Low Testosterone in Men: What the Evidence Actually Supports

Dernière mise à jour : août 29, 2026

• WordPress
• LinkedIn
• Courrier

Few areas of men’s health attract as much confusion — and as much aggressive marketing — as low testosterone, or male hypogonadism. My training in this area is an Andrology Fellowship at Chang Gung Memorial Hospital Kaohsiung, Taiwan (2019), under Professor Po-Hui Chiang, and male hormonal health has been a large part of my clinic ever since. I am currently a Research Scholar and Clinical Observer at the Scott Department of Urology, Baylor College of Medicine — a research and observer appointment rather than a clinical fellowship, carrying no licence to practise in the United States. My practice is in Thailand.

In 2024 The Lancet Diabetes & Endocrinology published a review of the pathogenesis, diagnosis and management of male hypogonadism, drawing together the large clinical trials of the previous decade including TRAVERSE. Its conclusions are considerably more modest than the advertising, and considerably more useful. This article sets out what they mean for a man who suspects his testosterone is low.

Before Anything Else: Symptoms That Need Urgent Assessment

Most low testosterone is investigated calmly over weeks. Two situations are not.

If you are already taking testosterone, go to an emergency department the same day for chest pain or chest tightness, breathlessness at a level of exertion that never used to trouble you, pain or swelling in one calf, sudden sharp chest pain that is worse on breathing in, a rapid or irregular heartbeat, or any stroke symptom — face drooping, arm weakness, difficulty speaking — even if it settles within minutes. In Thailand, call 1669. These are ordinary emergency symptoms that anyone should act on; listing them here is prudence, not a claim that testosterone caused them.

If you have not started treatment, low testosterone accompanied by new persistent headaches, a change in vision — particularly loss of the outer edges of your visual field — or milky discharge from the nipples is not the ordinary picture. Those features point towards the pituitary gland rather than the testes, and they need investigating rather than treating with testosterone. Loss of body hair with shrinking testicles is also worth prompt assessment.

The Two Faces of Low Testosterone: Organic and Functional

Not all testosterone deficiencies are the same problem, and the distinction decides whether the answer is hormonal or metabolic.

1. Organic (classical) hypogonadism — an intrinsic and usually irreversible problem somewhere in the hypothalamic–pituitary–testicular axis. Klinefelter syndrome, previous testicular surgery, chemotherapy or radiotherapy, and pituitary disease all belong here. Testosterone replacement is the appropriate treatment.
2. Functional (late-onset) hypogonadism — testosterone suppressed by factors outside the reproductive axis, chiefly obesity, type 2 diabetes and accumulated illness, with no permanent structural damage. This is the form driving the global surge in testosterone prescriptions.

The crucial difference is that functional hypogonadism is frequently reversible when the underlying factors are corrected. That is why weight and metabolic health come before the prescription pad, and why the Lancet review names health optimisation as the gold-standard management strategy rather than a preliminary courtesy. Losing at least 5% to 10% of body weight raises testosterone measurably — on pooled data, by around 70 ng/dL with dietary weight loss and around 200 ng/dL after bariatric surgery, in proportion to the weight lost. The causal case is discussed in more detail in why your doctor recommends the gym before the pharmacy.

Quels sont les vrais symptômes ?

Many men arrive at the clinic attributing general fatigue or low mood to low testosterone. Those can be part of it, but the evidence is clear that sexual symptoms are the most specific indicators.

• The most specific: a definite fall in libido, loss of spontaneous morning erections, and erectile dysfunction.
• Supporting but non-specific: reduced muscle mass, increased central fat, unexplained anaemia, low mood and chronic fatigue. Each of these is caused far more often by something else — sleep apnoea, diabetes, thyroid disease, depression, poor sleep — than by testosterone.

This matters practically. A man treated with testosterone for fatigue that was actually caused by untreated sleep apnoea has been given a drug that can make sleep apnoea worse, while the real problem goes on unaddressed.

How the Test Should Be Done

One of the commonest problems I see is a diagnosis built on a badly taken sample.

• Timing and fasting. Blood should be drawn in the morning, ideally between 7:00 and 10:00, because testosterone peaks after sleep. Fasting matters too, as concentrations fall after a meal.
• Two samples, not one. Day-to-day variation is substantial, so a single low reading should always be confirmed on a separate day before anyone talks about treatment.
• Not during acute illness. Infection, injury, surgery and severe stress all suppress testosterone temporarily. A level taken in hospital or in the week after an illness can be low in a man whose axis is entirely normal.
• Total, and sometimes free. Total serum testosterone is the principal test. In older men and in obesity, the binding protein SHBG shifts, which can make the total figure misleading in either direction; a calculated free testosterone helps in those situations.
• LH and FSH. These separate a testicular problem from a pituitary one, and they change the plan. They are also what determines whether an alternative such as hCG could work at all — it does nothing when the testes themselves have failed, as explained in the article on hCG for low testosterone.

