Testosterone Replacement: The Warning the FDA Removed, and the One It Added

In February 2025 the Food and Drug Administration did two things to testosterone product labelling on the same day. It removed the boxed warning about cardiovascular risk, and it added a new warning about blood pressure. Clinics selling testosterone have spent the time since quoting the first half of that sentence.

How is low testosterone properly diagnosed?

The American Urological Association sets the diagnostic threshold at a total testosterone below 300 ng/dL, which it calls a reasonable cut-off rather than a bright line. That is a moderate recommendation on Grade B evidence, and the hedging is deliberate.

The next statement is not hedged at all. Diagnosis should be made only after two separate total testosterone measurements, both taken early in the morning. That is a strong recommendation on Grade A evidence, the firmest category the guideline has. Testosterone follows a daily rhythm and peaks somewhere between three and eight in the morning. An afternoon sample can read low in a man whose levels are entirely normal, and a single sample of any kind can be low by chance.

This matters commercially. A telehealth service that diagnoses you from one convenient blood draw, at whatever hour you booked it, has skipped the strongest recommendation in the guideline. What the number actually means depends entirely on when it was taken and how many times.

What does TRT actually improve?

The guideline splits this cleanly, and the split is not what the advertising implies.

Look at the right-hand column and then look at any testosterone clinic’s homepage. Energy and fatigue are the two symptoms the entire industry is built on selling, and they sit in the column where the AUA says the evidence does not support a conclusion. Sex drive and erectile function, which men are often embarrassed to lead with, sit in the column where it does.

That is not the same as saying nothing happens. It means that when a man on testosterone says he feels less tired, nobody can currently tell him how much of that is the hormone. The mood question has been tested directly, and the result there is more interesting than either side usually admits.

Is TRT safe for your heart?

For roughly a decade the honest answer about testosterone and the heart was that nobody knew. TRAVERSE was built to settle it. More than 5,200 men aged 45 to 80, all with symptoms, all with testosterone below 300 ng/dL, and all with existing cardiovascular disease or a high risk of it, were randomised to testosterone gel or placebo. Mean treatment lasted 21.7 months and mean follow-up ran to 33 months.

The primary result is a genuine reassurance. A hazard ratio of 0.96 with a confidence interval from 0.78 to 1.17 met the trial’s non-inferiority threshold, in the population most likely to be harmed if harm existed. The decade of uncertainty about heart attacks has a reasonable answer now, and the answer is no.

The three rows underneath are the part that gets left out. Each is a small absolute increase, and each is real. An extra 1.1 percentage points of atrial fibrillation across a population is not a trivial finding, and pulmonary embolism nearly doubled off a low base. These are the sort of risks a man can reasonably accept for a treatment he needs. They are not risks anyone should accept for a treatment sold to him on the promise of more energy.

What did the FDA change on the label?

On 28 February 2025 the FDA acted on TRAVERSE and on a set of postmarket blood pressure studies. Four things changed at once:

• The cardiovascular boxed warning language was removed from testosterone products.
• A new warning about increased blood pressure was added to products that did not already carry one.
• Product-specific blood pressure data was added where ambulatory monitoring studies had been completed. The postmarket studies found the class raises blood pressure.
• The limitation of use for age-related hypogonadism was retained, and the TRAVERSE results were added to every label.

That fourth point is the one to hold on to. The FDA has not approved testosterone for men whose levels are low because they are older. It said so before TRAVERSE and it repeated it afterwards. A very large part of the testosterone market consists of exactly that prescription, written for exactly that reason.

And the swap itself is worth sitting with. A warning about heart attacks came off, and a warning about blood pressure went on, for a class of drug taken mostly by men in their fifties and sixties. Blood pressure is the highest-value number in men’s health, and a treatment that raises it is not a free trade even when the cardiac endpoint is clean.

What does TRT do to your fertility?

Exogenous testosterone suppresses the signal that tells the testicles to make sperm. This is not a rare side effect or an idiosyncratic reaction. It is how the drug works, and in some men the suppression does not fully reverse.

The AUA gives this the strongest weight it has. Clinicians should discuss the long-term impact on sperm production with any man who may want children, a strong recommendation on Grade A evidence. And testosterone should not be prescribed to men who are currently trying to conceive, also strong, also Grade A.

Men in their thirties and forties are prescribed testosterone routinely by services that ask nothing about their plans. If you are in that group and the subject has not come up, that is not a detail somebody forgot. The causes of low sperm count are worth understanding before you add a reversible one on purpose.

One route may be different. Nasal gel is dosed three times a day and cleared quickly, and a single uncontrolled trial found sperm production largely continued on it — a finding worth knowing about and worth reading sceptically. We set it against the other seven routes in injections, gel or pellets.

What monitoring does TRT require?

Testosterone thickens the blood. The guideline requires haemoglobin and haematocrit to be measured before treatment starts and the risk of polycythaemia explained, a strong recommendation on Grade A evidence. The pooled trial data behind it is stark: 19 events in 1,094 men on testosterone against a single event in 1,093 on placebo. Injectable preparations produce the largest rises.

A PSA is also expected before starting in men over 40. On the prostate itself the guideline is clearer than the folklore, and it runs the other way: clinicians should inform patients of the absence of evidence linking testosterone therapy to the development of prostate cancer. The old belief that testosterone causes prostate cancer is not supported, which is a rare case of the evidence being more permissive than the reputation.

None of this is an argument against testosterone for the men who need it. For a man with symptoms, two confirmed low morning readings, and no wish to father children, it is a reasonable treatment with a now-reasonable safety file. The argument is against the version being sold to men who have none of those things, on the promise of the one benefit the guideline says the evidence cannot confirm. If low energy is the symptom, the cheaper explanations are worth excluding first.

This article is for information only and is not medical advice. It cannot account for your individual circumstances. Talk to a doctor about your own situation, particularly before starting or stopping any prescription medication.