TRT, fertility and long-term safety

Does TRT cause infertility?

It suppresses sperm production, yes, and that's a predictable effect of the treatment rather than a rare complication.

The mechanism is straightforward. Taking testosterone from outside tells the pituitary it doesn't need to send its own signals, so LH and FSH fall. FSH is what drives sperm production, and testosterone inside the testis has to be far higher than in the bloodstream for it to work. Replace one and you suppress the other.

Product labelling describes it as reversibly interrupting or reducing spermatogenesis, and notes that it may lead to azoospermia, meaning no measurable sperm at all.

Is it reversible?

In the best available data, sperm production usually recovers after testosterone is stopped, but the timing and completeness of recovery vary.

The strongest evidence is an integrated analysis of 30 studies covering 1,549 men. It pooled recovery after hormonal suppression of sperm production. Median time to recover a sperm concentration of 20 million per millilitre was 3.4 months. Two thirds of men reached that within six months, 90 per cent within twelve months, 96 per cent within sixteen months, and effectively all within twenty-four.

There's an important limitation. Those men were healthy young volunteers in male contraceptive trials, not middle-aged men on long-term replacement therapy, so we'd treat the curve as encouraging rather than as a promise.

One finding in it is directly relevant here. Recovery was faster in men of Asian origin, alongside older age, shorter treatment duration and shorter-acting preparations. The analysis found an association between Asian ethnicity and faster recovery, although this finding should not be interpreted as a guarantee for an individual patient.

What if I want children?

The first and clearest option is not to start, and every guideline says so.

The three main guidelines agree, and all three treat it as a firm position rather than a caution.

The most cited work on giving hCG alongside testosterone is a retrospective review of 26 men with no control group. Semen parameters held steady, no man became azoospermic, and nine of the 26 contributed to a pregnancy. Promising, and nowhere near established.

Does it raise heart risk?

In the TRAVERSE trial, testosterone was not associated with a higher incidence of major adverse cardiovascular events than placebo in the studied population.

TRAVERSE randomised 5,246 men aged 45 to 80 who had hypogonadism symptoms and either existing cardiovascular disease or high risk of it. They were given transdermal testosterone gel or placebo and followed for a mean of 33 months.

A primary cardiovascular event occurred in 7.0 per cent on testosterone and 7.3 per cent on placebo, a hazard ratio of 0.96 with a confidence interval from 0.78 to 1.17. The trial met its non-inferiority endpoint.

On the strength of that, the FDA removed the cardiovascular language from the boxed warning on testosterone products in February 2025. So if you read somewhere that testosterone carries a boxed warning for heart risk, the FDA's 2025 class-wide labelling changes should be considered when interpreting older information.

Three things sit alongside that result. TRAVERSE reported higher rates of some secondary events on testosterone than placebo.

Separately, a Testosterone Trials substudy scanned coronary arteries in 138 men and found non-calcified plaque volume increased more on testosterone than placebo, a difference of 41 cubic millimetres. There were no cardiovascular events in either group, and this result sits in genuine tension with TRAVERSE. Both are real and we'd carry both.

Lastly, blood pressure. The FDA stated in 2025 that ambulatory monitoring studies confirmed an increase in blood pressure with testosterone across the class. For the oral form the reported rise was about 4.9 over 2.5 millimetres of mercury, and about 5.4 over 3.2 in men already on treatment for hypertension.

That is a reason to have your blood pressure watched rather than a reason not to treat.

What about prostate cancer?

The evidence does not show that testosterone therapy causes prostate cancer, and it also cannot rule it out.

The TRAVERSE prostate substudy followed 5,204 men for a mean of about 22 months. High-grade prostate cancer occurred in 5 men on testosterone and 3 on placebo, a hazard ratio of 1.62 with a confidence interval running from 0.39 to 6.77.

Read that interval rather than the point estimate. With eight events in total, the data are compatible with a modest protective effect and with a six-fold increase in risk. It is too imprecise to establish whether testosterone meaningfully changes the risk of high-grade prostate cancer.