What Treatment Delivers, and What It Does Not

For functional, obesity-related hypogonadism, weight loss and treatment of the underlying conditions remain the best management. When that is not enough, or when the deficiency is organic, testosterone therapy is a reasonable option — but it is worth being precise about the size of the benefit.

The pooled trial evidence describes improvements in sexual function that are statistically significant and modest. Desire responds fairly dependably. Erections respond far less reliably, because most erectile dysfunction is a blood-vessel problem rather than a hormonal one, and normalising a hormone does not repair an artery. Energy and mood improve for some men and not for others. Testosterone is not a tonic for ageing, and higher levels are not better than adequate ones.

An earlier version of this article described testosterone therapy as a highly effective treatment for sexual symptoms. That was stronger than the evidence it cited, and it has been corrected.

What Testosterone Therapy Asks of You in Return

This is the section that was missing from the original version of this article, and it is the part of the conversation that should happen before the first dose rather than after it.

• Fertility. Testosterone given from outside switches off the body’s own signal to the testes, and sperm production falls with it — frequently to zero. For a man who may want children, this is the single most important fact about the treatment. Recovery usually takes months and sometimes a couple of years, and it is not guaranteed. If fertility matters to you, say so before starting: a semen analysis beforehand is sensible, and there are alternatives that raise testosterone while preserving sperm production, including clomiphene citrate, hCG and an aromatase inhibitor.
• Erythrocytosis. Testosterone thickens the blood by raising the red cell count. This was by a wide margin the clearest adverse effect in the TRAVERSE trial — a difference of a completely different order from the signals discussed below. Haematocrit is the monitoring result that most often forces a dose reduction, a change of preparation or a pause. It is checked before starting and regularly for as long as you remain on treatment.
• PSA. Prostate assessment and a PSA before starting, then monitoring, are standard — not because testosterone is thought to cause prostate cancer, but because treatment should not be started blind in a man over 40.
• Sleep apnoea. Testosterone can worsen untreated obstructive sleep apnoea, which is common in exactly the men who present with low testosterone. It should be looked for first, not afterwards.
• Transfer from gels. Testosterone gel transfers by skin contact and can cause early virilisation in a child or unwanted effects in a partner. Cover the site, wash your hands, and keep the area away from others until it is absorbed.
• Commitment. For organic hypogonadism this is usually lifelong. Stopping abruptly, having suppressed your own production, generally feels worse than never starting.

Safety: What TRAVERSE Actually Found

TRAVERSE was the large cardiovascular safety trial the field had waited decades for. It randomised 5,246 men aged 45 to 80 with confirmed low testosterone who already had cardiovascular disease or carried multiple risk factors, and followed them for a mean of 33 months. It is usually quoted in half — and, as this page previously demonstrated, it is easy to quote the wrong half in either direction.

The main result

Testosterone did not increase heart attack, stroke or cardiovascular death. It met the criterion for non-inferiority to placebo on that combined endpoint. This is the finding that is quoted everywhere, and it is genuinely reassuring — particularly given the trial deliberately enrolled the men most likely to be harmed if testosterone were harmful.

What reached statistical significance

• Erythrocytosis — a rise in red cell count, several times more frequent on testosterone than on placebo. This was the clearest adverse effect in the trial, and it is why haematocrit is monitored for as long as treatment continues.
• Fractures — the fracture subtrial found a statistically significant increase of roughly 43% in clinical fractures on testosterone, around 3.5% of men versus 2.5% on placebo, with the curves separating within the first few months. Ribs, wrist and ankle were the commonest sites. Contrary to the older belief that testosterone protects bone, it should not be prescribed to prevent fractures.

What was numerically higher but not statistically significant

Atrial fibrillation, pulmonary embolism and acute kidney injury each occurred somewhat more often in the testosterone group. None of those differences reached statistical significance.

An earlier version of this page listed these three as findings the trial recorded, without that qualification. That reads as though they were established harms, and they are not. Describing a difference the trial could not distinguish from chance as a demonstrated risk is a real error — and it is the same error, pointed the other way, as calling a significant 43% rise in fractures merely slight.

