{"slug":"testosterone-men","name":"Total testosterone (men)","aliases":["Serum testosterone","Total T"],"kind":"biomarker","system":"metabolomics","unit":"ng/dL","unitNote":"Some laboratories report nmol/L; divide ng/dL by about 28.8 for nmol/L.","answer":"In pooled cohorts of 20,654 men, all-cause mortality was higher only below about 213 ng/dL and cardiovascular mortality only below about 153, with no difference between quintiles above that; sexual symptoms clustered below about 320 in the European Male Aging Study; and no trial has shown that raising a man without symptoms from the middle of the range to the top of it changes any outcome. The harmonized reference range for healthy nonobese men aged 19 to 39 is 264 to 916 ng/dL, and guidelines diagnose deficiency only with consistent symptoms and two low fasting morning values.","summary":"Total testosterone is the whole hormone in the blood, most of it bound to proteins. Laboratories print about 264 to 916 ng/dL, or an older 300 to 1,000. The literature is firm that very low values keep company with symptoms and higher mortality, that weight, sleep, illness and some drug classes move the number, and that a morning draw repeated once is the minimum before any conclusion. Whether a higher value inside the range is healthier than a middle one has not been tested in anyone without symptoms.","whatItIs":[{"text":"Testosterone is made in the testes under a signal from the pituitary (luteinizing hormone), and most of what circulates is bound to sex hormone-binding globulin (SHBG) and albumin. A total testosterone counts all of it. The Endocrine Society's position statement on measurement notes that laboratory proficiency should rest on measuring known concentrations accurately rather than on agreement among laboratories using the same method, and its 2018 guideline recommends a fasting morning total testosterone, measured with an accurate and reliable assay, as the initial test, with a free testosterone by equilibrium dialysis or a validated formula when the total is near the lower limit or a condition alters SHBG.","cites":["rosner2007","bhasin2018"],"grade":"A"},{"text":"Levels fall with age, slowly. In 890 men of the Baltimore Longitudinal Study of Aging, total testosterone declined by about 0.124 nmol/L a year, about 3.6 ng/dL, independent of health factors, and the share with values in the hypogonadal range by total-testosterone criteria reached about 20 percent of men over 60, 30 percent over 70 and 50 percent over 80. In the European Male Aging Study, 2,395 men aged 40 to 79 lost about 0.1 nmol/L a year, about 3 ng/dL, over 4.4 years while SHBG rose.","cites":["harman2001","camacho2013"],"grade":"B"},{"text":"A low number is common; the syndrome is not. Among 2,966 European men aged 40 to 79, 63 (2.1 percent) met the criteria for late-onset hypogonadism, meaning three sexual symptoms together with a low total and free testosterone. They were older and more obese, with lower muscle mass, heel bone density and hemoglobin in proportion to the degree of deficiency, and the end-organ findings concentrated in men with a total testosterone below about 8 nmol/L, about 230 ng/dL. Men with a low testosterone but no symptoms showed smaller associations with the same end points.","cites":["tajar2012"],"grade":"B"}],"origins":[{"text":"Testosterone's receptor is younger than the family it belongs to, and the family began with estrogen. Phylogenetic reconstruction of steroid receptors in basal vertebrates, with resurrection of the inferred ancestral proteins, indicates that the first steroid receptor was an estrogen receptor, followed by a progesterone receptor, and that the full set of mammalian receptors arose from those by two rounds of large-scale genome duplication, one before and one after the origin of jawed vertebrates; specific regulation by androgens is, on this account, a relatively recent innovation. The ancestral receptor is older than vertebrates: an estrogen receptor ortholog was isolated from the sea slug Aplysia, and the reconstructed ancestor of all steroid receptors had estrogen receptor-like function, placing its origin before the split of bilaterally symmetric animals, with later loss in the lines leading to arthropods and nematodes. A 2017 study reconstructed the ancestral ligand as well: an aromatized steroid that kept its side chain, which the authors named paraestrol A and which, once synthesized, bound and activated the resurrected ancestral receptor.","cites":["thornton2001","thornton2003","markov2017"],"grade":"C"},{"text":"The explanation offered for that order is called ligand exploitation: in a biosynthetic pathway the terminal product is the first to get a receptor, selection for that hormone also selects for synthesis of the intermediates, and when the receptor gene duplicates the copies evolve affinity for molecules already present. Testosterone is such an intermediate, one aromatase step short of estradiol, which is the argument for how an androgen receptor could arise later and find its hormone waiting. The argument has complications and critics. A comparative survey of complete animal genomes concluded that the enzyme machinery of steroid synthesis was elaborated independently in the three main bilaterian lineages and that the steroid-making cytochrome P450 enzymes descend from enzymes that detoxify foreign chemicals; another phylogenetic analysis places the first appearance of CYP11A, which cleaves the cholesterol side chain, and CYP19, the aromatase, in the common ancestor of amphioxus and vertebrates, alongside the first receptors closely resembling ours. The receptor story and the enzyme story are still being fitted together.","cites":["thornton2001","markov2009","baker2011"],"grade":"C"},{"text":"In the network, testosterone is a cholesterol derivative on the road to estradiol. CYP11A opens the pathway by cleaving the cholesterol side chain and CYP19 closes it by aromatizing androgens to estrogens, so a man's estradiol is made from his testosterone. In blood most of it is bound to SHBG and albumin, so a total testosterone is partly a report on the liver's SHBG output, which weight gain lowers and weight loss raises in step with the hormone itself; that is why the shelf reads Total testosterone (men) beside SHBG, Fasting insulin and HOMA-IR, and why the same total can mean different things in two men.","cites":["baker2011","markov2017","camacho2013"],"grade":"C"}],"labRange":{"typical":"About 