HRT & TRT: Protocols, evidence reversal, and the DACH reality

Doctors backed away from HRT because one trial scared them. In 2002 the Women's Health Initiative (WHI) reported higher risk of breast cancer (HR 1.26; a hazard ratio of 1.26 means 26% higher risk), heart disease (HR 1.29) and stroke (HR 1.41) ^[1]. The press turned that into a blanket warning. The catch: WHI enrolled women at a mean age of 63, a median 12 years past menopause ^[1]. The headline got applied to someone else, the symptomatic 51-year-old.

In 1999, US doctors wrote about 90 million prescriptions a year for menopausal hormones ^[3]. The best-known product was Premarin (conjugated equine estrogens, CEE). By 2010, menopausal hormone therapy (MHT) had flipped into one of the most under-prescribed effective treatments in mainstream medicine. How did that happen? One trial, one press conference, and fifteen years of careful re-analysis.

Start with the trial that did it: the Women's Health Initiative (WHI), published in JAMA in 2002. It enrolled 16,608 postmenopausal women aged 50 to 79, mean age at randomisation 63, and put them on CEE 0.625 mg/d plus medroxyprogesterone acetate (MPA) 2.5 mg/d versus placebo. It was stopped early at a mean of 5.2 years of follow-up ^[1].

Here is what the combined-therapy arm showed. Some risks went up, some went down.

• Invasive breast cancer: 26% higher (HR 1.26, 95% CI 1.00 to 1.59), about 8 more cases per 10,000 person-years
• Coronary heart disease (CHD): 29% higher (HR 1.29, 95% CI 1.02 to 1.63), about 7 more events per 10,000 person-years
• Stroke: 41% higher (HR 1.41, 95% CI 1.07 to 1.85), about 8 more strokes per 10,000 person-years
• Pulmonary embolism (PE): roughly double (HR 2.13, 95% CI 1.39 to 3.25), about 8 more PEs per 10,000 person-years
• Hip fracture: 34% lower (HR 0.66, 95% CI 0.45 to 0.98), about 5 fewer fractures per 10,000 person-years
• Colorectal cancer: 37% lower (HR 0.63, 95% CI 0.43 to 0.92), about 6 fewer cancers per 10,000 person-years

The trial concluded CEE+MPA "should not be initiated or continued for the primary prevention of CHD." ^[1]

Here is the caveat that did not survive the press cycle. The mean age at randomisation was 63. Women started HRT a median of 12 years after menopause. That is not the patient most people picture when they think about HRT. It is not the symptomatic 51-year-old considering estradiol for hot flashes. It is a fundamentally different physiology.

Now the part almost nobody heard. WHI also ran a second arm, published in JAMA in 2004, in 10,739 women who had already had a hysterectomy. They took CEE alone (no progestogen) versus placebo ^[2]. The numbers looked very different:

• CHD: HR 0.91 (95% CI 0.75 to 1.12), neutral
• Breast cancer: HR 0.77 (95% CI 0.59 to 1.01), trending downward
• Stroke: HR 1.39 (95% CI 1.10 to 1.77), higher
• Hip fracture: HR 0.61 (95% CI 0.41 to 0.91), lower
• Global index: HR 1.01 (95% CI 0.91 to 1.12), neutral overall

Read that breast cancer number again. 0.77, trending down. Estrogen alone, in women without a uterus, did not raise breast cancer risk. This is one of the most under-discussed findings in 20th-century women's medicine. The 2002 press cycle lumped CEE+MPA together with "HRT" as a category. They were not the same drug.

The prescribing collapse got documented too (JAMA, 2004) ^[3]. Prempro (CEE+MPA) fell 66% from Jan-Jun 2002 to Jan-Jun 2003. Premarin (CEE alone) fell 33% in the same window. Total US HRT prescriptions dropped from around 90 million (1999) to a projected 57 million (2003) ^[3]. The authors read this as practice reacting fast to the new evidence of harm ^[3]. That reaction was not sorted by age or symptoms. Symptomatic 51-year-olds were pulled off HRT alongside asymptomatic 70-year-olds.

That collapse was a mistake. Not the trial. The interpretation.

The long-term follow-up made that clear. At 13-year cumulative follow-up (JAMA, 2013) ^[4], the CEE-alone arm showed a cumulative breast cancer HR of 0.79 (95% CI 0.65 to 0.97), significantly reduced. The CEE+MPA arm showed a cumulative breast cancer HR of 1.28 (95% CI 1.11 to 1.48). Women aged 50 to 59 on CEE alone showed a directionally favourable all-cause mortality signal.

Then came the capstone (JAMA, 2017): 18-year cumulative follow-up, median 17.7 years, n = 27,347 pooled ^[5]. The plain takeaway is that hormone therapy did not change whether women lived or died.

• All-cause mortality (pooled): HR 0.99 (95% CI 0.94 to 1.03). HRT arm 27.1% vs placebo 27.6%
• Cardiovascular mortality (pooled): HR 1.00 (95% CI 0.92 to 1.08)
• Cancer mortality (pooled): HR 1.03 (95% CI 0.95 to 1.12)
• Age interaction (younger 50-59 vs older 70-79 starters, during the intervention): ratio of HRs 0.61 (95% CI 0.43 to 0.87)

The trial put it plainly: "hormone therapy with CEE plus MPA for a median of 5.6 years or with CEE alone for a median of 7.2 years was not associated with risk of all-cause, cardiovascular, or cancer mortality during a cumulative follow-up of 18 years." ^[5]

This is the all-cause mortality capstone. Hormone therapy, in either arm, over 18 years, did not raise total mortality. The 2002 narrative did not survive the long-term follow-up.

The fair reading is not "WHI was wrong." WHI was correct for the patients it enrolled and the regimen it tested: CEE 0.625 mg + MPA 2.5 mg in 63-year-olds. It got over-extrapolated to a population it did not represent: the symptomatic 51-year-old starting transdermal estradiol + micronized progesterone. The 2022 Menopause Society position statement, the 2024 International Menopause Society (IMS) White Paper, and the German DGGG S3 Leitlinie 015/062 (2020 version; a 2026 update was put out for public consultation, with comments until 30 September 2026; the final version is pending) all reflect that revised consensus ^{[14, 15, 17]}.

