If you want children — now, or at any point in the future — say so before you start testosterone.
Not after the first injection. Not at the three-month follow-up. Before.
This is the single most consequential conversation in testosterone medicine, and it is the one most often skipped.
Testosterone therapy is an effective male contraceptive. That is not a rhetorical framing — it is literally how the drug was studied for two decades. The World Health Organization ran multinational trials in the 1990s testing weekly testosterone injections as a reversible method of male birth control, and the reason those trials were run is that testosterone works.
Men are prescribed testosterone every day without being told this. A survey of 2,983 people conducted by Legacy and Carrot found that nearly three-quarters of men on hormone therapy were unaware of its potential effect on their fertility. That is survey data rather than a controlled study, and it should be read as such — but it matches what reproductive urologists see in clinic, which is men arriving having been given a fertility-suppressing drug that nobody described to them as one.
Why testosterone therapy suppresses fertility
Sperm production doesn’t run on the testosterone in your blood. It runs on the testosterone inside your testes, and the difference between the two is enormous.
In a study of 29 healthy men published in the Journal of Clinical Endocrinology & Metabolism, intratesticular testosterone measured approximately 1,234 nmol/L, against a serum concentration of around 14.1 nmol/L. Testosterone inside the testis sits roughly 80 to 100 times higher than the level in the bloodstream. Spermatogenesis requires that gradient.
The gradient is maintained by luteinizing hormone from the pituitary, which stimulates the Leydig cells to produce testosterone locally. When you inject testosterone, the pituitary reads the elevated blood level, concludes that no further stimulation is needed, and shuts down LH and FSH. Leydig cells stop producing. Intratesticular testosterone collapses.
In that same study, men given testosterone enanthate without any supporting treatment saw intratesticular testosterone fall to 72 nmol/L — a 94% reduction.
So the paradox that confuses most men resolves cleanly: your blood testosterone is now excellent, and your testes are in a hormonal environment that cannot sustain sperm production. A serum testosterone level tells you nothing whatsoever about intratesticular testosterone. A man on testosterone therapy can have a level of 900 ng/dL and no sperm at all.
FSH suppression compounds the problem, since FSH acts on the Sertoli cells that support developing sperm directly.
How reliably does this happen?
Reliably enough that it was pursued as contraception.
In the World Health Organization trials of weekly testosterone enanthate, 65% of men (157 of 241) became azoospermic — no sperm in the ejaculate at all — with a mean time to azoospermia of around 120 days. Most of the remainder were severely suppressed without reaching zero.
There is a striking population difference in these data that is worth knowing if it applies to you. East Asian men suppressed far more completely than European men on identical regimens: azoospermia in approximately 91% of Chinese men versus around 60% of non-Asian men. In a pooled analysis of 1,756 men, Caucasian men suppressed to below one million per millilitre in up to 80% of cases, East Asian men in up to 90%. Among men who reached azoospermia, East Asian men stayed azoospermic more consistently — 95% versus 68%.
The practical implication cuts both ways. Testosterone is not a reliable enough contraceptive to use as one — a substantial minority of men retain some sperm production, which is precisely why it never came to market as birth control. And it is far too reliable a suppressant to take casually if you want to conceive.
Is it reversible?
Usually. Not always. And “usually” is doing real work in that sentence.
The best data come from an integrated analysis by Liu and colleagues published in The Lancet in 2006, pooling recovery data across 30 studies of healthy men who had received testosterone in contraceptive trials.
Median time to recovery:
- to 3 million sperm/mL — 2.5 months
- to 10 million/mL — 3.0 months
- to 20 million/mL — 3.4 months
Probability of recovering to at least 20 million/mL:
- 67% by 6 months
- 90% by 12 months
- 96% by 16 months
- 100% by 24 months
Those are reassuring numbers, and the AUA cites essentially the same figures in its guideline: roughly two-thirds of men in contraceptive studies recovered sperm within six months, with about 10% not doing so until the second year.
But read the population carefully. These were healthy, young, eugonadal volunteers, given testosterone for a defined study period, under supervision. That is not the same as a 45-year-old man who has been on testosterone cypionate for six years because a clinic put him there.