264 to 916 ng/dL on assays calibrated to the CDC standard; many laboratories still print about 300 to 1,000, and some print an age-adjusted range","claims":[{"text":"The harmonized range comes from 9,054 community-dwelling men in four cohorts in the United States and Europe (Framingham, the European Male Aging Study, Osteoporotic Fractures in Men and the Male Sibling Study of Osteoporosis), whose assays were cross-calibrated to a CDC reference method. In healthy nonobese men aged 19 to 39, the 2.5th, 5th, 50th, 95th and 97.5th percentiles were 264, 303, 531, 852 and 916 ng/dL. The authors found that a substantial share of the differences between cohorts had been assay differences, which is why an older laboratory range of 300 to 1,000 and a newer one of 264 to 916 can describe the same men.","cites":["travison2017"],"grade":"B"},{"text":"The two US guidelines read the lower limit differently. The Endocrine Society's 2018 guideline uses the harmonized lower limit of 264 ng/dL (9.2 nmol/L), recommends measuring total testosterone on two separate fasting mornings, and recommends against routine screening of men in the general population. The American Urological Association's 2018 guideline names a total testosterone below 300 ng/dL as a reasonable cutoff, also on two early-morning measurements, and makes the diagnosis only when a low level is combined with symptoms or signs. Neither range is a treatment threshold on its own.","cites":["bhasin2018","mulhall2018"],"grade":"A"},{"text":"The reference range is a distribution of young men, not a map of risk. In the pooled mortality cohorts the thresholds that mattered sat well below it, at about 213 and 153 ng/dL, and the authors wrote that a threshold adds information beyond ranges based on the distributions of selected healthy men. The lower half of the printed range is where a measured total testosterone most often misleads, because SHBG rises with age and a normal total can hide a low free fraction.","cites":["yeap2024","antonio2016"],"grade":"B"}]},"evidence":{"headline":"Risk sat at the bottom. In 20,654 men from nine cohorts measured by mass spectrometry, all-cause mortality was higher only below about 213 ng/dL and cardiovascular death only below about 153, with no difference between quintiles above those lines; sexual symptoms clustered below about 320 in 3,369 European men; the one large treatment trial of men with symptoms and two values under 300 found cardiac events no more common than placebo but more atrial fibrillation, pulmonary embolism and kidney injury; and no trial has treated men without symptoms to a higher number.","scale":{"min":0,"max":1200,"ticks":[0,200,400,600,800,1000,1200]},"bands":[{"label":"Harmonized reference range (healthy nonobese men aged 19 to 39); many laboratories print about 300 to 1,000","low":264,"high":916,"tone":"lab","source":"Travison 2017; typical US laboratories"},{"label":"Higher cardiovascular mortality in pooled cohorts","low":0,"high":153,"tone":"caution","source":"Yeap 2024"},{"label":"Higher all-cause mortality in pooled cohorts","low":153,"high":213,"tone":"caution","source":"Yeap 2024"},{"label":"Below the harmonized lower limit (264) and the AUA cutoff (300); a diagnosis still needs symptoms and a second fasting morning draw","low":213,"high":300,"tone":"caution","source":"Travison 2017; Bhasin 2018; Mulhall 2018"},{"label":"Borderline: sexual symptoms rose below about 320 to 370 in EMAS; SHBG and free testosterone decide what a total means here","low":300,"high":350,"tone":"neutral","source":"Wu 2010; Antonio 2016; Bhasin 2018"},{"label":"No mortality difference by quintile above 213 in pooled cohorts; the cohort medians (357 to 588) and the trial treatment targets (350 to 750) sit here","low":350,"high":852,"tone":"evidence","source":"Yeap 2024; Lincoff 2023; Travison 2017"},{"label":"Above the 95th percentile of healthy nonobese young men; usually a timing, assay or exogenous-testosterone question; no outcome benefit shown","low":852,"high":1200,"tone":"neutral","source":"Travison 2017; Yeap 2024"}],"claims":[{"text":"The best mortality data pool individual records from nine prospective cohorts of community-dwelling men whose testosterone was measured by mass spectrometry and who were followed at least five years: 20,654 men, 255,830 person-years, 7,241 deaths. After adjustment for age, weight, smoking, alcohol, activity, blood pressure, diabetes, kidney function and lipids, the hazard ratios overlapped 1.0 at every quintile median; only men with a baseline testosterone below 7.4 nmol/L (213 ng/dL) had higher all-cause mortality, and only those below 5.3 nmol/L (153 ng/dL) had higher cardiovascular mortality. The cohorts' median testosterone ran from 357 to 588 ng/dL. A low level went with dying sooner; a higher level inside the range did not go with living longer.","cites":["yeap2024"],"grade":"A"},{"text":"The same analysis found that a lower SHBG, not a higher testosterone, was associated with lower all-cause mortality (hazard ratio 0.85 for the lowest versus the highest quintile) and lower cardiovascular mortality (0.81), and that dihydrotestosterone had a U-shaped relationship with death, higher at both the lowest and the highest concentrations. These are associations in observational cohorts; the authors list the design and the imputation of missing data as limitations.","cites":["yeap2024"],"grade":"B"},{"text":"An earlier meta-analysis of 12 observational studies (16,184 men for all-cause mortality, 11,831 for cardiovascular mortality, mean age 61, mean testosterone 487 ng/dL, mean follow-up 9.7 years) also found low endogenous testosterone associated with higher all-cause and cardiovascular death, but with so much heterogeneity between studies that the authors declined to give a summary estimate. The association was stronger in studies with older men, lower testosterone, shorter follow-up and blood drawn through the day, which they read as the effect of underlying health status differing between cohorts rather than of the hormone itself.","cites":["araujo2011"],"grade":"B"},{"text":"Symptoms have their own thresholds. Among 3,369 European men aged 40 to 79 with morning samples measured by mass spectrometry, poor morning erections, low sexual desire and erectile dysfunction became more likely as total testosterone fell through about 8.0 to 13.0 nmol/L (about 230 to 370 ng/dL), and only those three