Regulators have caught up. In November 2025 the FDA started removing the boxed warnings on heart disease, breast cancer and dementia from menopause hormone labels ^[36]. The boxed warning on endometrial cancer stays for systemic estrogen-only products ^[36]. The label now advises starting systemic HRT before 60 or within 10 years of menopause ^[36]. That is the same window this guide describes.

Yes, at least for your arteries. In the ELITE trial, estradiol started within 6 years of menopause slowed the thickening of the neck artery wall (0.0044 vs 0.0078 mm/yr, P=0.008) ^[11]. Started 10 or more years after menopause, it did nothing ^[11]. That is the timing hypothesis: confirmed on a surrogate marker, not on hard cardiovascular events like heart attacks or strokes. Two randomised trials after WHI tested it.

The cleanest answer comes from ELITE (NEJM, 2016) ^[11]. It enrolled 643 healthy postmenopausal women and split them by time since menopause: an early stratum (less than 6 years post-menopause) and a late stratum (10 or more years). Everyone got either oral micronized 17-beta-estradiol 1 mg/d plus sequential vaginal progesterone gel, or placebo, for a median of 5 years. The primary endpoint was the rate of carotid intima-media thickness (CIMT) progression. Think of CIMT as a measurable proxy for how fast plaque is building up in your neck arteries.

The timing made all the difference. Start early, and estradiol slowed plaque progression. Start late, and it did nothing.

• Early stratum: placebo 0.0078 mm/yr, estradiol 0.0044 mm/yr (P=0.008, a significant slowdown of atherosclerosis progression)
• Late stratum: placebo 0.0088 mm/yr, estradiol 0.0100 mm/yr (P=0.29, no benefit, trend slightly adverse)
• Time-by-treatment interaction: P=0.007

Coronary artery calcium did not differ in either stratum. ELITE put it plainly: "When estradiol therapy was initiated within 6 years after menopause, the rate of CIMT progression was significantly less in the estradiol group than in the placebo group, whereas when therapy was initiated 10 or more years after menopause, the rate of progression of CIMT was similar in the two groups." ^[11]

That is the empirical age-window evidence. The vascular benefit of estradiol is time-dependent. Start it early, it slows carotid IMT progression. Start it late, it does not.

One caveat. ELITE did not show reduced hard cardiovascular events (heart attack, stroke, CV death). CIMT is a surrogate marker. As of 2026, no randomised trial shows that HRT started early reduces hard CV endpoints.

The other trial, KEEPS (Annals of Internal Medicine, 2014), pushed back a little. It found no difference at all: CIMT grew 0.007 mm/yr in all three arms ^[12].

• Who: 727 healthy women aged 42 to 58, 6 to 36 months since their last period, no prior CV events
• What: oral CEE 0.45 mg/d, transdermal estradiol 50 micrograms/d, or placebo, all with cyclic micronized progesterone 12 days per month
• How long: 48 months
• Limits: lower doses than WHI (CEE 0.45 vs 0.625 mg), shorter, and underpowered (too small to detect differences in clinical CV events)

Its conclusion: "Four years of early MHT did not affect progression of atherosclerosis despite improving some markers of CVD risk." ^[12]

And the brain. A KEEPS side study on thinking and mood (KEEPS-Cog, PLoS Medicine, 2015) followed 693 women for 4 years ^[13]. MHT gave no cognitive benefit on any test: the Modified Mini-Mental, verbal learning and memory, attention and working memory, visual attention and executive function, or speeded language. The oral CEE arm showed modest reductions in depression and anxiety; transdermal estradiol did not.

This matters. The "HRT prevents dementia" claim cited in popular wellness media is not supported by KEEPS-Cog in recently menopausal women. The earlier WHI Memory Study (WHIMS) in women over 65 actually showed increased dementia incidence on CEE+MPA.

Where that leaves you: the age-window hypothesis holds on surrogate vascular endpoints (ELITE). It does not translate into a randomised hard-endpoint CV-prevention indication. HRT started in the 50-59 window is safe enough that the WHI extrapolation does not apply. But it is not a primary CV prevention drug, and KEEPS-Cog removed the dementia-prevention argument for recently menopausal women.

And how long can you stay on it? There is no fixed stop date. The Menopause Society sees no reason to stop routinely at 65 if symptoms persist, as long as you and your doctor review the choice every year ^[14].

HRT reliably cuts moderate-to-severe hot flashes and night sweats, by roughly 75 % ^[14]. It also treats the genitourinary syndrome of menopause (vaginal dryness, painful sex, recurrent UTIs) and protects bone in the right patient. For premature ovarian insufficiency (menopause before 40) it is standard of care. It does NOT reliably prevent heart disease, dementia or aging. The strongest indication is hot flashes and night sweats (vasomotor symptoms) in the 50-59 window. Everything else is secondary, neutral or unsupported.

If you are still mapping which symptoms are actually menopausal, the perimenopause guide covers staging and symptom interpretation before this protocol-level detail.

Here is the clean, evidence-aligned version of what MHT does at the population level.

HRT is established for:

Moderate-to-severe vasomotor symptoms (VMS, the hot flashes and night sweats). This is the only Level I, broadly accepted indication. The Menopause Society 2022 position statement (Menopause, 2022) ^[14] is verbatim: "Hormone therapy is the most effective treatment for vasomotor symptoms and the genitourinary syndrome of menopause." The effect size in symptomatic women is roughly 75 % fewer hot flashes.

Genitourinary syndrome of menopause (GSM, the vaginal atrophy, painful sex, and recurrent UTIs). Local low-dose vaginal estrogen (Vagifem, Estring, OeKolp estriol cream in DACH) is first-line. Systemic HRT also works. The systemic blood level from a vaginal tablet sits below the threshold that would require endometrial concern, so no progestogen is needed for local-only therapy. GSM, unlike VMS, does not resolve on its own. Left untreated, it tends to get worse over time.