Recovery was faster in men with shorter treatment duration, higher baseline sperm concentration, shorter-acting testosterone preparations, faster initial suppression, and lower baseline LH. Asian ethnicity also predicted faster recovery. Counterintuitively, in this analysis older age predicted faster recovery — a finding that sits awkwardly against clinical experience with long-term therapeutic users, and one reason not to over-extrapolate contraceptive-trial data to the clinic.
Data from men stopping clinical testosterone therapy rather than trial participation suggest a less rosy picture, with longer treatment duration and older age both associated with failure to recover. We have seen figures reported for this but have not been able to verify them against the primary publication, so we are not going to quote them. What can be said honestly: the longer you have been on testosterone, the less confident anyone can be about your recovery.
What “recovery” actually means
Two distinctions matter here and both get lost.
Sperm count recovery is not fertility. Returning to 20 million sperm per millilitre is a laboratory milestone. Whether you and your partner conceive depends on motility, morphology, DNA integrity, her age and fertility, and timing. A recovered count is a necessary condition, not a sufficient one.
Nothing happens quickly, even once the axis restarts. A full cycle of spermatogenesis takes approximately 74 days, and newly produced sperm then spend a further two to three weeks in epididymal transit before appearing in the ejaculate. So even in the best case, roughly three months elapse between the testes resuming production and that production becoming measurable. Patience is not optional; it is physiology.
For reference, the World Health Organization’s 2021 sixth-edition lower reference limits for semen analysis are: volume 1.4 mL, sperm concentration 16 million/mL, total sperm number 39 million per ejaculate, progressive motility 30%, vitality 54%, and normal morphology 4%. These are fifth-centile values derived from men who conceived within twelve months — they describe the bottom of the fertile range, not a target.
Sperm banking
If you have any intention of having biological children and you are about to start testosterone, bank sperm first.
It is inexpensive relative to the alternative, it takes very little time, and cryopreserved sperm remains viable for many years. Banking before treatment converts an irreversible-in-some-men risk into a manageable one. There is no clinical argument against it for a man who wants children and every practical argument for it.
The reason it doesn’t happen is not clinical disagreement. It is that the conversation never occurs.
hCG for fertility preservation
Human chorionic gonadotropin mimics LH. Given alongside testosterone, it stimulates the Leydig cells directly, maintaining intratesticular testosterone even while the pituitary is suppressed.
The mechanistic case is strong. In the dose-ranging study cited earlier, hCG restored intratesticular testosterone in a dose-dependent fashion in men whose gonadotropins had been suppressed by testosterone: 125 IU every other day brought it to roughly 883 nmol/L — still about 25% below baseline; 250 IU restored it to near baseline at about 1,092 nmol/L; 500 IU pushed it above baseline at around 1,478 nmol/L.
The clinical evidence is thinner than the mechanism suggests. The most-cited study, by Hsieh and colleagues in the Journal of Urology in 2013, was retrospective and included 26 men on testosterone plus hCG 500 IU every other day. Testosterone rose from 207 to 1,056 ng/dL. No patient became azoospermic, and sperm density, motility and volume did not significantly change over more than a year. Nine of the 26 achieved pregnancy.
That is a genuinely encouraging result. It is also 26 men, retrospective, uncontrolled. There is no randomized trial of hCG for fertility preservation during testosterone therapy. Dosing in practice — commonly 500 to 1,500 IU two or three times weekly — is extrapolated from small physiology studies rather than derived from dose-ranging trials with fertility endpoints.
Evidence: 🟡 Promising. Not Established. If a clinic presents hCG as a guaranteed solution that makes testosterone fertility-safe, they are overstating what 26 retrospective patients can support.
Clomiphene and enclomiphene
A different strategy: instead of replacing testosterone, stimulate the man’s own production by blocking estrogen feedback at the pituitary. LH and FSH rise, the testes produce more testosterone, and — critically — the intratesticular gradient is preserved because the stimulation is coming from above rather than being overridden from outside.
Clomiphene citrate is used off-label for this. Retrospective series report testosterone roughly doubling with sperm parameters maintained. The evidence is observational — no randomized controlled trials with hard endpoints, and essentially no data beyond a couple of years. Clomiphene is a mixture of two isomers: enclomiphene, which drives the testosterone increase, and zuclomiphene, which is weakly estrogenic, has a much longer half-life, and is implicated in the visual and mood side effects some men report.
