Testosterone in Midlife

What testosterone actually does

Testosterone is made mostly in the testes, on instruction from the brain. The hypothalamus releases GnRH, the pituitary answers with LH and FSH, and the testes respond by producing testosterone and supporting sperm production. The finished hormone then feeds back on the brain and turns the instruction down. It is a loop, and almost everything interesting about testosterone in midlife is a question about which part of the loop is failing.

What it does downstream is well established. It builds and maintains muscle mass, maintains bone density, drives red blood cell production, influences where the body stores fat, and is central to libido — desire, specifically. It also converts, in fat tissue and elsewhere, into estradiol, which men genuinely need: bone, libido and mood all depend on it. Testosterone is not the opposite of estrogen in men. They work together.

What it does not do is equally worth stating. It is not a general vitality hormone. It is not a treatment for growing older. It is not a reliable treatment for erectile dysfunction — erections depend mainly on blood vessels and nerves, and a man with normal testosterone and erectile difficulty is very unlikely to be helped by more of it. It is not an antidepressant. And in a man whose level is genuinely normal, adding more is a performance-enhancing drug, not a treatment.

Why “low testosterone” is a harder diagnosis than the ads suggest

Start with the symptom list, because it is the weakest link in the whole chain. Fatigue. Low mood. Poor concentration. Weight gain. Poor recovery from training. Reduced libido. Every single one of those belongs to at least five other conditions, several of which are more common than low testosterone and most of which are more treatable. A symptom list that fits everyone identifies no one.

Then the numbers. Testosterone does fall with age, but the fall is modest — a small percentage a year on average, beginning somewhere in the thirties — and most men stay inside the laboratory reference interval well into their seventies. The picture of a cliff edge at fifty, an inversion of menopause, is not what the population data shows. Men who fall a long way usually have a reason beyond age, and finding that reason is more useful than treating the number.

And then the part that unsettles people: the correlation between how a man feels and where his number sits is weak. There are men near the bottom of the range who feel entirely well, and men in the comfortable middle who feel terrible. Symptoms improve in placebo arms of testosterone trials at rates that should make everyone modest. This is why a credible diagnosis needs both halves — a consistent symptom picture and a genuinely low measurement, repeated. Either alone is not enough, and the number alone is the more dangerous of the two, because it looks objective.

How it is properly measured

In the morning. Testosterone follows a daily rhythm, highest in the early hours after waking and drifting down through the day, and the reference ranges every lab reports against were built on early-morning samples. An afternoon draw compared against a morning range is how healthy men get told they are deficient.

Fasting, and more than once. Eating lowers testosterone acutely. So does a bad night, a heavy training block, an infection, a course of steroids for a chest complaint, a week of heavy drinking. Day-to-day variation in the same man is wide enough that a single low result means very little. Two separate mornings is the minimum before anyone should be having a conversation about treatment.

Total is not the whole story. Most testosterone in the blood is not available to your tissues: a large share is bound tightly to sex hormone binding globulin, more is bound loosely to albumin, and only a small free fraction, plus what comes off albumin, is doing anything. SHBG is not fixed. It rises with age, with an overactive thyroid, with liver disease and with some medications; it falls with obesity, insulin resistance and an underactive thyroid. So a respectable total testosterone sitting on a high SHBG can hide a genuinely low free level, and a modest total on a low SHBG can be entirely adequate. The usual approach is to measure total testosterone alongside SHBG and albumin and calculate the free level from them — direct free testosterone immunoassays are unreliable enough that most guidelines discourage them. There is a fuller explanation in What SHBG Is, and Why It Changes Your Testosterone Result.

A proper workup rarely stops at testosterone. LH and FSH separate the two causes described below; prolactin, thyroid function, iron studies, a metabolic panel and hematocrit all either change the reading or catch something else entirely. That is what a baseline panel is for.

Primary or secondary — and why it changes everything

LH and FSH answer one question: is the brain asking, and are the testes able to respond?

Primary means the testes are the problem. The brain is shouting — LH and FSH come back high — and the output is still low. Causes include prior mumps orchitis, testicular injury or surgery, undescended testes in childhood, chemotherapy or radiation, and genetic conditions such as Klinefelter syndrome. It is usually permanent.

Secondary means the instruction is the problem. Testosterone is low and LH and FSH are low or unremarkable, which is the wrong answer — a healthy pituitary faced with a low testosterone should be shouting. The common causes are the ones nobody wants listed: obesity, obstructive sleep apnea, opioid medication, glucocorticoids, heavy alcohol use, chronic illness, severe energy deficit from training and under-eating, and a history of anabolic steroid use. Occasionally it is a pituitary problem, which is why prolactin gets checked.

The distinction matters for two reasons. The first is that secondary is often reversible, and treating the cause frequently raises testosterone without any hormone at all. The second is mechanical: the medications that stimulate your own production only work if the testes can respond. In primary hypogonadism there is nothing to stimulate, and replacement is the only route that does anything.

Replacing it, or stimulating it

Replacement supplies testosterone from outside — injection, gel, pellet or an oral formulation. It reliably raises the level, which is its great advantage. The cost is the loop. Your body reads the hormone arriving, concludes it has enough, and shuts its own production down: LH and FSH fall, the testosterone concentration inside the testes falls much further than the blood level suggests, and sperm production is suppressed — in many men to the point of infertility. The testes usually shrink. This is not a rare side effect; it is how the drug works.

Suppression is often reversible after stopping, but recovery takes months rather than weeks and is not guaranteed, and the odds get worse with longer use and older age. Anyone who might want children, or is not certain he will not, should have that conversation and a semen analysis before starting rather than after. A male fertility test is the honest starting point there, and Testosterone Therapy and Fertility covers what is known about recovery.

