Testosterone and cardiometabolic risk: what the evidence supports and what it does not license us to claim
The association between low testosterone and metabolic syndrome is robust, but association is not causation. A sober reading of the randomized trials and current guidelines, with particular attention to TRAVERSE.
Low testosterone and metabolic syndrome: a robust and reproducible association
The association exists and is consistent. What it does not tell us, on its own, is which variable moves the other.
Few relationships in metabolic endocrinology are as reproducible as the one linking low testosterone levels to markers of cardiometabolic risk in men. A meta-analysis of observational studies showed that men with metabolic syndrome consistently have lower total and free testosterone than controls without the syndrome (Corona, 2011). The finding repeats when the outcome is type 2 diabetes: a classic meta-analysis of endogenous sex hormones observed that higher testosterone is associated with a lower risk of diabetes in men (Ding, 2006).
It is tempting to read these data as a direct therapeutic invitation — if low testosterone accompanies metabolic disease, raising it should protect. That inference, however, ignores the architecture of the data. This is cross-sectional and prospective evidence of an associative nature, subject to reverse causation and to powerful confounders. Visceral adiposity and insulin resistance themselves lower testosterone, through suppression of the hypothalamic-pituitary-gonadal axis and greater aromatase activity in adipose tissue. The causal arrow, therefore, can point in both directions — and probably does.
The same work by Ding, 2006 brings a datum that should already make us cautious about generalizations: in women, the direction of the association with diabetes reverses. A hormone-disease relationship that changes sign according to sex is not the portrait of a simple, universal mechanism — it is the first sign that we are facing sex-specific biology, not a linear marker of health.
Association is not causation: what genetics and physiology impose on interpretation
Mendelian randomization and threshold analysis are the filters that separate correlation from mechanism.
When the randomization of a trial is not available, genetics offers a partial substitute. Mendelian randomization uses genetic variants associated with lifelong testosterone as a kind of natural allocation, less vulnerable to reverse causation. Applied to testosterone, it revealed sexually dimorphic effects: genetically higher testosterone reduces the risk of type 2 diabetes in men, but increases it in women (Ruth, 2020). This result supports some sex-specific causality in the testosterone-metabolism axis, and at the same time warns, forcefully, against extrapolating pooled observational associations to clinical management.
The shape of the relationship also matters. An individual participant data meta-analysis, bringing together multiple cohorts, indicated that the excess of all-cause and cardiovascular mortality concentrates in men with very low total testosterone — below approximately 7.4 nmol/L (about 213 ng/dL) —, and not along a continuous gradient (Yeap, 2024). The pattern is one of a threshold, not of a linear dose-response. This invalidates the "the higher, the better" logic and reallocates clinical attention to genuine deficiency, not to the percentile.
The synthesis of these two blocks is the interpretive axis of this entire review: the association between low testosterone and cardiometabolic risk is real, bidirectional and sex-specific. None of that amounts to demonstrating that administering testosterone to a man prevents diabetes, cardiovascular disease or death. That is a different question, and only randomized clinical trials can answer it.
| Competing explanation | Mechanism | Implication |
|---|---|---|
| Reverse causation | Visceral adiposity and insulin resistance suppress the gonadal axis and increase aromatization | Low testosterone may be a consequence, not a cause, of metabolic disease |
| Confounders | Age, comorbidities, sleep, medications and inflammation affect both sides | Observational associations overestimate the effect attributable to the hormone |
| Threshold effect | Excess risk concentrated in very low testosterone (< ~213 ng/dL) | No linear gradient; raising already-normal levels has no rationale for benefit |
| Sexual dimorphism | The direction of the genetic effect on diabetes reverses between men and women | Forbids generalizing pooled data or from one sex to the other |
Cardiovascular safety of replacement therapy: reading TRAVERSE for what it measures
Non-inferiority is a statement about safety, not about benefit. The distinction is the whole point.
For nearly a decade, the dominant concern was whether testosterone-replacement therapy (TRT) would increase cardiovascular events. The TRAVERSE trial was designed precisely to test safety. It randomized 5,246 men aged 45 to 80 with symptomatic hypogonadism — total testosterone below 300 ng/dL confirmed on two measurements — and preexisting cardiovascular disease or high risk, to transdermal testosterone gel or placebo, with a mean follow-up of approximately 33 months (Lincoff, 2023).
The primary result was non-inferiority for major adverse cardiovascular events (MACE: cardiovascular death, non-fatal myocardial infarction and non-fatal stroke). In practical terms, TRT did not detectably increase the risk of MACE in this high cardiovascular-risk population over the period studied. It is a reassuring datum — and it is important to name exactly what it authorizes: relative cardiovascular safety, within that context and time horizon. TRAVERSE did not demonstrate that testosterone improves cardiovascular outcomes, did not study eugonadal men and does not validate "anti-aging", vitality or performance use.
