Low testosterone (below ~213 ng/dL or 7.4 nmol/L) is consistently associated with increased all-cause and cardiovascular mortality in men, across 11 prospective studies totaling over 24,000 participants. The TRAVERSE trial (5,246 men, NEJM 2023) established that TRT does not increase major cardiovascular events — the FDA removed the boxed warning in February 2025. But there's a catch: a 2026 Mendelian randomization study found genetically higher lifelong testosterone associated with ~17% increased coronary artery disease risk, suggesting endogenous vs exogenous testosterone may carry different risk profiles. Optimal for longevity appears to be mid-physiologic: 550–700 ng/dL total T — not the highest levels you can achieve, not the "normal" bottom of the reference range.
Why Testosterone Matters for Longevity
Testosterone isn't a sex hormone. That's its most famous job, not its most important one. Testosterone is a systemic regulatory hormone that influences lean mass preservation, fat distribution, insulin sensitivity, bone mineral density, erythropoiesis, mood regulation, and cognitive function. When it declines — and it declines roughly 1–2% per year after age 30 — every one of those systems degrades.
The longevity relevance is direct: sarcopenia (muscle loss) is one of the strongest predictors of all-cause mortality in aging. Insulin resistance drives metabolic disease. Bone density determines fracture risk, which determines independence. Testosterone sits upstream of all of them.
The question has never been whether testosterone matters. The question is what levels are optimal, whether replacement therapy in deficient men improves outcomes, and whether pushing levels higher in the "normal range" buys anything for healthy men.
The Mortality Data: What Low Testosterone Does
The 24,000-Man Meta-Analysis
A systematic analysis of 11 prospective cohort studies followed over 24,000 community-dwelling men for a minimum of 5 years each. The finding was unambiguous: low baseline testosterone is associated with increased risk of all-cause mortality. Men with total testosterone below 7.4 nmol/L (~213 ng/dL) showed elevated cardiovascular death risk, with those below 5.3 nmol/L (~153 ng/dL) at the highest risk.
This isn't one study or one population. It's a consistent signal across multiple cohorts, measured with mass spectrometry (the gold standard), with follow-up periods long enough to capture real mortality differences.
A 9,000-Man Real-World Cohort
A 2026 retrospective cohort study of over 9,000 men on testosterone therapy showed that adequately treated hypogonadal men had comparable mortality rates to eugonadal controls — and significantly lower mortality than untreated hypogonadal men. This is the closest thing to a real-world confirmation that bringing low testosterone back to physiologic levels may reduce mortality risk.
What the Data Doesn't Show
The mortality data establishes that low testosterone is associated with higher death risk. It does not prove that raising testosterone causes lower death risk. Low testosterone may be a marker of poor health (obesity, metabolic syndrome, chronic disease) rather than a cause. The best available evidence suggests it's both — a consequence of declining health and an independent contributor to further decline — but no RCT has powered for mortality as a primary endpoint.
The TRAVERSE Trial: Cardiovascular Safety Settled
For over a decade, the primary concern about TRT was cardiovascular risk. Observational studies in the early 2010s suggested increased heart attack and stroke risk. The FDA added a boxed warning to all testosterone products in 2015.
The TRAVERSE trial (New England Journal of Medicine, 2023) was designed to answer this definitively:
TRAVERSE Trial Design
- Participants: 5,246 men aged 45–80 with documented hypogonadism (total T <300 ng/dL) and existing or high-risk cardiovascular disease
- Intervention: Testosterone gel vs placebo, mean 33 months follow-up
- Primary endpoint: Major adverse cardiovascular events (MACE: cardiovascular death, nonfatal MI, nonfatal stroke)
- Result: 7.0% testosterone group vs 7.3% placebo — HR 0.96 (95% CI 0.78–1.17), p < 0.001 for non-inferiority
Translation: TRT did not increase cardiovascular events, even in a population specifically selected for high cardiovascular risk. This was the highest-risk population you could test — if TRT were going to cause heart attacks, it would show up here. It didn't.