sexual symptoms had a syndromic association with the hormone; depression, fatigue and difficulty with vigorous activity related to the level but did not cluster with it. The authors defined late-onset hypogonadism as at least three sexual symptoms with a total testosterone below 11 nmol/L (about 320 ng/dL) and a free testosterone below 220 pmol/L. In the same cohort, 277 men with a normal total but a low calculated free testosterone were older, had higher SHBG and reported more sexual and physical symptoms, while 96 men with a low total but a normal free testosterone were younger, more obese, and lacked the features of deficiency.","cites":["wu2010","antonio2016"],"grade":"B"},{"text":"What treatment does for men who are both low and symptomatic was tested in the Testosterone Trials: 790 men aged 65 and older, screened from 51,085, with symptoms and an average testosterone under 275 ng/dL, randomized to testosterone gel or placebo for one year. Raising the level to the mid-normal range of men aged 19 to 40 produced a moderate improvement in sexual activity (treatment effect 0.58 on a 12-point scale), desire and erectile function, no benefit in vitality on the fatigue scale, and no significant difference in walking distance within the physical-function trial; the effect sizes for mood and depressive symptoms were all below 0.20. Seven men on testosterone and none on placebo reached a hemoglobin of 17.5 g/dL or more. The authors wrote that the trial was too small to judge risk.","cites":["snyder2016"],"grade":"B"},{"text":"Cardiovascular safety was tested in TRAVERSE: 5,246 men aged 45 to 80 with existing or high-risk cardiovascular disease, symptoms of hypogonadism and two fasting values under 300 ng/dL, randomized to a transdermal testosterone gel adjusted to hold the level between 350 and 750 ng/dL or to placebo, followed for a mean of 33 months. A major cardiac event (cardiovascular death, heart attack or stroke) occurred in 7.0 percent on testosterone and 7.3 percent on placebo (hazard ratio 0.96), meeting the trial's noninferiority margin; atrial fibrillation, acute kidney injury and pulmonary embolism were more common on testosterone. The result says treatment of deficient, symptomatic men did not raise the rate of heart attacks and strokes over about three years; it does not say anything about men above 300 or men without symptoms, who were not enrolled.","cites":["lincoff2023"],"grade":"B"},{"text":"Genetics gives the one causal reading. Using 2,571 genetic variants in 425,097 UK Biobank participants, a genetically higher testosterone in men lowered the odds of type 2 diabetes (odds ratio 0.86 per standard deviation) and raised the risk of prostate cancer, while in women it raised type 2 diabetes and polycystic ovary syndrome. The direction of effect differed between the sexes, and the authors note that the effects of testosterone on disease outcomes had otherwise been unknown.","cites":["ruth2020"],"grade":"B"},{"text":"The low level of a heavy, unwell middle-aged man is usually functional, not glandular. The Endocrine Society of Australia's position statement separates pathological hypogonadism, a disease of the pituitary or testes confirmed by hormone assays and warranting treatment at any age, from the lowered testosterone of obesity, metabolic syndrome and type 2 diabetes, which comes without a rise in LH and FSH, and for which, it says, clear evidence of the efficacy and safety of testosterone therapy is lacking. A review of the same men concludes that weight loss, treatment of comorbidities and withdrawal of offending medications bring clinical improvement and a modest rise in testosterone, that definitive randomized trials are lacking, and that testosterone should be considered primarily for men with significant clinical features and unequivocally low levels.","cites":["yeap2016","grossmann2017"],"grade":"A"},{"text":"Population levels have been falling for reasons that are not fully explained. In 1,532 Boston men sampled across 1987 to 2004, testosterone fell with calendar year independently of age, by more than the usual cross-sectional decline with aging, and the fall was not accounted for by smoking or obesity. In 4,045 adolescent and young adult men in US national surveys from 1999 to 2016, mean total testosterone was lower in the 2011 to 2016 cycles than in 1999 to 2000, lower still with higher body mass index, and the trend held in men of normal weight; the authors list changing assays and unmeasured environmental factors as limitations.","cites":["travison2007","lokeshwar2021"],"grade":"C"}]},"levers":[{"name":"Body weight","effect":"weight gain lowers it; weight loss raises it, in proportion to the loss","claim":{"text":"Across 24 studies, a low-calorie diet raised total testosterone by about 2.87 nmol/L (about 83 ng/dL) and bariatric surgery by about 8.73 nmol/L (about 250 ng/dL), with the degree of weight loss the best predictor of the rise and larger gains in younger, nondiabetic men with more obesity; SHBG-bound and unbound fractions both rose, estradiol fell and gonadotropins rose. In 2,395 European men followed 4.4 years, weight loss went with a proportional rise and weight gain with a proportional fall in testosterone and SHBG, and free testosterone moved only in men who changed weight by 15 percent or more. Excess weight is the most common correctable cause of a low total testosterone.","cites":["corona2013","camacho2013"],"grade":"B"}},{"name":"Sleep","effect":"one week of short sleep lowers it 10 to 15 percent","claim":{"text":"Ten healthy men aged about 24 who slept 5 hours a night for one week had daytime testosterone levels 10 to 15 percent lower than after 10-hour nights (16.5 versus 18.4 nmol/L, about 475 versus 530 ng/dL across the waking day, and 15.5 versus 17.9 in the afternoon and evening), with no rise in cortisol and a progressive fall in vigor scores. The authors compare the drop with the 1 to 2 percent a year of normal aging. Most of the day's testosterone is released during sleep, and the sample was small.","cites":["leproult2011"],"grade":"C"}},{"name":"Exercise","effect":"raises it for about 30 minutes after a hard session; training does not raise the resting level","claim":{"text":"Across 48 studies and 569 men, moderate and high-intensity exercise raised total and free testosterone acutely (standardized mean difference 0.74), measured at the end of the session and within 30 minutes but