Enclomiphene isolates the active isomer. The comparison data are striking. In phase II randomized trials involving 240 men, enclomiphene raised testosterone to around 450 ng/dL while sperm concentration was unchanged from baseline at 16 weeks, with oligospermia in fewer than 5%. Men on testosterone gel over the same period saw sperm concentration fall by around 60%, with oligospermia in more than half.
That is close to a clean demonstration of the principle: raise testosterone from above and fertility survives; replace it from outside and it doesn’t.
So why isn’t everyone using it? Enclomiphene is not FDA-approved. It was developed as Androxal, received multiple Complete Response Letters from the FDA — over assay issues, isomer-ratio variability, and manufacturing controls — and the required phase III programme was never completed. The company was acquired and the programme lapsed.
It remains available only through compounding pharmacies, which are not required to demonstrate the isomer purity that the FDA specifically found wanting. The defect that sank the approval — inconsistent zuclomiphene contamination — is exactly the defect a compounded product is least equipped to exclude.
Evidence: 🟠 Limited. The mechanism is sound and the short-term trial signal is genuinely good. There is no approved product, no long-term data, and no regulated supply chain. Men prescribed compounded enclomiphene should understand they are taking an unapproved drug of unverified composition, which may still be a reasonable decision — but it should be a decision, not an assumption.
Gonadotropin therapy
For men with secondary hypogonadism who want fertility, the more direct approach replaces the missing pituitary signals: hCG to substitute for LH, with FSH added — as human menopausal gonadotropin or recombinant FSH — when hCG alone doesn’t restore spermatogenesis.
This is established practice in reproductive endocrinology, particularly in congenital hypogonadotropic hypogonadism, where it can induce spermatogenesis in men who have never produced sperm. It requires specialist management, it is expensive, and it takes many months. It is the right answer for a specific population rather than a general alternative to testosterone.
When to see a reproductive urologist
Sooner than most men are referred. Specifically:
- Before starting testosterone, if fertility matters to you at all
- If you are on testosterone and now want to conceive
- If you have stopped testosterone and sperm have not returned after six to twelve months
- If you have a history of anabolic steroid use and suppressed gonadotropins
- If semen analysis is abnormal for any reason alongside low testosterone
- If you have low testosterone with low LH and FSH and want children — this is a fertility problem before it is a testosterone problem
If you have genuine deficiency and want children
This is the hardest version of the problem, and it deserves a straight answer rather than reassurance.
A man with unambiguous testosterone deficiency, real symptoms, and a desire for children in the near future faces a conflict that no option resolves perfectly.
The AUA guideline is unambiguous about one branch: testosterone therapy should not be prescribed to men who are currently trying to conceive. That is a strong recommendation, and it reflects the fact that giving a contraceptive to a man attempting conception is difficult to defend.
The realistic options are stimulating his own axis rather than replacing it — clomiphene or enclomiphene, with the evidence limitations described above; hCG monotherapy, which raises testosterone while maintaining intratesticular concentrations, though the monotherapy evidence is weaker than the adjunct evidence; testosterone with hCG co-administration, accepting that this rests on 26 retrospective patients; banking sperm and then treating, which is often the most honest compromise; or deferring treatment and accepting symptoms until the family is complete, which is sometimes the right answer and is almost never offered.
The AUA guideline permits clinicians to use aromatase inhibitors, hCG or selective estrogen receptor modulators in men who want fertility — a permissive statement rather than a directive, which fairly reflects an evidence base too thin to mandate anything.
What should not happen is that a man is started on testosterone, feels better, and discovers eighteen months later that the decision was made for him.
What remains uncertain
How well contraceptive-trial recovery data apply to long-term therapeutic users. The Liu analysis is the best dataset available and it studied healthy young volunteers on defined-duration treatment. Men on testosterone for five or ten years are a different population and are not well characterized.
How many men never recover. Recovery reached 100% by 24 months in the contraceptive data. Clinical experience with long-term users suggests a genuine non-recovery group exists. Its size is not established.
Whether hCG co-administration works as reliably as the mechanism suggests. Twenty-six retrospective patients is not a randomized trial, and one has never been done.
Optimal hCG dosing for fertility preservation. Current practice is extrapolated from physiology studies, not derived from trials with fertility endpoints.
Whether enclomiphene’s short-term sperm preservation persists over years. Sixteen weeks of preserved sperm concentration is encouraging and is not a multi-year fertility outcome.