Based on TRAVERSE, the FDA removed the cardiovascular boxed warning from testosterone products in February 2025. But it wasn't a clean victory — several important signals emerged:
| Adverse Event | Testosterone Group | Placebo Group | Implication |
|---|---|---|---|
| Major CV events (MACE) | 7.0% | 7.3% | No increase — safety confirmed |
| Nonfatal arrhythmias | 5.2% | 3.3% | Numerically higher — monitor |
| Atrial fibrillation | 3.5% | 2.4% | Signal — ECG monitoring warranted |
| Acute kidney injury | 2.3% | 1.5% | Monitor renal function |
| Clinical fractures | Higher (HR 1.43) | Lower | Unexpected — contradicts bone density data |
TRAVERSE found a 43% higher fracture risk in the testosterone group compared to placebo. This contradicts decades of bone density research showing testosterone improves BMD. The finding is unexplained — it may reflect increased physical activity (more falls), may be a statistical artifact in a trial not powered for fractures, or may indicate something about testosterone's effects on bone quality vs. density. This warrants attention but shouldn't be over-interpreted from a single trial powered for cardiovascular events, not fractures.
The Mendelian Randomization Complication
A 2026 Mendelian randomization study added an important nuance: genetically higher lifelong endogenous testosterone was associated with approximately 17% increased coronary artery disease risk, likely mediated through elevated blood pressure. The FDA responded by adding a new blood pressure warning to testosterone product labels.
The editorial accompanying the study clarified the distinction: this is an association between lifelong genetically-elevated endogenous testosterone and coronary risk. It does not necessarily translate to carefully monitored exogenous therapy in hypogonadal men, which is a different pharmacological context — controlled doses, regular monitoring, specific target ranges.
But it does suggest that "higher is better" is wrong. The optimal testosterone level for longevity is not the maximum achievable level — it's a physiologic range that delivers the metabolic, musculoskeletal, and cognitive benefits without the cardiovascular costs of chronic supraphysiologic exposure.
The Optimal Range: What to Target
| Marker | Lab "Normal" Range | Longevity Optimal | Notes |
|---|---|---|---|
| Total Testosterone | 264–916 ng/dL | 550–700 ng/dL | Mid-physiologic range. Symptom resolution + mortality benefit without supraphysiologic risk |
| Free Testosterone | 5–21 pg/mL (age-dependent) | 15–20 pg/mL | The bioactive fraction. More clinically relevant than total T in many cases |
| SHBG | 10–57 nmol/L | 20–40 nmol/L | Binds testosterone, reducing free T. Too high or too low is problematic |
| Estradiol (E2) | 8–35 pg/mL | 20–30 pg/mL | Aromatized from testosterone. Essential for bone health and brain function — don't crash it |
| LH | 1.7–8.6 mIU/mL | Context-dependent | Elevated LH with low T = primary hypogonadism. Normal LH with low T = secondary |
| FSH | 1.5–12.4 mIU/mL | Context-dependent | Helps distinguish primary vs secondary hypogonadism |
| Prolactin | 4–15 ng/mL | <15 ng/mL | Elevated prolactin suppresses GnRH → low T. Rule out pituitary issues |
Below ~350 ng/dL, most men develop symptoms (fatigue, loss of libido, declining muscle mass, mood changes). Below ~213 ng/dL, mortality risk rises significantly. Above ~900 ng/dL, you're approaching supraphysiologic territory where the Mendelian randomization signal suggests increased cardiovascular risk. The 550–700 range represents the zone where symptoms resolve, metabolic markers improve, lean mass is maintained, and you're not pushing into the territory where lifelong elevation shows harm. It's not a minimum threshold to barely clear — it's the operating range where the system runs best.
Free Testosterone: The Number That Actually Matters
Total testosterone is the headline number, but free testosterone is the fraction doing the work. Only 1–3% of circulating testosterone is "free" — unbound to SHBG or albumin and available to enter cells. The rest is bound and functionally inactive.
SHBG increases with age, thyroid hormone, and liver disease, and decreases with obesity, insulin resistance, and anabolic steroid use. A man with total testosterone of 600 ng/dL and high SHBG may have less bioavailable testosterone than a man at 450 ng/dL with normal SHBG.
This is why total T alone can mislead. A man presents with classic low-T symptoms, total T comes back at 500 ng/dL ("normal"), and the conversation ends. But if SHBG is 65 nmol/L, his free T may be well below optimal. Always test free testosterone alongside total — calculated free T from total T + SHBG + albumin is acceptable, equilibrium dialysis is the gold standard.