not after, with no change in SHBG; mild activity did not. Across 11 randomized trials of 421 insufficiently active men aged 19 to 75, a median of 12 weeks of aerobic, resistance or combined training had no effect on resting total testosterone (standardized mean difference 0.00), whatever the mode, age or body mass. Training earns its place for other reasons; a resting testosterone is not one of them, except through the weight it helps remove.","cites":["dandrea2020","potter2021"],"grade":"B"}},{"name":"Alcohol","effect":"heavy use lowers it; ordinary intake does not show a consistent effect","claim":{"text":"A meta-analysis of 17 observational studies found no significant overall difference in testosterone between men who drank and men who did not (standardized mean difference 0.03), with the authors reserving their concern for the notable alterations seen in alcohol use disorder. The signal is in heavy drinking, not the occasional glass, and the studies are observational.","cites":["moosazadeh2024"],"grade":"C"}},{"name":"Opioids and glucocorticoids","effect":"both lower it, often reversibly, without the rise in LH and FSH that marks testicular failure","claim":{"text":"Across 17 studies with 2,769 participants (800 opioid users), men using opioids regularly had significantly lower mean testosterone than controls, whatever the opioid, including methadone, with high heterogeneity and low overall quality of evidence. The Endocrine Society of Australia names glucocorticoid and opioid use as causes of possibly reversible reductions in testosterone without elevated LH and FSH, and suggests endocrine review where the drug must continue. Neither should be stopped over a testosterone result; the finding is a conversation with the prescriber.","cites":["bawor2015","yeap2016"],"grade":"B"}}],"caveats":[{"text":"The hour of the draw changes the answer. In 66 Boston men sampled at six visits, testosterone at 4 p.m. was 20 to 25 percent lower than at 8 a.m. in men aged 30 to 40, with the difference narrowing to about 10 percent by age 70; 17 men who had at least one value under 300 ng/dL in the afternoon were normal at all three morning visits. Both guidelines want a fasting early-morning draw, and both want it repeated on a second morning.","cites":["brambilla2009","bhasin2018","mulhall2018"],"grade":"A"},{"text":"One value is not a level. In 132 men sampled in pairs 1 to 3 days apart and again at 3 and 6 months, biological variation outweighed assay variation, and the day-to-day component outweighed the long-term one; for the hormones studied, two measurements on the same man differed by more than 18 to 28 percent about half the time and by more than 27 to 54 percent a quarter of the time. The authors concluded that one sample is generally not enough to characterize a man's level and that more than three is probably not warranted.","cites":["brambilla2007"],"grade":"B"},{"text":"Assays disagree, most at the low end. When 122 pedigreed samples were run on six immunoassays against liquid chromatography-tandem mass spectrometry, one automated platform read about 90 ng/dL low and another about 99 ng/dL high across the range, and at concentrations below 100 ng/dL the immunoassays scattered in both directions; most adult-male results fell within 20 percent of the reference method. The Endocrine Society's 2018 guideline notes the better performance of mass spectrometry, especially in the low range, and says direct analog free-testosterone immunoassays should not be used, because they are inaccurate. A borderline value deserves a mass-spectrometry repeat before it is believed.","cites":["wang2004","rosner2007","bhasin2018"],"grade":"A"},{"text":"A total testosterone can mislead in both directions through SHBG. SHBG rises with age, so an older man can have a normal total and a low free fraction with real symptoms, while a heavy younger man can have a low total from a low SHBG with a normal free fraction and no deficiency at all; in 3,334 European men both patterns were common. A calculated free testosterone with a measured SHBG belongs beside any borderline total. A future entry on SHBG will take this up on its own.","cites":["antonio2016","bhasin2018"],"grade":"B"},{"text":"Illness lowers it, in proportion to severity. In 59 men admitted to critical care, admission testosterone was below healthy controls in every severity group, lowest in the sickest, and the gap widened as the stay went on; the guideline lists acute illness and nutritional deficiency among the things that lower a measurement. A testosterone drawn during or soon after an illness, a crash diet or a period of heavy training describes that week, not the man, and should be repeated when well.","cites":["spratt1993","bhasin2018"],"grade":"B"},{"text":"The number is not the diagnosis. Both guidelines make the diagnosis only in men with symptoms or signs consistent with deficiency and unequivocally, consistently low morning values; and in the European cohort, men with a low testosterone and no symptoms showed smaller end-organ associations than men with both. Nothing on this page applies to a man taking testosterone or anabolic steroids, whose measured level reflects the dose, or to adolescents, whose ranges are their own.","cites":["bhasin2018","mulhall2018","tajar2012"],"grade":"A"}],"seeAPhysician":["A fasting morning total testosterone below about 264 to 300 ng/dL together with low desire, poor morning erections or erectile dysfunction, which is the pattern the guidelines ask to be confirmed on a second morning with SHBG, LH and a free testosterone.","Any value below about 150 to 200 ng/dL, with or without symptoms, since that is where mortality rose in the cohorts and where a pituitary or testicular cause needs to be looked for.","A low testosterone with a high LH and FSH, small or shrinking testes, infertility, a history of testicular injury, chemotherapy or mumps orchitis, which points to the testes themselves.","A low testosterone with a low or normal LH together with headaches, vision changes, loss of body hair, milky nipple discharge, or symptoms of other pituitary hormones, which points upstream.","A value well above the laboratory range in a man not taking testosterone, or any value in a man who is taking testosterone, anabolic steroids or supplements marketed to raise it, where hematocrit, PSA and the reason for the level need review.","Anyone considering testosterone, who should