Reading the trial honestly also requires disclosing its warning signals. TRAVERSE itself reported a higher incidence, in the testosterone group, of atrial fibrillation, pulmonary embolism and acute kidney injury (Lincoff, 2023). These findings do not nullify the message of non-inferiority in MACE, but together with it they make up the complete picture that any risk-benefit discussion must include.
| Dimension | Finding | Correct reading |
|---|---|---|
| Primary outcome (MACE) | Non-inferiority of testosterone vs placebo at ~33 months | Signal of cardiovascular safety, not of benefit |
| Safety signals | More atrial fibrillation, pulmonary embolism and acute kidney injury in the testosterone group | Risks to disclose alongside the non-inferiority |
| Population | Men aged 45-80, symptomatic hypogonadism (T < 300 ng/dL, two measurements), high CV risk | Not extrapolable to eugonadal men or off-label use |
| Unsupported claims | Cardioprotection, longevity, performance | None was tested or demonstrated |
Testosterone and diabetes prevention: T4DM and the brake of the TRAVERSE substudy
Two randomized trials, divergent results, one prudent conclusion: no general metabolic indication.
The hypothesis that testosterone could prevent diabetes gained its best argument in the T4DM trial. In 1,007 men with abdominal obesity (waist circumference ≥ 95 cm) and impaired glucose tolerance or newly diagnosed diabetes, injectable testosterone undecanoate added to a structured lifestyle program reduced the proportion of men with diabetes on a glucose tolerance test at two years, compared with placebo — approximately 12% versus 21% (Wittert, 2021). It is randomized-trial evidence and deserves noting.
It equally deserves context. T4DM used injectable testosterone in a regimen that produces supraphysiological levels, within an intensive lifestyle program, in rigorously selected men. It is not the portrait of routine TRT, and its authors were careful not to present it as a general clinical indication for diabetes prevention.
The prudence was confirmed by the counterweight. The prespecified prediabetes substudy of TRAVERSE did not reproduce the finding: testosterone replacement did not significantly reduce the progression from prediabetes to diabetes nor improve glycemic control versus placebo (Bhasin, 2024). Faced with two randomized trials with divergent results — in distinct populations, formulations and co-interventions —, the defensible reading is not to pick the more encouraging one. It is to recognize that the metabolic benefit is not generalizable and that preventing diabetes is not, today, a clinical indication for prescribing testosterone.
Diagnostic rigor: when hypogonadism justifies treatment
The most common error is not one of therapy, it is one of diagnosis: treating those who never met the criteria.
The entire preceding discussion converges on a practical point: the indication for TRT is anchored in the diagnosis of hypogonadism, and this has demanding criteria. The reference guideline of the Endocrine Society recommends diagnosing hypogonadism only in the presence of compatible symptoms combined with unequivocally low morning total testosterone, confirmed on a repeat measurement — and not treating on the basis of age or nonspecific symptoms alone (Bhasin, 2018).
The rigor of measurement is not a formality. Testosterone has diurnal variation and is sensitive to acute illness, fasting and sleep; a single afternoon measurement, without repetition and without complementary assessment of LH and SHBG, produces overdiagnosis and leads to treating men who would not benefit. It is precisely the population of genuine deficiency — and not the man with testosterone at the lower limit of normal and nonspecific complaints — that the safety trials and the guidelines have in mind.
No randomized trial has demonstrated a reduction in mortality with TRT, and the guidelines restrict the indication to diagnosed hypogonadism (Bhasin, 2018). Positioning testosterone replacement as a therapy for longevity, cardioprotection or performance exceeds the available evidence and contradicts the standard of care.
| Element | Recommended standard | Pitfall to avoid |
|---|---|---|
| Symptoms | Presence of compatible manifestations, not merely nonspecific ones | Treating for age or a vague complaint in isolation |
| Hormone measurement | Morning total testosterone, fasting, confirmed on a new sample | A single afternoon sample, without repetition |
| Complementary assessment | LH and SHBG to characterize the origin and interpret the value | Interpreting total testosterone in isolation |
| Indication | Restricted to confirmed hypogonadism | Anti-aging, vitality or performance use |
Clinical synthesis and framing of the authorial perspectives
Safety in a selected population is not synonymous with benefit in any patient.
Gathering the threads: the association between low testosterone and cardiometabolic risk is solid (Corona, 2011; Ding, 2006), but bidirectional and sex-specific, as genetics makes clear (Ruth, 2020). The excess risk concentrates in frank deficiency, in a threshold pattern (Yeap, 2024). On the terrain of trials, TRT proved non-inferior to placebo in MACE in high-risk hypogonadal men, with safety signals to consider (Lincoff, 2023), and the metabolic benefit remains uncertain and not generalizable (Wittert, 2021; Bhasin, 2024).