That is not a reason to dismiss them. These are serious events, the numbers moved in the same direction, and a trial this size failing to reach significance is not proof of no effect. It is reasonable to mention a previous blood clot, a clotting disorder or a known arrhythmia before starting, so it can be weighed. It is not reasonable to refuse treatment to a man who needs it on the strength of them.

What the trial looked for and did not find

• Diabetes. Smaller earlier studies suggested testosterone might stop prediabetes progressing. The prespecified TRAVERSE diabetes substudy found no significant difference in progression, and no difference in glucose or HbA1c. The American Diabetes Association does not recommend testosterone for diabetes prevention. It is not a treatment for diabetes and not a substitute for weight loss.
• Prostate. No increase in prostate cancer was seen. The important qualification is that TRAVERSE, like the earlier T-Trials, excluded men with a history of prostate cancer — so it does not answer the question that men already treated for prostate cancer most want answered.

That last gap is one our group has contributed to directly. A single-centre cohort of 46 hypogonadal men aged 80 and over starting testosterone therapy, published in Journal de médecine sexuelle in 2026, found that 39% had a history of prostate cancer and that no biopsy-confirmed recurrence occurred during follow-up. The limits deserve as much weight as the finding: the median follow-up was around 21 months rather than the full three years, only about a third of the men were followed for the whole 36 months, and there was no untreated comparison group, so the study cannot measure relative risk. It is a contribution to the evidence rather than an answer to it — and the shift in thinking it belongs to is set out in testosterone and prostate cancer: the saturation model and in the honest balance of benefits and risks.

A safety trial that finds no increase in the feared harm has shown the absence of that harm. It has not shown a benefit. Those are different results and they are often reported as the same one. The fuller account of the cardiovascular question is in testosterone and the heart.

What Is Available Here

At Bangkok Hospital Headquarters I provide testosterone as gels and injections. Pellets are common in the United States and are almost never used in Thailand; oral testosterone products vary by country and are not part of my practice. Where preserving fertility is the priority, clomiphene citrate, hCG or an aromatase inhibitor may be more appropriate than testosterone itself. Practical technique for the injectable route is covered in the guide to subcutaneous self-injection, and the wider metabolic picture in testosterone, weight loss and men’s metabolic health.

Testosterone bought online or from a gym is a different proposition altogether. The contents are not guaranteed, the dose is often far above replacement, and nobody is monitoring your haematocrit.

Le point essentiel

If you are over 40 with a genuine fall in sexual desire, loss of morning erections and fatigue, that is worth investigating properly rather than self-treating. The investigation is a correctly taken morning test repeated on a second day, with LH, FSH and a look for the conditions that suppress testosterone in the first place. Distinguishing organic from functional hypogonadism is what determines whether the answer is hormonal or metabolic, and that distinction is the starting point of the évaluation et traitement de la baisse de testostérone pathway. A fuller discussion of fertility, prostate and heart questions is in thérapie à la testostérone aujourd'hui.

Appointments. Consultations are arranged through the hospitals: Hôpital de Bangkok Siège social, or Samitivej Sriracha Hospital in Chonburi on 088-022-1445.

La télémédecine de Bangkok Hospital est disponible pour les patients qui ne peuvent pas se déplacer en personne, y compris les patients internationaux — organisez-la à l'avance par e-mail auprès du service d'urologie à bhquro@bdms.co.th. Samitivej Sriracha is in-person only. Enquiries about cost are answered by the hospital rather than through this website; please email the same address.

Foire aux questions sur la testostérone basse

Les symptômes les plus courants d'un faible taux de testostérone sont : * Baisse de la libido (désir sexuel) * Difficultés d'érection * Fatigue et manque d'énergie * Perte de masse musculaire * Augmentation de la masse graisseuse * Baisse de la densité osseuse * Changements d'humeur (irritabilité, dépression) * Perte de pilosité corporelle * Diminution de la production de spermatozoïdes (pouvant affecter la fertilité)

The most specific symptoms are sexual — a definite fall in libido, loss of spontaneous morning erections, and erectile dysfunction. Fatigue, low mood, reduced muscle mass and increased central fat can accompany it, but each of those is caused far more often by sleep apnoea, diabetes, thyroid disease or depression than by testosterone, so they are not enough on their own to make the diagnosis.

Quelle est la différence entre l'hypogonadisme organique et fonctionnel ?

Organic hypogonadism involves permanent damage somewhere in the hormonal axis and usually needs testosterone replacement. Functional hypogonadism is testosterone suppressed by outside factors such as obesity or type 2 diabetes, with no structural damage, and it is frequently reversible when those factors are corrected.