hear, before a prescription, the fertility effect, the hematocrit, prostate and sleep-apnea checks the guidelines require, and the atrial fibrillation, clotting and kidney signals from TRAVERSE."],"openQuestions":["No randomized trial has taken men without symptoms whose testosterone sits inside the reference range and raised it to a higher number to see whether anything improves, so any optimal zone above the symptom thresholds is a judgment, not a tested goal.","The pooled cohorts found no mortality gradient above about 213 ng/dL and the genetic data point in two directions at once (less diabetes, more prostate cancer with genetically higher levels), so whether a higher natural level is protective, neutral or costly over a lifetime is unresolved.","The long-term safety of treatment beyond the roughly three years of TRAVERSE, in men younger than 45, in men above 300 ng/dL, and in men without cardiovascular risk, has not been studied in a trial of adequate size; the atrial fibrillation, pulmonary embolism and kidney signals have not been explained.","Why population and young-adult testosterone levels have fallen across recent decades beyond what weight explains is unknown; assays, environmental exposures, sleep, activity and birth-cohort effects are all candidates and none is established.","Whether free testosterone, SHBG or a combination should replace the total as the first test is argued: one European analysis found free testosterone tracks symptoms better, while the Australian position statement advises against calculated free testosterone for decisions."],"questions":[{"q":"Is a testosterone of 350 ng/dL low?","a":"By the harmonized reference range it is inside normal (264 to 916 for healthy nonobese men aged 19 to 39), and above the AUA's 300 cutoff. It is also near the level where sexual symptoms became more likely in the European Male Aging Study (about 320 to 370), so the answer depends on three things the number does not carry: whether it was a fasting morning draw, whether a second draw agrees, and whether SHBG and free testosterone are low. A 350 at 4 p.m. after a bad week can be a 450 at 8 a.m. after a good one. Without symptoms, no trial has shown that doing anything about a 350 helps."},{"q":"What is an optimal testosterone level for a 45-year-old man?","a":"No study defines one. The median of healthy nonobese young men was 531 ng/dL, the cohorts that followed men for mortality had medians of about 357 to 588, and across that whole span the risk of dying did not differ by quintile; it rose only below about 213. Trials that treated deficient men aimed for 350 to 750 and the mid-normal range. A physician's preferred zone inside the range is a judgment about symptoms, body composition and SHBG, and this page says so."},{"q":"Does low testosterone shorten life?","a":"Men with very low levels die sooner in the cohorts: all-cause mortality rose below about 213 ng/dL and cardiovascular death below about 153 in pooled data on 20,654 men, and an earlier meta-analysis found the same direction with wide differences between studies. Whether the hormone causes that or marks illness, obesity and age is not settled; the association weakened after adjustment, and the treatment trials were not designed to measure survival. No cohort found that higher levels inside the range went with living longer."}],"citations":[{"id":"travison2017","authors":"Travison TG, Vesper HW, Orwoll E, Wu F, Kaufman JM, Wang Y, Lapauw B, Fiers T, Matsumoto AM, Bhasin S","title":"Harmonized Reference Ranges for Circulating Testosterone Levels in Men of Four Cohort Studies in the United States and Europe","journal":"Journal of Clinical Endocrinology and Metabolism","year":2017,"pmid":"28324103","doi":"10.1210/jc.2016-2935","type":"laboratory","finding":"In 9,054 men of four cohorts cross-calibrated to a CDC reference method, healthy nonobese men aged 19 to 39 had 2.5th, 5th, 50th, 95th and 97.5th percentiles of 264, 303, 531, 852 and 916 ng/dL; much of the intercohort variation was assay difference.","url":"https://pubmed.ncbi.nlm.nih.gov/28324103/"},{"id":"bhasin2018","authors":"Bhasin S, Brito JP, Cunningham GR, Hayes FJ, Hodis HN, Matsumoto AM, Snyder PJ, Swerdloff RS, Wu FC, Yialamas MA","title":"Testosterone Therapy in Men With Hypogonadism: An Endocrine Society Clinical Practice Guideline","journal":"Journal of Clinical Endocrinology and Metabolism","year":2018,"pmid":"29562364","doi":"10.1210/jc.2018-00229","type":"guideline","finding":"Diagnose hypogonadism only with consistent symptoms and unequivocally, consistently low fasting morning total testosterone measured by an accurate assay and repeated; harmonized lower limit 264 ng/dL; free testosterone when total is borderline or SHBG is altered; against routine screening of the general population.","url":"https://pubmed.ncbi.nlm.nih.gov/29562364/"},{"id":"mulhall2018","authors":"Mulhall JP, Trost LW, Brannigan RE, Kurtz EG, Redmon JB, Chiles KA, et al.","title":"Evaluation and Management of Testosterone Deficiency: AUA Guideline","journal":"Journal of Urology","year":2018,"pmid":"29601923","doi":"10.1016/j.juro.2018.03.115","type":"guideline","finding":"A total testosterone below 300 ng/dL on two early-morning measurements is a reasonable cutoff; the diagnosis is made only when a low level is combined with symptoms or signs.","url":"https://pubmed.ncbi.nlm.nih.gov/29601923/"},{"id":"wu2010","authors":"Wu FC, Tajar A, Beynon JM, Pye SR, Silman AJ, Finn JD, et al.; EMAS Group","title":"Identification of late-onset hypogonadism in middle-aged and elderly men","journal":"New England Journal of Medicine","year":2010,"pmid":"20554979","doi":"10.1056/NEJMoa0911101","type":"cross-sectional","finding":"In 3,369 European men aged 40 to 79, three sexual symptoms clustered with total testosterone below 11 nmol/L (about 320 ng/dL) and free testosterone below 220 pmol/L; symptom probabilities rose through about 8.0 to 13.0 nmol/L.","url":"https://pubmed.ncbi.nlm.nih.gov/20554979/"},{"id":"tajar2012","authors":"Tajar A, Huhtaniemi IT, O'Neill TW, Finn JD, Pye SR, Lee DM, et al.; EMAS Group","title":"Characteristics of androgen deficiency in late-onset hypogonadism: results from the European Male Aging Study (EMAS)","journal":"Journal of Clinical Endocrinology and Metabolism","year":2012,"pmid":"22419720","doi":"10.1210/jc.2011-2513","type":"cross-sectional","finding":"Among 2,966 men, 63 (2.1 percent) met late-onset hypogonadism