Prevention of osteoporotic fracture in women at risk who cannot tolerate or have a contraindication to bisphosphonates. WHI 2002 hip fracture HR 0.66 (0.45 to 0.98) ^[1]. Bone is one of the few endpoints where the WHI primary results were directionally favourable. HRT is not first-line for established osteoporosis (bisphosphonates and denosumab are), but it is an evidence-based option in the right patient.

Premature ovarian insufficiency (POI, menopause before age 40). HRT to physiological age 50-51 is standard of care and not optional. This is a different population from "menopausal HRT" and rarely gets covered in commercial guides. If you reached menopause before 40, you should be on HRT unless contraindicated. That is the consensus across NAMS, IMS, ESHRE, and DGGG.

HRT is NOT reliably established for:

Primary CV prevention. ELITE shows attenuated CIMT but not hard events. ESC and ACC do not recommend HRT for primary CV prevention as of 2026.

Dementia or cognitive prevention. KEEPS-Cog showed no cognitive effect in recently menopausal women ^[13]. The WHI Memory Study (WHIMS) in women >65 showed increased dementia incidence on CEE+MPA. The "estrogen prevents Alzheimer's" claim is not supported in 2026 evidence.

General "anti-aging." No major society endorses MHT as a longevity drug. The 18-year Manson 2017 all-cause mortality data is neutral (HR 0.99) ^[5], not favourable.

Mood beyond perimenopausal mood symptoms. Some evidence for perimenopausal depressive symptoms in transitioning women. Not for general depressive disorder. SSRIs and SNRIs remain first-line for major depression, even during the menopausal transition.

Put together: the strongest indication is VMS in the 50-59 window. Everything else is either secondary (GSM, bone), neutral (all-cause mortality), or not supported (CV prevention, dementia, anti-aging). A symptomatic woman who wants symptom relief deserves a real conversation about HRT. An asymptomatic woman looking for a longevity drug should not be sold MHT.

Cannot or do not want to take hormones? Since December 2023 the EU has licensed a non-hormonal pill for hot flashes, fezolinetant (Veoza). In the SKYLIGHT 1 trial it cut hot flash frequency and severity more than placebo ^[41]. NICE also recommends cognitive behavioural therapy (CBT) for hot flashes and low mood ^[34].

Micronized progesterone, not a synthetic progestin. Progestogen is the umbrella term for the second hormone a woman with a uterus takes alongside estrogen. Which one she takes drives much of the extra breast cancer signal in combined HRT. In the French E3N cohort, synthetic progestins raised ductal breast cancer risk (relative risk, RR 1.6) and lobular risk (RR 2.0) ^[8]. Micronized progesterone showed no significant overall increase, though risk crept up with longer use ^[8]. A 2018 review puts the reassuring window at up to 5 years ^[9]. For a woman with an intact uterus, that makes micronized progesterone the default.

This is the single most important, and most under-discussed, lever in modern MHT.

The clearest read on this comes from the French E3N cohort (Journal of Clinical Oncology, 2008), which tracked around 80,000 postmenopausal women over a mean of 8.1 years ^[8]. When you break the breast cancer risk down by which progestogen women took, the picture splits sharply:

• Estradiol + dydrogesterone (Duphaston class): lobular cancer RR 1.7 (95% CI 1.1 to 2.6)
• Estradiol + "other progestins" (synthetic norpregnane or nortestosterone-derived): ductal cancer RR 1.6 (95% CI 1.3 to 1.8) and lobular RR 2.0 (95% CI 1.5 to 2.7)
• Estradiol + micronized progesterone: no significant overall breast cancer risk increase, though lobular and ER+/PR- risk rose with longer use ^[8]

The progestogen identity matters more than the estrogen identity for breast cancer risk.

A systematic review (Climacteric, 2018) ^[9] reached the same conclusion: "Estrogens combined with oral (approved) or vaginal (off-label use) micronized progesterone do not increase breast cancer risk up to 5 years of treatment duration. There is limited evidence that estrogens combined with oral micronized progesterone applied for more than 5 years are associated with an increased breast cancer risk."

The big observational counterweight is the Million Women Study (Lancet, 2003), which followed 1,084,110 UK women aged 50 to 64 ^[7]. Comparing current HRT users to never users:

• Any current HRT use: RR 1.66 (95% CI 1.58 to 1.75)
• Estrogen alone: RR 1.30 (95% CI 1.21 to 1.40)
• Estrogen + progestagen: RR 2.00 (95% CI 1.88 to 2.12)

In absolute terms, 10 years of HRT use leads to +5 breast cancers per 1,000 (estrogen-only) or +19 per 1,000 (combined) in women aged 50 to 64. Million Women is observational with documented healthy-user and detection-bias concerns (a 2012 critique in JFPRHC laid these out), but combined with WHI it still anchors the modern breast cancer risk estimate.

The modern DACH recommendation. A woman with an intact uterus who needs systemic estrogen gets micronized progesterone (Utrogest from Besins Healthcare, Famenita from Exeltis Germany GmbH) 100 mg nightly continuous, or 200 mg cyclic days 14 to 25. Synthetic progestins (MPA in Clinofem; norethisterone acetate; levonorgestrel; dienogest, marketed as Visanne by Bayer) carry larger breast cancer signals.

One terminology note that matters for DACH readers. "Bioidentical" simply means molecularly identical to the body's own hormone. Estradiol patches (Estradot from Sandoz/Hexal, Estramon from Hexal, Climara from Bayer) and micronized progesterone (Famenita, Utrogest) are bioidentical. They are also licensed, evidence-based, GMP-manufactured pharmaceuticals.

