0.5%). These are not trivial findings. A separate sub-study from the earlier TTrials consortium found that testosterone therapy increased non-calcified coronary artery plaque volume, a marker some cardiologists view as a precursor to future events that may take years to manifest. The trial's sponsor was AbbVie, the leading testosterone manufacturer, which had an obvious financial stake in the outcome. And the 33-month mean follow-up cannot rule out cardiovascular risks that emerge only after five or ten years of continuous therapy.
What We Didn't Prove
TRAVERSE tested transdermal testosterone gel in men with confirmed hypogonadism who were already at high cardiovascular risk. It did not study testosterone injections, which produce higher peak serum levels than gel. It did not enroll men under 45 or men without cardiovascular disease or risk factors. The 61% discontinuation rate in both groups reflects real-world adherence but also means the effective treatment exposure was lower than a fully compliant population would experience. The noninferiority margin of 1.5 for the hazard ratio means the trial was not designed to detect a true risk increase smaller than about 17%. And the atrial fibrillation, kidney injury, and venous thromboembolic signals deserve dedicated follow-up trials with longer observation windows before they can be fully characterized.
The Bottom Line
For nearly a decade, a cardiovascular warning built on two deeply flawed observational studies suppressed testosterone prescribing across the United States. When the FDA got the randomized trial it demanded, enrolling the highest-risk patients it could specify, the result was clear: testosterone therapy did not increase heart attacks, strokes, or cardiovascular death. The remaining safety signals around atrial fibrillation and blood clots are real and worth monitoring, but they are a different category of risk from the catastrophic cardiac narrative that drove a decade of clinical paralysis.
What You Can Do
If you have symptoms of low testosterone and have avoided evaluation because of cardiovascular concerns, discuss the TRAVERSE results with your physician. The evidence now supports that testosterone-replacement therapy does not increase the risk of major cardiac events, even in men with preexisting heart disease. If you are currently receiving testosterone therapy, ask about periodic monitoring for atrial fibrillation and kidney function, both of which showed elevated rates in the trial. If you are a clinician still operating under the assumptions of the 2015 warning, review the Androgen Society's 2024 position paper and the European Expert Panel's consensus statement. The clinical question has shifted: it is no longer whether testosterone therapy is safe for the heart, but whether it is the right treatment for each individual patient.