criteria, with lower muscle mass, heel bone density and hemoglobin in proportion to deficiency, concentrated below 8 nmol/L; men with low testosterone but no symptoms showed smaller associations.","url":"https://pubmed.ncbi.nlm.nih.gov/22419720/"},{"id":"antonio2016","authors":"Antonio L, Wu FC, O'Neill TW, Pye SR, Ahern TB, Laurent MR, et al.","title":"Low Free Testosterone Is Associated with Hypogonadal Signs and Symptoms in Men with Normal Total Testosterone","journal":"Journal of Clinical Endocrinology and Metabolism","year":2016,"pmid":"26909800","doi":"10.1210/jc.2015-4106","type":"cross-sectional","finding":"In 3,334 men, 277 with normal total but low calculated free testosterone were older, had higher SHBG and more sexual and physical symptoms; 96 with low total but normal free testosterone were younger, more obese and lacked features of deficiency.","url":"https://pubmed.ncbi.nlm.nih.gov/26909800/"},{"id":"yeap2024","authors":"Yeap BB, Marriott RJ, Dwivedi G, Adams RJ, Antonio L, Ballantyne CM, et al.","title":"Associations of Testosterone and Related Hormones With All-Cause and Cardiovascular Mortality and Incident Cardiovascular Disease in Men: Individual Participant Data Meta-analyses","journal":"Annals of Internal Medicine","year":2024,"pmid":"38739921","doi":"10.7326/M23-2781","type":"meta-analysis","finding":"In individual data on 20,654 men from nine mass-spectrometry cohorts (255,830 person-years), all-cause mortality was higher only below 7.4 nmol/L (213 ng/dL) and cardiovascular mortality only below 5.3 nmol/L (153 ng/dL); hazard ratios overlapped 1.0 at every quintile median; lower SHBG went with lower mortality.","url":"https://pubmed.ncbi.nlm.nih.gov/38739921/"},{"id":"araujo2011","authors":"Araujo AB, Dixon JM, Suarez EA, Murad MH, Guey LT, Wittert GA","title":"Clinical review: Endogenous testosterone and mortality in men: a systematic review and meta-analysis","journal":"Journal of Clinical Endocrinology and Metabolism","year":2011,"pmid":"21816776","doi":"10.1210/jc.2011-1137","type":"meta-analysis","finding":"Across 12 observational studies (16,184 men for all-cause and 11,831 for cardiovascular mortality, mean age 61, mean testosterone 487 ng/dL), low testosterone was associated with higher mortality, with heterogeneity too great for a summary estimate and stronger in older, lower-testosterone, shorter and non-morning-sampled studies.","url":"https://pubmed.ncbi.nlm.nih.gov/21816776/"},{"id":"ruth2020","authors":"Ruth KS, Day FR, Tyrrell J, Thompson DJ, Wood AR, Mahajan A, et al.","title":"Using human genetics to understand the disease impacts of testosterone in men and women","journal":"Nature Medicine","year":2020,"pmid":"32042192","doi":"10.1038/s41591-020-0751-5","type":"mendelian-randomization","finding":"In 425,097 UK Biobank participants, genetically higher testosterone reduced type 2 diabetes in men (odds ratio 0.86 per standard deviation) and raised prostate cancer risk, while raising type 2 diabetes (1.37) and polycystic ovary syndrome (1.51) in women.","url":"https://pubmed.ncbi.nlm.nih.gov/32042192/"},{"id":"travison2007","authors":"Travison TG, Araujo AB, O'Donnell AB, Kupelian V, McKinlay JB","title":"A population-level decline in serum testosterone levels in American men","journal":"Journal of Clinical Endocrinology and Metabolism","year":2007,"pmid":"17062768","doi":"10.1210/jc.2006-1375","type":"cohort","finding":"In 1,532 Boston men observed across 1987 to 2004, testosterone fell with calendar year independently of age, by more than the cross-sectional age decline, unexplained by smoking or obesity.","url":"https://pubmed.ncbi.nlm.nih.gov/17062768/"},{"id":"lokeshwar2021","authors":"Lokeshwar SD, Patel P, Fantus RJ, Halpern J, Chang C, Kargi AY, et al.","title":"Decline in Serum Testosterone Levels Among Adolescent and Young Adult Men in the USA","journal":"European Urology Focus","year":2021,"pmid":"32081788","doi":"10.1016/j.euf.2020.02.006","type":"cross-sectional","finding":"In 4,045 adolescent and young adult men in NHANES 1999 to 2016, mean total testosterone was lower in 2011 to 2016 than in 1999 to 2000, lower with higher BMI, and the trend persisted at normal BMI.","url":"https://pubmed.ncbi.nlm.nih.gov/32081788/"},{"id":"harman2001","authors":"Harman SM, Metter EJ, Tobin JD, Pearson J, Blackman MR; Baltimore Longitudinal Study of Aging","title":"Longitudinal effects of aging on serum total and free testosterone levels in healthy men. Baltimore Longitudinal Study of Aging","journal":"Journal of Clinical Endocrinology and Metabolism","year":2001,"pmid":"11158037","doi":"10.1210/jcem.86.2.7219","type":"cohort","finding":"In 890 men, total testosterone fell by 0.124 nmol/L a year independent of health factors; hypogonadal values by total-testosterone criteria reached about 20 percent over 60, 30 percent over 70 and 50 percent over 80.","url":"https://pubmed.ncbi.nlm.nih.gov/11158037/"},{"id":"corona2013","authors":"Corona G, Rastrelli G, Monami M, Saad F, Luconi M, Lucchese M, et al.","title":"Body weight loss reverts obesity-associated hypogonadotropic hypogonadism: a systematic review and meta-analysis","journal":"European Journal of Endocrinology","year":2013,"pmid":"23482592","doi":"10.1530/EJE-12-0955","type":"meta-analysis","finding":"Across 24 studies, a low-calorie diet raised total testosterone by 2.87 nmol/L and bariatric surgery by 8.73 nmol/L; the degree of weight loss was the best determinant of the rise.","url":"https://pubmed.ncbi.nlm.nih.gov/23482592/"},{"id":"camacho2013","authors":"Camacho EM, Huhtaniemi IT, O'Neill TW, Finn JD, Pye SR, Lee DM, et al.; EMAS Group","title":"Age-associated changes in hypothalamic-pituitary-testicular function in middle-aged and older men are modified by weight change and lifestyle factors: longitudinal results from the European Male Ageing Study","journal":"European Journal of Endocrinology","year":2013,"pmid":"23425925","doi":"10.1530/EJE-12-0890","type":"cohort","finding":"In 2,395 men followed 4.4 years, testosterone fell 0.1 nmol/L a year and SHBG rose; weight loss raised and weight gain lowered testosterone and SHBG proportionally; free testosterone changed only with weight change of 15 percent or more.","url":"https://pubmed.ncbi.nlm.nih.gov/23425925/"},{"id":"grossmann2017","authors":"Grossmann M, Matsumoto AM","title":"A Perspective on Middle-Aged and Older Men With Functional Hypogonadism: Focus on Holistic Management","journal":"Journal