How should testosterone be tested correctly?

Blood is drawn in the morning between 7:00 and 10:00 while fasting, and a single low reading is confirmed on a separate day. It should not be measured during an acute illness or shortly after one, because that suppresses the level temporarily. LH and FSH are measured alongside it to separate a testicular cause from a pituitary one.

Does testosterone therapy affect fertility?

Yes, and substantially. Testosterone given from outside suppresses the body’s own signal to the testes and sperm production falls with it, frequently to zero. Recovery after stopping usually takes months and sometimes a couple of years, and it is not guaranteed. If you may want children, raise it before the first dose — a semen analysis beforehand is sensible, and clomiphene citrate, hCG or an aromatase inhibitor can raise testosterone while preserving sperm production.

Which adverse findings in TRAVERSE were statistically significant?

Two. Erythrocytosis — a rise in red cell count — was by far the clearest, which is why haematocrit is monitored throughout treatment. Clinical fractures also rose significantly, by roughly 43%, in the fracture subtrial. Atrial fibrillation, pulmonary embolism and acute kidney injury were each numerically higher on testosterone but did not reach statistical significance, and should not be described as established harms. An earlier version of this page listed those three without that qualification.

Is testosterone therapy safe for the heart and the prostate?

For the heart, the TRAVERSE trial found no increase in heart attack, stroke or cardiovascular death, meeting the criterion for non-inferiority. For the prostate, no increase in prostate cancer was seen — but the trial excluded men with a previous prostate cancer, so it does not answer that question. A safety trial showing the absence of a feared harm is not the same as a trial showing benefit.

Un taux de testostérone bas peut-il être inversé sans médicaments ?

In functional hypogonadism, often yes. Losing at least 5% to 10% of body weight raises testosterone — on pooled data, by around 70 ng/dL with dietary weight loss and around 200 ng/dL after bariatric surgery, in proportion to the weight lost. Treating underlying conditions such as type 2 diabetes and obstructive sleep apnoea helps too, and this remains the best management strategy for that group.

Which forms of testosterone are used in your practice?

Gels and injections. Pellets are common in the United States and almost never used in Thailand, and oral testosterone products vary by country; neither is part of my practice. Where fertility is the priority, clomiphene citrate, hCG or an aromatase inhibitor may be the better route.

When should I seek urgent medical attention?

If you are on testosterone therapy: chest pain, breathlessness on less exertion than usual, pain or swelling in one calf, sharp chest pain worse on breathing in, an irregular or racing heartbeat, or any stroke symptom even if it resolves — same-day emergency department, and 1669 in Thailand. If you have not started treatment: low testosterone with new headaches, loss of peripheral vision, or milky nipple discharge points to the pituitary and needs assessment rather than a testosterone prescription.

Références

• De Silva NL, Papanikolaou N, Grossmann M, Antonio L, Quinton R, Anawalt BD, Jayasena CN. Male hypogonadism: pathogenesis, diagnosis, and management. Lancet Diabetes Endocrinol. 2024;12(10):761–774.
• Lincoff AM, Bhasin S, Flevaris P, et al. Cardiovascular safety of testosterone-replacement therapy (TRAVERSE). N Engl J Med. 2023;389(2):107–117.
• Snyder PJ, Bauer DC, Ellenberg SS, et al. Testosterone treatment and fractures in men with hypogonadism. N Engl J Med. 2024;390(3):203–211.
• Bhasin S, Lincoff AM, Nissen SE, et al. Effect of testosterone on progression from prediabetes to diabetes in men with hypogonadism: a substudy of the TRAVERSE randomized clinical trial. JAMA Intern Med. 2024;184(4):353–362.
• Coady PJ, Walia A, Oppenheimer A, Hernandez BS, Chidananda R, Hinojosa-Gonzalez DE, Weerasopone S, Kohn T, Khera M. Testosterone replacement therapy in male octogenarians: clinical outcomes and adverse event rates. J Sex Med. 2026;23(9):qdag258.

Avis de non-responsabilité : This content is written and reviewed by Dr. Soarawee Weerasopone, a board-certified urologist at Bangkok Hospital Headquarters. It is intended for general education only and does not constitute medical advice, diagnosis or treatment for any individual. No advice, diagnosis or prescription is given through personal messaging channels or social media, and Dr. Soarawee operates no public social media account. Always consult a qualified doctor about your own symptoms. In an emergency in Thailand, call 1669.