Why Testosterone Declines (and What Accelerates It)
The 1–2% annual decline after 30 is an average, not a destiny. Some men maintain robust testosterone into their 70s. Others are clinically hypogonadal by 45. The difference is largely driven by modifiable factors:
| Factor | Effect on Testosterone | Magnitude |
|---|---|---|
| Excess body fat (visceral) | Aromatase converts T → estradiol; adipose tissue = more aromatase | Each 1-point BMI increase ~2% lower T |
| Sleep deprivation (<6 hours) | Suppresses pulsatile LH release → reduced T production | 10–15% lower T vs 7–9 hour sleepers |
| Insulin resistance | Impairs Leydig cell function; bidirectional with low T | HOMA-IR inversely correlated with total T |
| Chronic stress / elevated cortisol | HPA axis suppresses HPG axis directly | Variable — chronic exposure most impactful |
| Alcohol (>2 drinks/day chronic) | Direct testicular toxicity + increased SHBG | Dose-dependent suppression |
| Zinc deficiency | Required for testosterone synthesis | Up to 75% reduction when severely deficient |
| Opioid use (chronic) | Potent GnRH suppressant | Clinically significant hypogonadism common |
| Sedentary lifestyle | Reduced androgen receptor sensitivity + less acute T release | Variable but consistent |
The uncomfortable truth: most age-related testosterone decline is not "aging" — it's the metabolic consequences of how men age. Gaining visceral fat, sleeping poorly, becoming sedentary, and developing insulin resistance are the primary drivers. Address these before assuming your testosterone decline requires pharmaceutical intervention.
Natural Optimization: First-Line Protocol
Before TRT, these interventions produce measurable testosterone increases in men who aren't severely hypogonadal. Combined, they can deliver 20–40% improvement over 3–6 months in men with suboptimal lifestyle factors:
Tier 1: Non-Negotiable Foundation
- Resistance training 3–4x/week: Compound lifts (squat, deadlift, bench, overhead press) at challenging loads. Resistance training increases androgen receptor density and produces acute T elevations. Evidence suggests 15–20% increase over 8–12 weeks of consistent progressive training
- Sleep 7–9 hours consistently: Testosterone is produced primarily during sleep, with peak secretion during REM cycles. Men sleeping <6 hours show 10–15% lower testosterone. Sleep optimization may be the single highest-ROI intervention
- Reduce body fat to 12–18%: Visceral fat drives aromatization. Every percentage point of body fat reduced improves the T:E2 ratio. Weight loss of 5–10% of body weight is associated with meaningful testosterone recovery
- Manage blood glucose: Insulin resistance directly impairs Leydig cell function. Keeping fasting glucose <85 mg/dL and HOMA-IR <1.0 supports endogenous T production
Tier 2: Targeted Supplementation
- Vitamin D: 2,000–4,000 IU daily if serum 25(OH)D is below 40 ng/mL. Vitamin D is technically a hormone precursor, and deficiency is associated with lower testosterone. Supplementation in deficient men shows modest T improvement — it will not raise T meaningfully if you're already replete
- Zinc: 15–30mg daily (zinc picolinate or citrate). Zinc is required for testosterone synthesis at the enzymatic level. Supplementation significantly raises T in deficient men. Most men eating adequate meat and shellfish are not deficient; supplementation in the already-replete does nothing
- Magnesium: 400–600mg daily (glycinate or threonate for bioavailability). Magnesium supports over 300 enzymatic reactions including testosterone-related pathways. Deficiency is common — RBC magnesium testing is more accurate than serum
- Boron: 6–10mg daily. Small studies show modest free testosterone increase through SHBG reduction. Evidence base is thin but the risk profile is essentially zero at these doses
- Ashwagandha (KSM-66): 600mg daily. Multiple RCTs show modest testosterone increases (12–15%) in stressed/overweight men. Likely works through cortisol reduction rather than direct T stimulation. Well tolerated but effects are modest
TRT: When Natural Optimization Isn't Enough
Diagnostic Criteria
TRT is appropriate when two conditions are met simultaneously:
- Symptoms: Fatigue, reduced libido, loss of muscle mass/strength, mood changes, erectile dysfunction, cognitive fog, depression, decreased bone density
- Labs: Two morning total testosterone values below 300 ng/dL (or below 350 ng/dL with significant symptoms and low free T), with LH and FSH to distinguish primary from secondary hypogonadism, and prolactin to rule out pituitary pathology
A single low reading is not enough — testosterone fluctuates daily and is suppressed by acute illness, poor sleep the night before, and recent exercise. Two confirmed low values on separate occasions, drawn between 7–10 AM (peak diurnal secretion), establish the diagnosis.