How much of the testicular volume loss on testosterone fully reverses. Testicular atrophy on testosterone is well described and largely reversible, but we could not verify quantitative figures for the degree of shrinkage or the completeness of recovery, so we are not stating any.
Whether sperm quality — DNA fragmentation, morphology — recovers as completely as count. Count is what gets measured. It is not the whole of fertility.
Questions patients ask
TRT makes you infertile.
It suppresses fertility in most men, usually reversibly — which is not the same as making you infertile, and not as harmless as it sounds either.
What the evidence showsIn WHO contraceptive trials, 65% of men on weekly testosterone became azoospermic, at a mean of around 120 days. Most of the rest were severely suppressed. Recovery occurred in 67% by six months and 90% by twelve.
What remains uncertainThose data come from healthy young volunteers on defined-duration treatment. Recovery after years of clinical therapy is less well characterized.
Bottom lineAssume testosterone will stop your sperm production, and plan accordingly.
Strong
TRT-related infertility is permanent.
For most men it isn't, but the risk of prolonged or incomplete recovery is real and rises with duration of use.
What the evidence showsThe Lancet analysis found 90% of men recovering to at least 20 million sperm/mL by twelve months and 100% by twenty-four. The AUA notes roughly 10% did not recover until the second year.
What remains uncertainWhether a genuine non-recovery group exists among long-term therapeutic users, and how large it is.
Bottom lineProbably reversible. Not guaranteed, and not quickly.
Moderate
You can take TRT and still have children.
Some men do, but not usually while continuing testosterone alone.
What the evidence showsTestosterone alone suppresses spermatogenesis in the majority. Strategies that preserve fertility involve either co-administered hCG, or stimulating the axis with a SERM instead of replacing testosterone.
What remains uncertainHow reliably hCG co-administration preserves fertility — the main study had 26 patients and no control group.
Bottom linePossible with the right strategy, planned in advance. Not by accident.
Limited
hCG prevents infertility on TRT.
It helps considerably. "Prevents" overstates what the evidence supports.
What the evidence showshCG maintains intratesticular testosterone dose-dependently in men whose gonadotropins are suppressed. In 26 men on testosterone plus hCG 500 IU every other day, none became azoospermic and sperm parameters were maintained beyond a year; nine achieved pregnancy.
What remains uncertainNo randomized trial exists. Optimal dosing is extrapolated. Whether it works equally well in all men is unknown.
Bottom lineThe best available fertility-preservation strategy for a man who needs testosterone. Not a guarantee.
Moderate (🟡 Promising)
Clomiphene can replace testosterone.
For some men with secondary hypogonadism, yes — and it preserves fertility, which testosterone doesn't.
What the evidence showsRetrospective series report testosterone roughly doubling on clomiphene with sperm parameters maintained. It only works if the pituitary and testes can still respond, so it is not an option in primary testicular failure.
What remains uncertainNo randomized trials with hard endpoints, and very little data beyond two years.
Bottom lineA legitimate alternative for the right man, on weaker evidence than testosterone itself.
Limited
Enclomiphene is a safer version of TRT.
It works differently and preserves fertility better. "Safer" is a claim nobody has the data to make.
What the evidence showsIn phase II trials of 240 men, enclomiphene raised testosterone with sperm concentration unchanged at 16 weeks and oligospermia under 5%, versus a roughly 60% fall in sperm concentration and over 50% oligospermia on testosterone gel.
What remains uncertainIt is not FDA-approved, having received multiple Complete Response Letters and never completed phase III. No long-term safety data exist. Compounded product is not required to demonstrate the isomer purity the FDA found wanting.
Bottom lineGenuinely promising for fertility preservation, genuinely unapproved, and the supply chain is the weak link.
Limited (🟠)
Sperm production always recovers after TRT.
Usually, and not always quickly.
What the evidence showsMedian time to 20 million/mL was 3.4 months in contraceptive-trial participants, with 67% recovered at six months and 90% at twelve. Shorter treatment duration and higher baseline sperm concentration predicted faster recovery.
What remains uncertainLong-term therapeutic users are underrepresented in this data.
Bottom lineLikely, but plan on months rather than weeks, and don't count on it if you've been on treatment for years.
Moderate
If I stop TRT my fertility comes back in a few weeks.