of Clinical Endocrinology and Metabolism","year":2017,"pmid":"28359097","doi":"10.1210/jc.2016-3580","type":"review","finding":"Functional hypogonadism in obese, comorbid middle-aged and older men is potentially reversible; weight loss, treatment of comorbidities and withdrawal of offending drugs bring clinical improvement and a modest rise in testosterone; definitive randomized trials are lacking.","url":"https://pubmed.ncbi.nlm.nih.gov/28359097/"},{"id":"yeap2016","authors":"Yeap BB, Grossmann M, McLachlan RI, Handelsman DJ, Wittert GA, Conway AJ, et al.","title":"Endocrine Society of Australia position statement on male hypogonadism (part 1): assessment and indications for testosterone therapy","journal":"Medical Journal of Australia","year":2016,"pmid":"27510348","doi":"10.5694/mja16.00393","type":"consensus","finding":"Pathological hypogonadism warrants treatment at any age; obesity, metabolic syndrome, diabetes, glucocorticoids and opioids lower testosterone without raising LH and FSH, and clear evidence for treating that state is lacking; calculated free testosterone is not recommended for decisions.","url":"https://pubmed.ncbi.nlm.nih.gov/27510348/"},{"id":"leproult2011","authors":"Leproult R, Van Cauter E","title":"Effect of 1 week of sleep restriction on testosterone levels in young healthy men","journal":"JAMA","year":2011,"pmid":"21632481","doi":"10.1001/jama.2011.710","type":"rct","finding":"In 10 healthy young men, one week of 5-hour nights lowered daytime testosterone by 10 to 15 percent (16.5 versus 18.4 nmol/L) with lower vigor and no change in cortisol.","url":"https://pubmed.ncbi.nlm.nih.gov/21632481/"},{"id":"brambilla2009","authors":"Brambilla DJ, Matsumoto AM, Araujo AB, McKinlay JB","title":"The effect of diurnal variation on clinical measurement of serum testosterone and other sex hormone levels in men","journal":"Journal of Clinical Endocrinology and Metabolism","year":2009,"pmid":"19088162","doi":"10.1210/jc.2008-1902","type":"cross-sectional","finding":"In 66 men aged 30 to 80, testosterone at 4 p.m. was 20 to 25 percent lower than at 8 a.m. in men aged 30 to 40 and about 10 percent lower at 70; 17 men with an afternoon value under 300 ng/dL were normal at all three morning visits.","url":"https://pubmed.ncbi.nlm.nih.gov/19088162/"},{"id":"brambilla2007","authors":"Brambilla DJ, O'Donnell AB, Matsumoto AM, McKinlay JB","title":"Intraindividual variation in levels of serum testosterone and other reproductive and adrenal hormones in men","journal":"Clinical Endocrinology","year":2007,"pmid":"18052942","doi":"10.1111/j.1365-2265.2007.02976.x","type":"laboratory","finding":"In 132 men sampled over 6 months, biological variation exceeded assay variation; two measurements differed by more than 18 to 28 percent about half the time and by more than 27 to 54 percent a quarter of the time; one sample is not enough, more than three probably unwarranted.","url":"https://pubmed.ncbi.nlm.nih.gov/18052942/"},{"id":"wang2004","authors":"Wang C, Catlin DH, Demers LM, Starcevic B, Swerdloff RS","title":"Measurement of total serum testosterone in adult men: comparison of current laboratory methods versus liquid chromatography-tandem mass spectrometry","journal":"Journal of Clinical Endocrinology and Metabolism","year":2004,"pmid":"14764758","doi":"10.1210/jc.2003-031287","type":"laboratory","finding":"Across 122 samples on six immunoassays versus LC-MS/MS, one platform read 90 ng/dL low and another 99 ng/dL high; below 100 ng/dL the immunoassays were imprecise and biased; over 60 percent of adult-male results fell within 20 percent of the reference method.","url":"https://pubmed.ncbi.nlm.nih.gov/14764758/"},{"id":"rosner2007","authors":"Rosner W, Auchus RJ, Azziz R, Sluss PM, Raff H","title":"Position statement: Utility, limitations, and pitfalls in measuring testosterone: an Endocrine Society position statement","journal":"Journal of Clinical Endocrinology and Metabolism","year":2007,"pmid":"17090633","doi":"10.1210/jc.2006-1864","type":"consensus","finding":"Laboratory proficiency testing should rest on accurate measurement of known concentrations, not agreement among users of the same method; normative values should be established once standardization is in place, accounting for the many variables that influence testosterone.","url":"https://pubmed.ncbi.nlm.nih.gov/17090633/"},{"id":"bawor2015","authors":"Bawor M, Bami H, Dennis BB, Plater C, Worster A, Varenbut M, et al.","title":"Testosterone suppression in opioid users: a systematic review and meta-analysis","journal":"Drug and Alcohol Dependence","year":2015,"pmid":"25702934","doi":"10.1016/j.drugalcdep.2015.01.038","type":"meta-analysis","finding":"Across 17 studies with 2,769 participants (800 opioid users), men using opioids had significantly lower mean testosterone than controls regardless of opioid type, including methadone; women were unaffected; heterogeneity high and evidence quality low.","url":"https://pubmed.ncbi.nlm.nih.gov/25702934/"},{"id":"potter2021","authors":"Potter NJ, Tomkinson GR, Dufner TJ, Walch TJ, Roemmich JN, Wilson PB, Fitzgerald JS","title":"Effects of Exercise Training on Resting Testosterone Concentrations in Insufficiently Active Men: A Systematic Review and Meta-Analysis","journal":"Journal of Strength and Conditioning Research","year":2021,"pmid":"35134000","doi":"10.1519/JSC.0000000000004146","type":"meta-analysis","finding":"Across 11 randomized trials of 421 insufficiently active men aged 19 to 75, a median 12 weeks of aerobic, resistance or combined training had no effect on resting total testosterone (standardized mean difference 0.00).","url":"https://pubmed.ncbi.nlm.nih.gov/35134000/"},{"id":"dandrea2020","authors":"D'Andrea S, Spaggiari G, Barbonetti A, Santi D","title":"Endogenous transient doping: physical exercise acutely increases testosterone levels-results from a meta-analysis","journal":"Journal of Endocrinological Investigation","year":2020,"pmid":"32297287","doi":"10.1007/s40618-020-01251-3","type":"meta-analysis","finding":"Across 48 studies and 569 men, moderate and high-intensity exercise raised total testosterone acutely (standardized mean difference 0.74) within 30 minutes but not after; mild activity did not; SHBG did not change.","url":"https://pubmed.ncbi.nlm.nih.gov/32297287/"},{"id":"moosazadeh2024","authors":"Moosazadeh M, Heydari K, Rasouli K, Azari S, Afshari M, Barzegari S, et al.","title":"Association of