Rule out reversible causes first: sleep apnea (treat it and testosterone often normalizes), obesity (lose the weight first), opioid use (taper if possible), medications that suppress T (some antidepressants, ketoconazole, spironolactone), chronic illness. If your testosterone is 280 ng/dL because you sleep 5 hours and weigh 260 lbs, TRT is treating the downstream effect while leaving the cause intact. Fix the upstream problem first — you may not need TRT at all.
TRT Protocol Options
| Protocol | Dose | Frequency | Pros | Cons |
|---|---|---|---|---|
| Testosterone cypionate injection | 100–200mg/week | Weekly or split 2x/week | Precise dosing, cost-effective, reliable levels | Injections, potential peaks/troughs if weekly |
| Testosterone enanthate injection | 100–200mg/week | Weekly or split 2x/week | Near-identical to cypionate pharmacokinetics | Same as cypionate |
| Transdermal gel (AndroGel, Testim) | 50–100mg daily | Daily application | Steady levels, no injections, proven in TRAVERSE | Transfer risk to partners/children, daily application, cost |
| Testosterone pellets (Testopel) | 450–900mg | Every 3–6 months | Set-and-forget convenience | Minor procedure, difficult to adjust dose, pellet extrusion possible |
| hCG monotherapy | 1,500–3,000 IU | 2–3x/week | Preserves fertility and testicular size | Variable response, may not achieve target levels in all men |
Exogenous testosterone shuts down the HPG axis: LH and FSH drop to near zero, spermatogenesis ceases, and testes atrophy. This is functionally reversible in most men (recovery takes 3–12 months after cessation), but fertility is essentially zero while on TRT. If you plan to conceive within 1–2 years, do not start testosterone. hCG monotherapy or hCG + low-dose clomiphene can stimulate endogenous production while maintaining fertility, though efficacy varies and these are off-label applications.
Monitoring on TRT
| Marker | Frequency | Target | Why |
|---|---|---|---|
| Total testosterone | 6–8 weeks after starting, then every 3–6 months | 550–700 ng/dL (trough) | Confirm therapeutic levels; draw at trough (day before next injection) |
| Free testosterone | Same as total T | 15–20 pg/mL | Bioactive fraction — may need SHBG context |
| Estradiol (sensitive assay) | Every 3–6 months | 20–30 pg/mL | Excess aromatization → gynecomastia, water retention, mood issues |
| Hematocrit / hemoglobin | Every 3–6 months | Hematocrit <52% | Testosterone stimulates erythropoiesis. Hematocrit >54% = donate blood or reduce dose |
| PSA | Baseline, then every 6–12 months | Stable; <4.0 ng/mL | Rapid PSA rise warrants urological evaluation. TRAVERSE showed no prostate cancer increase |
| Blood pressure | Every visit | <130/80 mmHg | New FDA warning — T can modestly raise BP. Monitor and treat if elevated |
| Lipid panel + ApoB | Every 6 months | See ApoB guide | Monitor cardiovascular risk profile on therapy |
| Metabolic panel (liver/kidney) | Every 6 months | Within normal limits | TRAVERSE showed slightly elevated AKI risk |
Estradiol: The Most Mismanaged Hormone in TRT
Testosterone aromatizes to estradiol (E2) via the aromatase enzyme, primarily in adipose tissue. This is not a problem — it's a feature. Estradiol is essential for bone mineral density, cardiovascular protection, brain function, libido, and joint health in men. The male body is designed to produce estradiol from testosterone.