This is not the same thing as US-style "compounded bioidentical hormone therapy" (cBHT) with pellet implants, custom troche compounds, and saliva-tested protocols. The US National Academies of Sciences, Engineering, and Medicine 2020 report on the clinical utility of compounded bioidentical hormone therapy explicitly recommended against routine cBHT due to dose inconsistency and lack of safety data ^[31]. In DACH, when a gynaecologist says "bioidentisch," she almost always means a licensed estradiol + Famenita regimen, which is fine and evidence-based. Do not let US-influencer cBHT marketing creep into the DACH framing.

Yes. And the difference is big enough that the route is now the single biggest VTE-safety lever in modern MHT.

A meta-analysis of fifteen observational studies (Journal of Clinical Endocrinology & Metabolism, 2015) compared oral to transdermal estrogen head to head ^[10]. Oral lost on clotting:

• Venous thromboembolism (VTE, the blood clots in deep veins or the lungs): RR 1.63 (95% CI 1.40 to 1.90) for oral vs transdermal
• Deep vein thrombosis specifically: RR 2.09 (95% CI 1.35 to 3.23)
• Stroke looked slightly higher on oral in a single study; heart attack risk did not differ. All of this is observational, low-certainty data ^[10]

Why? A pill has to pass through your liver before it reaches your blood. Doctors call this the first-pass effect. With oral CEE or oral estradiol, the liver responds by making more clotting proteins (factor VII, prothrombin, fibrinogen), more of the inflammation marker CRP, and more sex hormone binding globulin (SHBG). A patch or gel skips the liver and goes straight into the bloodstream, so that liver response does not show up.

What that means in DACH: First-line MHT in DACH 2026 is transdermal estradiol (patch or gel) + oral micronized progesterone. Oral estradiol is second-line, reserved for women with no VTE risk who prefer pills. CEE (Premarin in the US, Presomen historically in DE) has been largely phased out in DACH in favour of 17-beta-estradiol. Transdermal is dominant.

The guidelines line up behind this. The DGGG S3 Leitlinie 015/062 ("Peri- und Postmenopause, Diagnostik und Interventionen," 2020; English short version 2021) says women must be told that the thromboembolism risk "associated with oral estrogen intake is higher than for transdermal applications" ^[17]. The Menopause Society 2022 statement reinforces it: "Transdermal routes of administration and lower doses of hormone therapy may decrease risk of venous thromboembolism." ^[14]

Practical translation. Got any of the following? Past VTE, factor V Leiden or other thrombophilia, BMI over 30, active smoking, age over 60 at initiation, prolonged immobility. Then transdermal estradiol is the default. If you have no risk factors and a strong preference for oral, low-dose oral estradiol with micronized progesterone is reasonable. CEE is not.

Only for one thing: low sexual desire. The 2019 Global Consensus, endorsed by the Endocrine Society, IMS, NAMS, and EMAS, found that hypoactive sexual desire disorder (HSDD) is the single evidence-based indication, and that there are insufficient data for bone, mood, cognition, body composition, or energy ^[19]. No FDA-, EMA-, or DACH-licensed female formulation exists. This is one of the most misrepresented topics in current wellness marketing for women, including in broader longevity advice for women.

A review (Lancet Diabetes & Endocrinology, 2015) ^[18] laid the groundwork: sexual dysfunction (low desire) is the only validated indication, and there is no FDA-approved female formulation.

Then the field reached consensus. The Global Consensus Position Statement on testosterone therapy for women (Journal of Clinical Endocrinology & Metabolism, 2019) was co-published in Climacteric, Maturitas and the Journal of Sexual Medicine ^[19]. It was endorsed by the Endocrine Society, the International Menopause Society (IMS), the North American Menopause Society (NAMS), the European Menopause and Andropause Society (EMAS), ISSM, ISSWSH, RCOG and RANZCOG, among others ^[19].

The headline, in its own words: "The only evidence-based indication for testosterone for women is for HSDD" (HSDD, hypoactive sexual desire disorder; the formal name for clinically significant low libido in postmenopausal women). And: "There are insufficient data for using testosterone for any other symptom/condition or for disease prevention." That explicitly rules out bone, mood, cognition, well-being, body composition, and energy as indications ^[19].

No FDA-approved or EMA-approved female testosterone formulation exists in the US, EU, or DACH. AndroFeme was first licensed in Australia (TGA, 2020), and in 2025 the UK MHRA granted it a national marketing authorisation for HSDD in postmenopausal women (the first licensed female testosterone product in the UK). The UK is post-Brexit non-EU, so this is not an EU or DACH authorisation: there is still no licensed female testosterone product in Germany, Austria, or Switzerland, and none approved by the EMA.

Off-label prescribing of approved male formulations at female-physiological doses is reasonable if serum concentrations stay in the premenopausal female range (roughly 0.5 to 2.5 nmol/L total testosterone, with the actual target depending on assay). Compounded products (saliva-tested, custom-troche, pellet) are explicitly not recommended by the 2019 Global Consensus unless approved alternatives are unavailable ^[19]. Baseline testosterone measurement, repeat at 3 to 6 weeks, then 6-monthly; discontinue if no benefit by 6 months.

DACH reality 2026. No licensed female testosterone product exists in Germany, Austria, or Switzerland. The typical Selbstzahler workaround is off-label use of male Testogel (Besins) or Tostran (Advanz Pharma GmbH) at roughly one-tenth of the male dose (one pump dose every 4 to 7 days, applied to the thigh or lower abdomen). This is not GKV-reimbursed, requires a Privatrezept, and should be done by an endocrinologist or experienced gynaecologist with proper laboratory monitoring. Not via the direct-to-consumer telemedicine "menopause hormone" platforms that have entered the DACH market since 2024.

One word on the marketing claims. Recent influencer and wellness-platform messaging often presents testosterone as a generalised "vitality, mood, energy, and body-composition" optimiser for menopausal women. The 2019 Global Consensus is explicit that the evidence does not support this. HSDD is the only validated indication. A woman who is doing well sexually but considering testosterone for energy or body composition is being sold a non-evidence-based intervention.