No. Three months is the physiological floor, and the realistic timeframe is six to twelve.
What the evidence showsSpermatogenesis takes about 74 days, plus two to three weeks of epididymal transit — so roughly three months elapse between the testes restarting and sperm appearing. Median recovery to 20 million/mL was 3.4 months, with 90% recovered by twelve.
What remains uncertainIndividual variation is wide and hard to predict in advance.
Bottom lineStop well before you plan to conceive, not when you start trying.
Moderate
I should freeze sperm before starting TRT.
If you might want biological children, yes — this is close to a free option.
What the evidence showsCryopreserved sperm remains viable for many years. Banking converts a small but real risk of non-recovery into a manageable one, at modest cost.
What remains uncertainNothing much. This is one of the least controversial recommendations on this site.
Bottom lineBank first. The men who regret not doing it substantially outnumber those who regret doing it.
Strong (as a risk-management decision)
hCG can raise testosterone naturally.
It raises your own testosterone production, which is different from "naturally" but a meaningful distinction.
What the evidence showshCG mimics LH and stimulates Leydig cells directly, raising both serum and intratesticular testosterone. Evidence for hCG as monotherapy is limited to small uncontrolled series.
What remains uncertainLong-term efficacy and safety of hCG monotherapy are not established.
Bottom lineReal mechanism, thin evidence as a standalone treatment.
Limited
You can combine TRT and hCG.
Yes, and for men who want to preserve fertility it is the most common approach.
What the evidence showshCG restores intratesticular testosterone dose-dependently alongside exogenous testosterone. In the 26-man series, combination therapy maintained sperm parameters for over a year.
What remains uncertainOptimal dose, whether it works for everyone, and long-term outcomes — none established by trial.
Bottom lineReasonable, mechanistically sound, and resting on much less evidence than its popularity implies.
Moderate
TRT permanently shrinks the testicles.
Testicular volume does decrease, and it largely recovers after stopping.
What the evidence showsSuppressed LH means reduced Leydig cell activity and reduced intratesticular testosterone, and the testis is mostly seminiferous tubule — so when spermatogenesis stops, volume falls. hCG co-administration substantially mitigates this.
What remains uncertainWe were unable to verify quantitative figures for how much volume is lost or how completely it returns, so we are not giving any.
Bottom lineExpect shrinkage. Expect most of it to reverse. Don't expect a precise number from anyone who hasn't measured it.
Moderate on direction; Limited on magnitude
Fertility treatment can't work after years on TRT.
It often can, but it takes longer and works less predictably than in men treated briefly.
What the evidence showsLonger treatment duration consistently predicts slower recovery. Gonadotropin therapy — hCG with FSH added where needed — can restore spermatogenesis in men whose axis has not recovered spontaneously.
What remains uncertainSuccess rates specifically after prolonged testosterone therapy are not well characterized.
Bottom lineNot a lost cause. Get to a reproductive urologist rather than waiting to see what happens.
Limited
Men with low testosterone who want children should never receive TRT.
Not never — but not while trying to conceive, and not without the conversation.
What the evidence showsThe AUA guideline states as a strong recommendation that testosterone should not be prescribed to men currently trying to conceive, and that fertility implications should be discussed with any man interested in future fertility. Alternatives that preserve fertility exist.
What remains uncertainThe best strategy for a man with genuine deficiency and a five-year family timeline is a judgment call, not a protocol.
Bottom lineThe question isn't whether he can ever have testosterone. It's whether anyone asked him about children first.
Strong (guideline position)
My testosterone level tells me whether my fertility is affected.
It tells you nothing at all about it.
What the evidence showsIntratesticular testosterone runs roughly 80 to 100 times higher than serum, and exogenous testosterone reduces it by around 94% while serum levels look excellent. Sperm production depends on the intratesticular concentration, which is not measurable by any routine blood test.
What remains uncertainNothing about the principle.
Bottom lineA good testosterone level on treatment is not evidence that your fertility is intact. Only a semen analysis tells you that.
Strong
Where to go next
If you are deciding whether to start: Should You Start Testosterone?
If you want to understand the alternatives in depth: Emerging and Experimental Therapies
If you used anabolic steroids and your axis hasn’t recovered: Performance Enhancement and Anabolic Steroids
If you’re wondering whether you need testosterone at all: When Testosterone Isn’t the Answer