The problem is when the longevity and TRT communities reflexively suppress aromatization with AI drugs (anastrozole, exemestane) to keep E2 artificially low. Crashing estradiol causes joint pain, bone loss, mood disturbances, cognitive impairment, and paradoxically worsens lipid profiles. It's one of the most common and most harmful mistakes in male hormone optimization.
Estradiol Management
- Target E2: 20–30 pg/mL (sensitive LC-MS/MS assay, not the standard immunoassay which is inaccurate in men)
- If E2 > 40 pg/mL with symptoms (gynecomastia, significant water retention, mood issues): first reduce testosterone dose or increase injection frequency (splitting the weekly dose into 2–3 injections reduces peak T and therefore peak aromatization). AI drugs are a last resort, not a first-line tool
- If E2 < 15 pg/mL: This is dangerously low. Your joints, bones, and brain are paying the price. Reduce or eliminate any AI. Consider that your testosterone dose may need to increase to produce adequate aromatization substrate
- Body composition matters most: Reducing visceral body fat reduces aromatase activity more sustainably than any drug. A lean man at 12–15% body fat rarely has aromatization problems on physiologic TRT doses
What Women Should Know
Testosterone matters for women too — it's the most abundant biologically active sex steroid in women at all stages of life, not estrogen. It declines ~50% between the 20s and 40s.
The 2019 Endocrine Society consensus identified hypoactive sexual desire disorder (HSDD) in postmenopausal women as the only evidence-based indication for female testosterone therapy, showing moderate improvements in desire and satisfying sexual events. No FDA-approved female testosterone formulation currently exists in the US, though international products and compounded formulations are available.
Female testosterone optimization is a different pharmacological world — doses are 10–20x lower than male doses, monitoring parameters differ, and the evidence base is far thinner. This article focuses on male optimization. We'll cover female hormone optimization separately.
The Decision Framework
Step 1: Test Properly
- Morning blood draw (7–10 AM) after adequate sleep
- Total testosterone, free testosterone (calculated or equilibrium dialysis), SHBG, estradiol (sensitive), LH, FSH, prolactin
- Add: CBC, metabolic panel, lipids + ApoB, thyroid (TSH + free T4), fasting insulin
- Two confirmatory draws on separate days if total T is below 350 ng/dL
Step 2: Optimize Before Medicating
- Fix sleep (7–9 hours, consistent schedule, treat sleep apnea if present)
- Reduce body fat below 18% (below 15% ideal)
- Lift heavy 3–4x/week
- Address insulin resistance (fasting glucose, HOMA-IR)
- Eliminate excessive alcohol
- Correct micronutrient deficiencies (vitamin D, zinc, magnesium)
- Retest in 3 months
Step 3: Consider TRT If
- Total T remains <300 ng/dL after 3 months of lifestyle optimization
- Or total T 300–350 ng/dL with low free T and persistent symptoms
- Reversible causes ruled out (sleep apnea, medications, pituitary pathology)
- Fertility considerations addressed
- Target: 550–700 ng/dL total T at trough, not "as high as possible"
TRT is not a lifestyle shortcut. It's a medical intervention for confirmed deficiency after lifestyle optimization has been given a genuine chance. If you're sleeping 5 hours, carrying 30% body fat, and sitting all day, testosterone is not your problem — everything upstream is.
References
- Lincoff AM, et al. "Cardiovascular Safety of Testosterone-Replacement Therapy." New England Journal of Medicine, 2023. (TRAVERSE Trial)
- Araujo AB, et al. "Endogenous testosterone and mortality in men: a systematic review and meta-analysis." Journal of Clinical Endocrinology & Metabolism, 2011.
- 2026 Mendelian randomization study: genetically higher lifelong testosterone and ~17% increased CAD risk. Andrology, 2026.
- FDA label update: removal of cardiovascular boxed warning from testosterone products, February 2025. Addition of blood pressure warning.
- 9,000-man retrospective cohort: real-world testosterone therapy outcomes and safety. World Journal of Men's Health, 2026.
- Bhasin S, et al. "Testosterone Therapy in Men With Hypogonadism: An Endocrine Society Clinical Practice Guideline." Journal of Clinical Endocrinology & Metabolism, 2018.
- Davis SR, et al. "Global Consensus Position Statement on the Use of Testosterone Therapy for Women." Journal of Clinical Endocrinology & Metabolism, 2019.