The headline first: when there are symptoms and a low testosterone level confirmed on two morning measurements. Both. Not one. If you are weighing TRT as part of a broader plan, the men 40-plus guide frames where hormones sit relative to the bigger levers. Here is the evidence.

Start with the Testosterone Trials, or T-Trials (NEJM, 2016) ^[20]. They enrolled 790 men aged 65 or older with serum T below 275 ng/dL (9.5 nmol/L), and gave them testosterone gel or placebo for one year. What helped, and how much:

• Sexual function: significant increases in sexual activity (P less than 0.001), desire, and erectile function, with a PDQ-Q4 score change of about 0.5 standard deviation
• Physical function (alone): no significant difference in 6-minute walking distance
• Physical function (all three trials pooled): the share of men with a 50 m or greater improvement differed, 20.5% on testosterone vs 12.6% on placebo (P=0.003)
• Vitality: no significant benefit on FACIT-Fatigue, with only modest mood improvement

What that adds up to: in older men with hypogonadism (testosterone below the clinical threshold, plus symptoms), testosterone produces a moderate sexual function benefit, a marginal physical function benefit, and no meaningful vitality benefit. This is not a vitality drug. The marketing claim that TRT will give you "the energy you had at 25" is not supported by the T-Trials.

The Endocrine Society Clinical Practice Guideline (Journal of Clinical Endocrinology & Metabolism, 2018) sets the diagnostic bar: "We recommend making a diagnosis of hypogonadism only in men with symptoms and signs consistent with testosterone deficiency and unequivocally and consistently low serum T concentrations." ^[21]

The diagnostic algorithm requires both:

1. Symptoms or signs consistent with testosterone deficiency (low libido, erectile dysfunction, loss of body or facial hair, reduced muscle mass and strength, fatigue, depressed mood)
2. Unequivocally and consistently low serum testosterone on two morning fasting measurements with accurate assays.

In borderline cases, or when SHBG (the protein that carries testosterone in your blood) is off because of obesity, thyroid disease or advanced age, the guideline recommends measuring free testosterone, by equilibrium dialysis or a validated formula ^[21].

The guideline recommends against starting TRT if you have ^[21]:

• active plans to father a child
• prostate or breast cancer
• a palpable prostate nodule or induration
• PSA over 4 ng/mL (or over 3 in high-risk men)
• high hematocrit
• untreated severe sleep apnea
• severe lower urinary tract symptoms
• uncontrolled heart failure
• a heart attack (MI) or stroke within the last 6 months
• thrombophilia (a tendency to form blood clots)

The urologists draw the line slightly differently. The AUA Guideline (Journal of Urology, 2018) ^[22] uses 300 ng/dL as the diagnostic threshold (versus the Endocrine Society's 264 ng/dL), based on a different cohort reference. Two morning fasting measurements are still required.

Where does 264 ng/dL come from? A harmonisation effort across four cohort studies (Journal of Clinical Endocrinology & Metabolism, 2017): Framingham Heart Study, EMAS, MrOS (Osteoporotic Fractures in Men Study), and the Male Sibling Study of Osteoporosis (MSOS), 9,054 community-dwelling men ^[23]. The harmonised 2.5th percentile lower normal limit in healthy non-obese men aged 19 to 39, using the CDC reference method, is 264 ng/dL (9.2 nmol/L). That is the basis for the Endocrine Society's diagnostic threshold and the global harmonisation effort.

And how common is real deficiency? The European Male Aging Study, or EMAS (NEJM, 2010) ^[24], enrolled 3,369 European men aged 40 to 79. It defined late-onset hypogonadism as at least three sexual symptoms (poor morning erections, low desire, ED) plus total T below 11 nmol/L (3.17 ng/mL) plus free T below 220 pmol/L. Population prevalence: roughly 2% of men aged 40 to 79.

The marketing claim that 25 to 40% of older men have "low T" rests on low total T alone, without symptom confirmation. The clinical truth from EMAS is much narrower. 2 percent. Not 25.

In European practice, including DACH, doctors usually follow the European Association of Urology (EAU) cut-off: about 12 nmol/L (around 346 ng/dL) total testosterone on two morning samples, plus symptoms ^[39]. That bar is higher, so less strict, than the Endocrine Society's 264 ng/dL, and close to the EMAS cut-off of 11 nmol/L ^[24]. There is no fixed GKV number. The Kasse pays when hypogonadism is confirmed by symptoms and lab values within the licensed indication. Below the cut-off, persistent symptoms (especially low libido, erectile dysfunction unresponsive to PDE5 inhibitors, or unexplained loss of muscle mass) make you a candidate for TRT. Above 12 nmol/L, TRT is generally not indicated.

TRAVERSE settled the big one: in 5,246 hypogonadal men with cardiovascular risk, TRT was non-inferior to placebo for major adverse cardiac events (MACE 7.0% vs 7.3%, HR 0.96, 95% CI 0.78 to 1.17) ^[25]. It led the FDA to drop its cardiovascular boxed-warning language in 2025 ^[37]. But it only covers confirmed hypogonadism, and it flagged modest signals for atrial fibrillation, pulmonary embolism, and acute kidney injury ^[25].

The TRT-and-cardiovascular-risk debate has its own ten-year arc.

It started with two scares. First, a VA cohort study (JAMA, 2013) of 8,709 men with T below 300 ng/dL ^[26]. It reported a 29% higher adjusted risk of death, heart attack or stroke with testosterone (HR 1.29, 95% CI 1.04 to 1.58) ^[26]. The 3-year event rates were 25.7% on testosterone vs 19.9% without, an absolute difference of 5.8% whose confidence interval crossed zero (95% CI -1.4% to +13.1%) ^[26]. The analysis drew heavy criticism and needed published data corrections, but it was the original heart scare on TRT. Second, an insurance claims analysis (PLoS One, 2014) reported a 2- to 3-fold increase in non-fatal heart attacks (MI) in the first 90 days of TRT among older men ^[27]. Its methods were criticised too.

Those two papers (2013 and 2014) drove the FDA's 2015 cardiovascular label warning on testosterone products ^[37].

