Hormones after 40: testosterone, estradiol, SHBG, DHEA, thyroid — and the honest version of TRT

What declines, and how much

Testosterone: what the cohorts say

The European Male Ageing Study (Wu et al. 2010, NEJM; 3,369 men aged 40–79) set out to define "late-onset hypogonadism" properly. It found that only three symptoms tracked with low testosterone — fewer morning erections, low libido and erectile difficulty — and only once total testosterone was below about 11 nmol/L and free testosterone below 220 pmol/L. By that definition just 2.1% of the men had it. Fatigue, low mood and poor concentration, the symptoms most often attributed to testosterone, were associated far more strongly with obesity, poor sleep and illness.

That matters because the most common cause of a low testosterone result in a man of fifty is not his testes — it is visceral fat, insulin resistance and untreated sleep apnoea, all of which suppress the axis and all of which reverse. Losing 10% of body weight raises total testosterone by roughly 2–3 nmol/L in trials; treating apnoea does something similar. A protocol that reaches for replacement before those are addressed is treating the thermometer.

TRT: the honest version

For men with symptoms and confirmed low levels on two morning samples, testosterone replacement does what it says: more lean mass, less fat, better bone, better libido and mood, and in the T Trials (Snyder 2016, NEJM; 790 men over 65) modest improvements in sexual function and walking distance. The safety question was answered by TRAVERSE (Lincoff 2023, NEJM): 5,246 men aged 45–80 with low testosterone and high cardiovascular risk, randomised to testosterone gel or placebo for a mean of 22 months. Testosterone was non-inferior for major cardiovascular events — the fear of a decade was not borne out — but the treated group had more atrial fibrillation, more pulmonary embolism and more acute kidney injury. Not dangerous; not free.

The three things every man considering it should hear before the first injection:

• Haematocrit. Testosterone raises red-cell production. Above 0.50 the dose or the frequency changes; above 0.54 treatment stops until it falls. Injections produce bigger rises than gels; smaller, more frequent doses produce smaller rises. It is checked at three months and then every six.
• Fertility. Exogenous testosterone switches off the signal to the testes, and sperm production falls to near zero in most men within months. It usually recovers after stopping, over six to eighteen months, but not always. Any man who may want children needs that conversation first; there are alternatives a physician can offer that preserve it.
• Estradiol. Some of the testosterone converts to estradiol, and that is a feature: it is what protects bone and much of the libido and mood benefit. Adding an aromatase inhibitor to "control E2" because of a number is one of the most common and most damaging mistakes in men's health — it produces joint pain, low mood, bone loss and a worse lipid profile. Symptoms, not the number, decide.

Also on the list: testicular shrinkage, acne, sleep apnoea getting worse, and the fact that it is a commitment — the axis does not restart easily after years of suppression. None of this is a reason not to do it. All of it is a reason to do it with a physician who measures.

Women: the perimenopause basics

Perimenopause begins on average in the mid-forties and lasts four to eight years. It is not a slow fade; it is a period of erratic, often high estradiol swinging against falling progesterone, which is why the early symptoms — sleep that breaks at 3 a.m., anxiety, heavier or unpredictable cycles, migraines — are so often misattributed. Hot flushes come later. A single FSH or estradiol measurement in this phase is close to meaningless because the values change week to week; the diagnosis is clinical, by age and symptoms.

On hormone therapy the evidence has settled considerably since the 2002 Women's Health Initiative headlines. The WHI population averaged 63 years old and more than a decade past menopause; the re-analyses and later trials support the timing hypothesis: for women who start within ten years of menopause or before sixty, menopausal hormone therapy relieves symptoms, protects bone and does not raise cardiovascular risk, and the breast-cancer signal with modern transdermal estradiol and micronised progesterone is small. For women with vasomotor symptoms it is the most effective treatment there is. Testosterone, at a tenth of the male dose, has evidence for libido in postmenopausal women and is increasingly prescribed. All of this is a specialist conversation; the point of this paragraph is that it is a conversation worth having rather than enduring.

What to measure

• Men: total testosterone, SHBG and calculated free testosterone, estradiol (by a sensitive assay), LH and FSH, prolactin if testosterone is low, haematocrit, PSA from 45. Drawn between 7 and 10 a.m., fasting, rested, and repeated once before any conclusion.
• Women: FSH and estradiol are useful before 45 and after menopause, unreliable in between; ferritin and thyroid matter more in perimenopause than hormones do.
• Both: TSH and free T4; DHEA-S once as a baseline; IGF-1 if a GH-axis protocol is in view; HbA1c and fasting insulin, because insulin resistance suppresses the axis in both sexes.
• Sleep and apnoea screen before any hormone is interpreted. A night of five hours cuts morning testosterone by 10–15%.
• DEXA for bone density: the thing both testosterone and estradiol protect.

Where peptides and therapies fit

The longevity goal page describes the protocol we write for people over forty whose complaint is energy, sleep and body composition rather than a single hormone, and it starts with this panel. The compounds in the research library that touch the hormonal axis are the growth-hormone secretagogues — CJC-1295 / ipamorelin and tesamorelin — which are studied for raising the body's own GH pulse rather than replacing it, and are gated on IGF-1 and HbA1c. Testosterone and menopausal hormone therapy are prescription medicines and a physician's decision; what a protocol can do is make sure the labs, the sleep and the body composition are in place so that decision is made on real numbers. The baseline-labs guide lists the panel.

Common mistakes

• One afternoon testosterone result. It is a morning, fasting, rested test, done twice.
• Treating total testosterone and ignoring SHBG. A total of 14 with an SHBG of 70 is a free testosterone problem.
• Replacing before fixing sleep, weight and apnoea. The commonest cause of low T at fifty is reversible.
• Crushing estradiol with an aromatase inhibitor because a number looked high.
• Starting TRT without a fertility conversation.
• Treating a mildly raised TSH in someone over 70. It is usually normal ageing and the trial showed no benefit.
• For women: a single FSH in perimenopause used to say "it is not your hormones".