There is a second limitation that matters more for you personally. The trial excluded men whose PSA was above 3.0 nanograms per millilitre. So it says nothing about that group, which is exactly the group most likely to be asking.

The AUA position is that clinicians should inform patients of the absence of evidence linking testosterone therapy to the development of prostate cancer. The Endocrine Society recommends against testosterone in men with prostate cancer.

The EAU treats locally advanced or metastatic disease as an absolute contraindication, and for men treated for localised disease suggests waiting at least a year with an undetectable PSA and no sign of recurrence.

The FDA moved in June 2026 to narrow the label contraindication to metastatic prostate cancer only. So the label and the specialty guidelines are currently pointing in slightly different directions, which is a reason to make this a conversation rather than a rule.

Does it protect my bones?

It improves bone density. It did not reduce fractures, and this is the single most important nuance in the whole safety picture.

The Testosterone Trials measured bone properly, with volumetric imaging in 211 men over twelve months. Trabecular bone density rose 7.5 per cent at the spine and 10.8 per cent at the hip compared with placebo, and estimated bone strength increased at both sites. Those are substantial numbers.

Then a TRAVERSE subtrial looked at what actually happens to bones, in 5,204 men over a median of 3.19 years. Clinical fractures occurred in 3.50 per cent on testosterone and 2.46 per cent on placebo, a hazard ratio of 1.43 with a confidence interval from 1.04 to 1.97. The incidence also appeared higher for every other fracture endpoint measured.

Bone density improved, but clinical fractures were more frequent in the testosterone group in TRAVERSE. The reason for this discrepancy is not established.

In practice, we don't offer testosterone as a bone treatment, and if osteoporosis is the problem, there are drugs with fracture data behind them.

What about blood count?

It changes it reliably, and it's the effect that is most likely to change your dose.

Pooled analysis found haemoglobin rose by about 0.8 grams per decilitre and haematocrit by about 3.2 percentage points on treatment. For most men, that's harmless. For some, it pushes the blood thicker than we want it.

The action threshold is clear. A haematocrit of 54 per cent or above requires stopping treatment, reducing the dose, changing formulation, or venesection.

A baseline of elevated haematocrit warrants assessment before treatment, and a baseline haematocrit above the guideline threshold is generally a reason to withhold treatment until the cause is evaluated and corrected.

This is why haematocrit is checked at baseline, three, six and twelve months, then annually, rather than being a one-off. The effect builds over the first year, peaking somewhere between nine and twelve months.

There's also a fair case for checking whether venous thromboembolism runs in your family before starting, which the EAU recommends, given the TRAVERSE numbers above.

What else should I know?

Three smaller things that come up.

Untreated severe obstructive sleep apnoea is a contraindication or important caution in major guidelines, and testosterone therapy may worsen sleep-disordered breathing in some patients.

Gynaecomastia, meaning breast tissue development, is listed as uncommon in trial data, which in labelling convention means somewhere between one in a thousand and one in a hundred. It can persist.

And if you're using a gel rather than an injection, it transfers by skin contact. That can raise testosterone levels in a partner or a child, so the site gets washed and covered and you wait at least an hour before close contact. It sounds fussy until you consider who is most likely to be affected.

What if I stop suddenly?

Your own production doesn't switch straight back on, so testosterone levels and endogenous production may remain suppressed for a period after stopping, and the original symptoms may return. The duration varies with the formulation, duration of treatment and individual factors.

You'll find label wording describing a withdrawal syndrome with low mood, fatigue, irritability, and reduced libido lasting weeks or months. Read in context, that describes men taking doses far above replacement, which is a different situation from stopping replacement therapy.

There is no trial data on how quickly symptoms return after a planned stop at replacement doses. In practice your doctor steps the dose down and monitors you.

The practical instruction is simpler. If you want to stop, tell us and we'll plan it. Stopping without telling anyone is how a manageable transition becomes a bad three months.

The bottom line

Fertility is the objection that holds, and if you want children soon this is not the treatment to start. The cardiovascular picture improved with TRAVERSE, and the fracture and prostate readings are less tidy, so both belong in the conversation.