Then came TRAVERSE (NEJM, 2023) ^[25]. It enrolled 5,246 men aged 45 to 80 with pre-existing or high cardiovascular risk plus documented hypogonadism (two morning T levels below 300 ng/dL plus symptoms). They got transdermal 1.62% testosterone gel (dose-adjusted to a 350 to 750 ng/dL target) or placebo. Mean treatment was 21.7 months; mean follow-up was 33 months.

The primary result cleared the heart concern:

• Primary MACE outcome (MACE, the composite endpoint of CV death, heart attack, or stroke): 7.0% on testosterone vs 7.3% on placebo. HR 0.96 (95% CI 0.78 to 1.17). P less than 0.001 for non-inferiority

The secondary CV composites told the same story. The conclusion, in its own words: "Testosterone-replacement therapy was noninferior to placebo with respect to the incidence of major adverse cardiac events." ^[25]

In February 2025 the FDA acted on this. It removed the heart-risk language from the boxed warning on all testosterone products and added the TRAVERSE results to the labels ^[37]. It also added a class-wide warning that testosterone can raise blood pressure ^[37]. The label still says testosterone is not approved for low testosterone caused by aging alone ^[37].

TRAVERSE settles the major cardiovascular mortality question for hypogonadal men. The 2015 warning, in retrospect, was overstated for the symptomatic, properly diagnosed hypogonadal population. But TRAVERSE also flagged real signals that must be in informed consent:

• Atrial fibrillation: 3.5% testosterone vs 2.4% placebo
• Pulmonary embolism: 0.9% vs 0.5%
• Acute kidney injury: 2.3% vs 1.5%
• Prostate: PSA rose more on testosterone, but high-grade cancer (0.19% vs 0.12%), any prostate cancer and biopsy rates did not differ significantly. Men with PSA over 3 were excluded ^[38]

Critical caveats. TRAVERSE was in hypogonadal men with confirmed deficiency. It does NOT validate TRT for men with normal testosterone, for athletic enhancement, or for general anti-aging. The 33-month follow-up does not address 10-plus year exposure questions. The AF, AKI, and PE signals are real and should change the conversation, not the prescription, for properly indicated patients.

The arc: the 2013 and 2014 scares led to the FDA's 2015 label warning. TRAVERSE 2023 then reversed the major heart concern, and in 2025 the FDA followed ^[37]. But TRAVERSE also confirmed TRT is not risk-free. Where things stand in 2026: TRT in confirmed hypogonadism does not raise MACE. It does raise AF, PE and AKI modestly, and PSA rises more, without a significant difference in prostate cancer or biopsies ^[38]. The size of those signals is acceptable for symptomatic patients with confirmed deficiency. It is not acceptable for asymptomatic eugonadal men.

Estradiol belongs in a man's body. Suppressing it prophylactically with an aromatase inhibitor on TRT works against you: Finkelstein 2013 showed that estrogen deficiency, not testosterone deficiency, mainly drives fat gain in men, and that estradiol also contributes to sexual function ^[28]. Reserve aromatase inhibitors for confirmed, clinically significant gynaecomastia only, never as routine prophylaxis.

The cleanest evidence here comes from Finkelstein 2013 (NEJM), an elegant dose-ranging design ^[28]. Healthy men aged 20 to 50 had their endogenous testosterone suppressed with a GnRH analogue (goserelin), then received graded testosterone gel doses with or without anastrozole (an aromatase inhibitor, the drug that blocks the conversion of testosterone to estradiol). The point was to isolate the independent effects of testosterone versus aromatised estradiol.

The findings were unusually clean. Body composition: estrogen deficiency (the anastrozole arm) drove fat accumulation, not testosterone deficiency per se. Lean mass and strength: testosterone-dependent. Sexual desire: both testosterone and estradiol contributed.

Translation: estradiol, not just testosterone, is essential to male body composition and sexual function. It matters for male bone too.

This matters because a section of online TRT culture treats estradiol as the enemy. "Crashed E2," anastrozole prescribed prophylactically alongside TRT, exemestane to "keep E2 low." The bro-science position is that estradiol is a feminising side-effect to be suppressed. Finkelstein 2013, in healthy men whose hormones were switched off and dosed back, showed that removing estradiol raised body fat, including belly fat, and lowered sexual function ^[28]. Crashing estradiol with an aromatase inhibitor works against you.

Modern TRT counselling: do NOT use aromatase inhibitors prophylactically. Reserve them for refractory, clinically significant gynaecomastia confirmed on examination. Not for cosmetic concerns. Not for E2 numbers that look high on a lab printout when the patient has no symptoms.

Some estradiol in TRT is desired, not a side effect. The reference range matters. E2 levels in TRT typically run higher than the male baseline reference because exogenous testosterone gives more substrate for aromatisation. A man on TRT with E2 at the upper end of the male reference (40-60 pg/mL) and no symptoms is doing fine. A man on TRT with E2 suppressed below 20 pg/mL by AI use is potentially harming himself.

The DACH clinical position aligns. Anastrozole is not licensed for male hypogonadism in Germany, Austria, or Switzerland. It is licensed for breast cancer in women. Its off-label use as a TRT adjunct is not GKV-reimbursed and requires a Privatrezept. Most reputable endocrinologists in DACH avoid it prophylactically, for the reasons Finkelstein 2013 makes clear.

Short answer: no. Neither one extends life at the population level on the all-cause mortality endpoint. Here is what the data actually says.

The biggest dataset is Yeap 2024 (Annals of Internal Medicine) ^[29], an individual participant data meta-analysis of nine prospective cohorts (255,830 participant-years) plus summary data from eleven more (24,109 men).

The headline thresholds: men with baseline endogenous total T <7.4 nmol/L (213 ng/dL) had higher all-cause mortality. Men with T <5.3 nmol/L (153 ng/dL) had higher CV mortality. LH >10 IU/L and very low estradiol (<5.1 pmol/L) also independently predicted mortality.