the Effect of Alcohol Consumption on Luteinizing Hormone (LH), Follicle-Stimulating Hormone (FSH), and Testosterone Hormones in Men: A Systematic Review and Meta-Analysis","journal":"International Journal of Preventive Medicine","year":2024,"pmid":"39867254","doi":"10.4103/ijpvm.ijpvm_81_24","type":"meta-analysis","finding":"Across 17 observational studies, testosterone did not differ significantly between drinkers and nondrinkers overall (standardized mean difference 0.03); the authors reserve concern for the notable alterations seen in alcohol use disorder.","url":"https://pubmed.ncbi.nlm.nih.gov/39867254/"},{"id":"spratt1993","authors":"Spratt DI, Cox P, Orav J, Moloney J, Bigos T","title":"Reproductive axis suppression in acute illness is related to disease severity","journal":"Journal of Clinical Endocrinology and Metabolism","year":1993,"pmid":"8501163","doi":"10.1210/jcem.76.6.8501163","type":"cohort","finding":"In 59 men admitted to critical care, admission testosterone was below healthy controls in every severity group and lowest in the sickest (APACHE II above 15), with the differences widening during hospitalization.","url":"https://pubmed.ncbi.nlm.nih.gov/8501163/"},{"id":"snyder2016","authors":"Snyder PJ, Bhasin S, Cunningham GR, Matsumoto AM, Stephens-Shields AJ, Cauley JA, et al.; Testosterone Trials Investigators","title":"Effects of Testosterone Treatment in Older Men","journal":"New England Journal of Medicine","year":2016,"pmid":"26886521","doi":"10.1056/NEJMoa1506119","type":"rct","finding":"In 790 symptomatic men aged 65 and older with testosterone under 275 ng/dL, one year of testosterone gel moderately improved sexual activity (treatment effect 0.58), desire and erectile function, with no benefit in vitality or walking distance and effect sizes for mood below 0.20; too few men to judge risk.","url":"https://pubmed.ncbi.nlm.nih.gov/26886521/"},{"id":"lincoff2023","authors":"Lincoff AM, Bhasin S, Flevaris P, Mitchell LM, Basaria S, Boden WE, et al.; TRAVERSE Study Investigators","title":"Cardiovascular Safety of Testosterone-Replacement Therapy","journal":"New England Journal of Medicine","year":2023,"pmid":"37326322","doi":"10.1056/NEJMoa2215025","type":"rct","finding":"In 5,246 men aged 45 to 80 with cardiovascular disease or high risk, symptoms and two fasting values under 300 ng/dL, testosterone gel titrated to 350 to 750 ng/dL was noninferior to placebo for major adverse cardiac events (7.0 versus 7.3 percent, hazard ratio 0.96) over a mean 33 months, with more atrial fibrillation, acute kidney injury and pulmonary embolism.","url":"https://pubmed.ncbi.nlm.nih.gov/37326322/"},{"id":"thornton2001","authors":"Thornton JW","title":"Evolution of vertebrate steroid receptors from an ancestral estrogen receptor by ligand exploitation and serial genome expansions","journal":"Proceedings of the National Academy of Sciences of the United States of America","year":2001,"pmid":"11331759","doi":"10.1073/pnas.091553298","type":"laboratory","finding":"The first steroid receptor was an estrogen receptor, then a progesterone receptor; the mammalian set arose by two genome expansions, before and after jawed vertebrates; androgen and corticoid regulation are relatively recent; the ligand exploitation model, in which the terminal ligand of a pathway gets the first receptor.","url":"https://pubmed.ncbi.nlm.nih.gov/11331759/"},{"id":"thornton2003","authors":"Thornton JW, Need E, Crews D","title":"Resurrecting the ancestral steroid receptor: ancient origin of estrogen signaling","journal":"Science","year":2003,"pmid":"14500980","doi":"10.1126/science.1086185","type":"laboratory","finding":"An estrogen receptor ortholog from the mollusk Aplysia and a resurrected ancestral steroid receptor show the family diversified before bilaterally symmetric animals, with estrogen receptor-like function; the gene was lost in arthropods and nematodes.","url":"https://pubmed.ncbi.nlm.nih.gov/14500980/"},{"id":"markov2017","authors":"Markov GV, Gutierrez-Mazariegos J, Pitrat D, et al.","title":"Origin of an ancient hormone/receptor couple revealed by resurrection of an ancestral estrogen","journal":"Science Advances","year":2017,"pmid":"28435861","doi":"10.1126/sciadv.1601778","type":"laboratory","finding":"Cladistic analysis of steroid pathways: side-chain cleavage is common to most vertebrate steroids and aromatization was co-opted from an older pathway; the ancestral aromatized steroids kept a side chain (paraestrols), and synthesized paraestrol A binds and activates the ancestral steroid receptor.","url":"https://pubmed.ncbi.nlm.nih.gov/28435861/"},{"id":"markov2009","authors":"Markov GV, Tavares R, Dauphin-Villemant C, Demeneix BA, Baker ME, Laudet V","title":"Independent elaboration of steroid hormone signaling pathways in metazoans","journal":"Proceedings of the National Academy of Sciences of the United States of America","year":2009,"pmid":"19571007","doi":"10.1073/pnas.0812138106","type":"laboratory","finding":"Comparative genomics of complete metazoan genomes: steroidogenesis was independently elaborated in the three main bilaterian lineages, and steroidogenic cytochrome P450 enzymes descend from xenobiotic-detoxifying ones.","url":"https://pubmed.ncbi.nlm.nih.gov/19571007/"},{"id":"baker2011","authors":"Baker ME","title":"Origin and diversification of steroids: co-evolution of enzymes and nuclear receptors","journal":"Molecular and Cellular Endocrinology","year":2011,"pmid":"20654689","doi":"10.1016/j.mce.2010.07.013","type":"review","finding":"CYP11A and CYP19, enzymes of adrenal and sex steroid synthesis, first appear in the common ancestor of amphioxus and vertebrates, correlating with the first receptors closely similar to vertebrate steroid receptors; other steroidogenic CYP450s trace to invertebrate xenobiotic detoxification.","url":"https://pubmed.ncbi.nlm.nih.gov/20654689/"}],"written":"2026-10-10","updated":"2026-10-10","reviewed":true,"reviewedOn":"2026-10-12","related":{"essays":["testosterone-200-apob-120-what-a-chatbot-cannot-do"],"glossary":["SHBG","Reference range","Optimal range","Bioidenticals"],"symptoms":["low-libido","erectile-dysfunction","fatigue","hormonal-imbalance"],"entries":["fasting-insulin","homa-ir","hba1c","tsh","ferritin","vitamin-d"]},"url":"https://taggehealth.com/labs/testosterone-men","author":"Daniel Tagge, MD","publisher":"Tagge Health","license":"Free to read and cite with attribution to the URL above."}