Stimulation works the other way. Enclomiphene and clomiphene block estrogen feedback at the hypothalamus and pituitary, so the brain reads the situation as lower than it is and asks louder. LH and FSH rise, and the testes make more of their own testosterone — which means testicular function and sperm production stay in play rather than being switched off. The catch is the one above: this only works in secondary hypogonadism, where the testes are capable. Clomiphene is a mixture of two isomers, one of which lingers in the body for a long time and carries much of the side-effect burden; enclomiphene is the isolated isomer without it. Both are used off-label in men, and enclomiphene is not available as an FDA-approved product — it is prescribed as a compounded preparation.

Neither is simply better. Replacement is more certain and more suppressive. Stimulation preserves the axis and does less, less predictably, in a narrower group of men. Testosterone Therapy or Enclomiphene: How They Differ sets the two side by side, and the clinical decision — including which form, if replacement is right — belongs with a clinician who has your labs, on the men’s hormone page.

What the evidence actually supports

This is the section most pages in the category leave out, so it is worth being specific. Everything below applies to men with confirmed low testosterone. None of it applies to men with normal levels who feel below par.

Sexual function. The most consistent finding in the literature. Libido and sexual activity improve, modestly and fairly reliably. Erectile function improves less, and testosterone is not a treatment for erectile dysfunction in its own right.

Energy and mood. Inconsistent. The Testosterone Trials, the best-designed set of studies in older men, found a small improvement in mood and depressive symptoms and no meaningful effect on vitality — the thing most men come in asking for. Some men report a clear change. The average effect across a trial population is small.

Body composition. Lean mass goes up and fat mass goes down, consistently. Whether that converts into strength or function you would notice is much less clear, and it does not happen without training.

Bone and blood. Bone density improves; whether that prevents fractures has never been tested. Anemia improves in men who were anemic. Cognition does not improve.

Cardiovascular safety. For a decade this was genuinely unresolved — some observational work suggested harm, some suggested benefit, one trial was stopped early for cardiovascular events, and regulators required a warning. TRAVERSE, reported in 2023, randomized more than five thousand middle-aged and older men with low testosterone and existing cardiovascular risk to testosterone gel or placebo, and found no excess of major adverse cardiac events. That is a real reassurance and the best evidence available. It is not a clean bill of health: the same trial found more atrial fibrillation, more pulmonary embolism and more acute kidney injury in the treated group, and it says nothing about men outside that population or about decades of use.

What gets monitored, and why

Testosterone therapy is a monitored treatment, not a prescription you collect and forget. Three things get watched for specific reasons.

Hematocrit. Testosterone stimulates red blood cell production, and in some men it overshoots. A rising hematocrit thickens the blood, and it is the single most common reason therapy gets reduced or paused. It is checked before starting and at intervals afterwards — see Hematocrit: Why It Gets Watched on Hormone Therapy.

Prostate. The old belief that testosterone causes prostate cancer has not held up. The residual concern is that it can accelerate a cancer already present, so a baseline PSA and a conversation about prostate risk come before treatment, with a recheck in the first year and periodically after.

Estradiol. Testosterone aromatizes into estradiol, and raising one raises the other. Men need estradiol, so the goal is not to crush it. When it runs high, breast tenderness, fluid retention and mood changes can follow, and the reflex to add an aromatase inhibitor is usually the wrong first move — over-suppressed estradiol causes bone loss, joint pain and the loss of the libido the treatment was meant to help. High Estradiol in Men: Why It Shows Up explains the picture.

Alongside those: sleep apnea can worsen, so anyone with symptoms should be assessed before starting; blood pressure and lipids are tracked; and there should be a defined point at which everyone agrees whether anything has changed. If nothing has after a fair trial, the honest answer is to stop rather than to keep adjusting.

The things that look like low testosterone

Every item on this list produces the symptom picture at the top of the page, and several of them also genuinely lower testosterone — which means treating them can raise it without a prescription.

Obstructive sleep apnea is the one most often missed. It fragments sleep, blunts the overnight testosterone rise, and produces exactly the fatigue, irritability and low libido that get attributed to hormones. It is common in men over forty-five who have put on weight, and it is treatable. See what we look at with sleep.

Excess weight, particularly visceral fat. Fat tissue converts testosterone into estradiol, which increases the feedback signal telling the brain to stop asking. It is a genuine loop, and weight loss raises testosterone more dependably than most interventions aimed at the hormone directly.

Alcohol suppresses the axis at both ends, and the amount required is lower than most men assume. Depression shares almost the entire symptom list and is treated differently. Thyroid disease produces the same fatigue and weight change and also moves SHBG, which distorts the testosterone result itself. Iron deficiency and the other common causes of fatigue belong on the same list.

So do medications. Opioids are strongly suppressive. So are glucocorticoids. Some antidepressants, finasteride and the gabapentinoids all affect libido or function independently of any hormone level. A medication review is free, quick, and skipped constantly.

If you are trying to decide

A reasonable path through this looks the same for most men. Measure properly — morning, fasting, twice, with SHBG and the pituitary hormones rather than a single total testosterone. Look hard at the reversible causes first, because they are common and the answers there are better. Only then decide between replacement and stimulation, with fertility, the monitoring burden and the honest evidence all visible on the table.

For a meaningful number of men the conclusion is not yet, or not this. That is not a failed consultation. A man who treats his sleep apnea, loses the visceral weight and finds his energy back has solved the actual problem, and he has done it without committing to a therapy that suppresses his own production. Testosterone is a good treatment for the men who need it, and that is a smaller group than the advertising implies.

Frequently asked questions