This is the largest natural-history dataset on endogenous testosterone and mortality available in 2026. The threshold at which low T becomes mortality-relevant is around 7 nmol/L total. That is well below even the EAU cut-off of 12 nmol/L, let alone the higher "optimal" ranges wellness clinics promote.

And here is the part that gets lost. The Yeap 2024 signal is for endogenous testosterone, not exogenous TRT. The observation that low-T men die more does not mean correcting their T pharmacologically extends their life. The leading interpretation is reverse causation: low T is partly a biomarker of underlying poor health (obesity, chronic inflammation, depression, multimorbidity). The men with the lowest T are sick from causes that exogenous testosterone does not fix.

Layered on the women's side: Manson 2017 (above). 18-year all-cause mortality on MHT, HR 0.99 (0.94 to 1.03) ^[5]. MHT does not extend life at the population level.

So: Neither HRT nor TRT is a longevity drug at the all-cause mortality endpoint. Both are symptom-modifying or hormone-deficiency-correcting therapies that, when properly indicated, have favourable benefit-risk in their indicated populations. Both get mis-marketed as longevity tools by commercial telemedicine clinics and influencer-driven content. That is the core point of this guide.

What is true:

• HRT is the most effective treatment for VMS and the only Level I indication. Secondary indications are GSM, fracture prevention in selected patients, and POI.
• TRT is the appropriate treatment for confirmed hypogonadism per Endocrine Society 2018 and EAU criteria ^{[21, 39]}, and TRAVERSE 2023 established it does not increase MACE in this population ^[25].
• The 18-year all-cause mortality on MHT is HR 0.99. Neutral, not favourable.
• Endogenous T <7.4 nmol/L predicts mortality, but this does not mean exogenous T extends life in men with normal T.
• Neither therapy is licensed in DACH or anywhere else for "longevity" or "healthy aging." Both are licensed for symptoms (VMS, GSM) or confirmed deficiency (hypogonadism).

What is being marketed beyond this ("hormone optimisation," "perimenopausal anti-aging stack," "andropause reversal," "bioidentical wellness") is consumer wellness positioning, not medicine. A symptomatic woman in the 50-59 window has a real conversation to have about HRT for her quality of life. An asymptomatic 55-year-old man with a single-sample T of 13 nmol/L being sold "hormone optimisation" by an online clinic does not.

Statutory insurance (GKV) pays for HRT prescribed for menopausal symptoms. You pay only the 5-10 € co-pay (Zuzahlung). It also pays for TRT once hypogonadism is documented: symptoms plus two morning fasting testosterone results below threshold (in European practice usually <12 nmol/L) ^[39]. Off-label or wellness-positioned hormone use for asymptomatic, eugonadal people is Selbstzahler. HRT is one of the few longevity-adjacent therapies that is straightforwardly GKV-covered; for the broader picture of what you pay out of pocket, see the Selbstzahler longevity guide.

HRT / MHT in Germany 2026, the licensed product lineup:

GKV (statutory insurance) coverage in Germany. All the licensed MHT preparations listed above (estradiol patches, gels, oral; micronized progesterone Utrogest/Famenita; combined products) are verschreibungspflichtig (prescription-only) and generally covered by GKV when prescribed for the licensed indication of menopausal symptoms. The standard 5 € or 10 € Zuzahlung applies. HRT is one of the few longevity-adjacent therapies that is straightforwardly GKV-covered in Germany. No special documentation hurdle.

Approximate Selbstzahler / Privatrezept HRT prices (Shop-Apotheke, Apotal, DocMorris snapshot 2026; verify before purchase):

• Famenita 100 mg, 90 capsules: ~€30-40 (3 months at 1/day)
• Estradot 50 mcg, 8 patches: ~€20-25 (~1 month)
• Gynokadin Dosiergel 80 g: ~€25-30 (~2 months)

TRT in Germany 2026, the licensed product lineup:

GKV coverage for TRT in Germany. GKV pays for TRT when documented hypogonadism is established by:

1. Symptoms consistent with androgen deficiency.
2. Two morning fasting total testosterone measurements below threshold. In European practice that is usually the EAU cut-off of about 12 nmol/L (~346 ng/dL) ^[39], which is less strict than the Endocrine Society's 264 ng/dL. There is no fixed GKV number; what counts is confirmed hypogonadism within the licensed indication.
3. Documented in the medical record. Off-label or "wellness" TRT for asymptomatic eugonadal men is Selbstzahler.

Approximate Selbstzahler / Privatrezept TRT prices:

• Nebido 1000 mg/4 ml ampoule: about €140-170 per ampoule (every 10-14 weeks)
• Testogel sachets 50 mg, 30 sachets: ~€60-80 per month
• Testavan pump: ~€60-80 per month
• Plus physician fees (the Nebido injection itself is usually done at a Hausarzt or urology practice)

Telemedicine HRT/TRT providers active in DE/AT/CH 2026 (verified web presence at research date; listed for transparency only, not endorsed by this guide):

For women / HRT: Hormone Online Clinic (Dr Sheila de Liz), Hormonic Care, Hormona/MyHormona, and various other DTC platforms that have entered the market since 2024.

For men / TRT: Swiss TRT (swiss-trt.ch) for Switzerland; various German private urology practices marketing TRT under "Männergesundheit" banners. US telehealth chains are arriving too: Hims & Hers bought the European platform ZAVA in 2025, which also serves Germany.

The regulatory context. Telemedicine HRT/TRT providers face structural constraints in DACH: §10 Heilmittelwerbegesetz (HWG) bans advertising prescription drugs to the public, and §9 HWG limits advertising for remote treatment. Remote prescribing must follow the doctors' professional code (§7(4) MBO-Ä), which allows it only when a doctor judges it medically justifiable. Cross-border prescribing is restricted. Some platforms operate via Wahlarzt-style Privatrezept and avoid GKV entirely.