The management that this evidence supports is sober: investigate and treat genuine hypogonadism according to rigorous criteria (Bhasin, 2018), follow cardiometabolic risk with the interventions of established efficacy — weight loss, physical activity, glycemic and lipid control — and not promise the patient outcomes that no trial has demonstrated.
A note of editorial transparency. The author maintains his own reviews on this topic, open-access and non-indexed. They are mentioned here as authorial production and perspective, and were deliberately not used as an evidentiary basis for this review. The evidence presented rests exclusively on the high-impact indexed sources cited throughout the text — a condition for avoiding the conflict of self-citation and preserving the hierarchy of evidence.
Why this matters for your care
This content has a strictly educational and scientific-updating purpose, in accordance with CFM norms, and does not replace individualized medical evaluation — no information here constitutes a prescription, a promise of results or a treatment indication. The decision to investigate hypogonadism or to start testosterone replacement depends on rigorous diagnostic criteria and on each person's clinical context. To organize your symptoms and history before talking with your doctor, you can use the Functional Self-Assessment, and to go deeper into other evidence-based metabolic medicine topics, consult the Library.
References
- Corona G, Monami M, Rastrelli G, Aversa A, Sforza A, Lenzi A, Forti G, Mannucci E, Maggi M. Testosterone and Metabolic Syndrome: A Meta-Analysis Study. The Journal of Sexual Medicine 8(1):272-283. 2011. doi:10.1111/j.1743-6109.2010.01991.x
- Ding EL, Song Y, Malik VS, Liu S. Sex Differences of Endogenous Sex Hormones and Risk of Type 2 Diabetes: A Systematic Review and Meta-analysis. JAMA 295(11):1288-1299. 2006. doi:10.1001/jama.295.11.1288
- Ruth KS, Day FR, Tyrrell J, Thompson DJ, Wood AR, Mahajan A, Beaumont RN, Wittemans L, Martin S, Busch AS, et al. Using human genetics to understand the disease impacts of testosterone in men and women. Nature Medicine 26(2):252-258. 2020. doi:10.1038/s41591-020-0751-5
- Yeap BB, Marriott RJ, Dwivedi G, Adams RJ, Antonio L, Ballantyne CM, Bauer DC, Bhasin S, Biggs ML, Cawthon PM, et al. Associations of Testosterone and Related Hormones With All-Cause and Cardiovascular Mortality and Incident Cardiovascular Disease in Men: Individual Participant Data Meta-analyses. Annals of Internal Medicine 177(6):768-781. 2024. doi:10.7326/M23-2781
- Lincoff AM, Bhasin S, Flevaris P, Mitchell LM, Basaria S, Boden WE, Cunningham GR, Granger CB, Khera M, Thompson IM, Wang Q, Wolski K, Davey D, Kalahasti V, Khan N, Miller MG, Snabes MC, Chan A, Dubcenco E, Li X, Yi T, Huang B, Pencina KM, Travison TG, Nissen SE (TRAVERSE Study Investigators). Cardiovascular Safety of Testosterone-Replacement Therapy. New England Journal of Medicine 389(2):107-117. 2023. doi:10.1056/NEJMoa2215025
- Wittert G, Bracsh T, Gebski V, Yeap BB, Allan CA, Armstrong S, Conway AJ, Daniel M, Handelsman DJ, Inder WJ, Jenkins AJ, McLachlan RI, Robledo KP, Zajac JD, Grossmann M (T4DM Study Investigators). Testosterone treatment to prevent or revert type 2 diabetes in men enrolled in a lifestyle programme (T4DM): a randomised, double-blind, placebo-controlled, 2-year, phase 3b trial. The Lancet Diabetes & Endocrinology 9(1):32-45. 2021. doi:10.1016/S2213-8587(20)30367-3
- Bhasin S, Lincoff AM, Nissen SE, Wannemuehler K, McDonnell ME, Peters AL, Khan N, Snabes MC, Li X, Li G, Buhr K, Pencina KM, Travison TG (TRAVERSE Study Investigators). Effect of Testosterone on Progression From Prediabetes to Diabetes in Men With Hypogonadism: A Substudy of the TRAVERSE Randomized Clinical Trial. JAMA Internal Medicine 184(4):353-362. 2024. doi:10.1001/jamainternmed.2023.7862
- Bhasin S, Brito JP, Cunningham GR, Hayes FJ, Hodis HN, Matsumoto AM, Snyder PJ, Swerdloff RS, Wu FC, Yialamas MA. Testosterone Therapy in Men With Hypogonadism: An Endocrine Society Clinical Practice Guideline. The Journal of Clinical Endocrinology & Metabolism 103(5):1715-1744. 2018. doi:10.1210/jc.2018-00229
Educational and scientific content. It does not constitute diagnosis, prescription or individual clinical guidance, and does not replace a medical consultation. Management decisions must be individualized by a physician.