For TRT specifically, German prescription data documented in Pharmazeutische Zeitung (September 2025) shows a 415 % increase in testosterone prescriptions (Verordnungen) 2005-2023. The site mann-und-gesundheit.com put it bluntly: testosterone medicines are only licensed for specific conditions, which raises the suspicion of misuse ("Testosteron-haltige Arzneimittel sind eigentlich nur bei bestimmten Erkrankungen zugelassen, Verdacht auf eine missbräuchliche Verwendung"). That is the German parallel to the US low-T-clinic phenomenon. Put simply, GKV-reimbursed TRT requires the confirmed-hypogonadism criteria above. Anything else is private-pay, off-label-positioned, and operates in a clinical grey zone.

Both therapies have real risk profiles. A consent conversation that skips them is not consent.

Breast cancer (women on combined MHT). WHI CEE+MPA: HR 1.26 (1.00 to 1.59), absolute excess ~8 cases per 10,000 person-years ^[1]. Million Women: +19/1,000 over 10 years for combined HRT in women aged 50-64 ^[7]. Micronized progesterone partly mitigates this signal (the E3N cohort, JCO 2008, and the Climacteric review, 2018) ^{[8, 9]}. There is no risk-free window. In the 2019 Lancet pooled analysis, combined HRT raised breast cancer risk from the first years of use (RR 1.60 in years 1 to 4), and the risk rose with duration ^[35]. Five years of combined HRT from age 50 means roughly 14 to 20 extra breast cancers per 1,000 women over the next 20 years (DGGG S3) ^[17]. Estrogen-only carries a smaller excess ^[35].

A 2019 Lancet meta-analysis from the Collaborative Group on Hormonal Factors in Breast Cancer (CGHFBC) found higher risk with every systemic MHT type, rising with duration and partly persisting for over 10 years after stopping ^[35]. The EMA PRAC recommendation of May 2020 updated product labelling accordingly ^[32]. The takeaway: real signal, magnitude varies by progestogen type, and the absolute numbers in the symptomatic 50-55-year-old window remain favourable on benefit-risk balance.

Venous thromboembolism (women on oral estrogen). The Mohammed meta-analysis (J Clin Endocrinol Metab, 2015): oral vs transdermal RR 1.63 (1.40 to 1.90) ^[10]; transdermal estradiol is the safer route. For women with thrombophilia, prior VTE, BMI >30, smoking, or a family history of VTE, transdermal is mandatory, not optional.

Endometrial cancer (women with intact uterus on unopposed estrogen). Unopposed systemic estrogen in women with a uterus drives endometrial hyperplasia and adenocarcinoma. The risk rises several-fold. Progestogen is non-negotiable for women with an intact uterus on systemic estrogen. Vaginal low-dose estrogen for GSM does not require progestogen because systemic absorption stays below the threshold of endometrial concern.

Stroke (women on systemic MHT, oral routes especially). WHI CEE+MPA: HR 1.41 ^[1]. WHI CEE-alone: HR 1.39 ^[2]. The risk is concentrated in older starters and oral routes, and mitigated by early initiation and the transdermal route.

Polycythemia / hematocrit elevation (men on TRT). The threshold to pause TRT is typically hematocrit >54 % ^[21]. More common with IM testosterone (Nebido) than gels. More common in obese men, smokers, and those with sleep apnea. Periodic CBC monitoring (every 3-6 months in the first year, then annually) is part of competent TRT care.

Prostate signals (men on TRT). In TRAVERSE, PSA rose more on testosterone, but high-grade cancer, any prostate cancer and biopsy rates did not differ significantly ^[38]. Men with PSA over 3 were not enrolled. A pre-existing palpable prostate nodule or PSA >4 (or >3 in high-risk men) is a contraindication per Endocrine Society 2018 ^[21].

Blood pressure (men on TRT). Since 2025 the FDA requires every testosterone product to warn that it can raise blood pressure. Ambulatory monitoring studies found the rise across all products ^[37]. Check your blood pressure before you start and at every follow-up.

Pulmonary embolism (men on TRT). TRAVERSE: 0.9 % vs 0.5 % ^[25]. Modest but real.

Atrial fibrillation (men on TRT). TRAVERSE: 3.5 % vs 2.4 % ^[25]. The biggest "new" TRAVERSE signal. AF risk should be discussed before initiation, particularly in men with hypertension, sleep apnea, or a pre-existing arrhythmia history.

Acute kidney injury (men on TRT). TRAVERSE: 2.3 % vs 1.5 % ^[25].

The ADT mirror. Androgen deprivation therapy (ADT) for prostate cancer increases cardiovascular morbidity, about 20% more serious heart and vessel events (Saigal 2007, Cancer) ^[30]. The implication for the longevity guide: suppressing testosterone in men carries CV harm, which frames the symmetric question. Boosting testosterone in normal men is also not free of risk. Hormonal homeostasis is not casually moved up or down.

Compounded products and pellet implants. The US National Academies of Sciences, Engineering, and Medicine 2020 report on the clinical utility of compounded bioidentical hormone therapy explicitly recommended against routine cBHT due to dose inconsistency and lack of safety data ^[31]. In DACH, pellet implants are rare; the term "bioidentisch" almost always refers to licensed Estradot + Famenita, which is fine and evidence-based. If a DACH provider is selling you saliva-tested compounded pellets or troches, you are outside the evidence base that supports modern MHT.

AndroFeme was first licensed in Australia (TGA, 2020) for female HSDD, and the UK MHRA granted it a national marketing authorisation in 2025, the first licensed female testosterone product in the UK. The UK is post-Brexit non-EU, so this is not an EU or DACH authorisation and there is still no EMA-approved female testosterone product.

Risks, summed up. HRT and TRT have well-characterised risk profiles. The risk magnitudes are acceptable for symptomatic women in the 50-59 window on transdermal estradiol + micronized progesterone, and for men with confirmed hypogonadism on properly monitored TRT. They are not acceptable for asymptomatic women or for eugonadal men chasing "hormone optimisation." Inform consent. Monitor. Be honest about what the trials actually measured.