Longevity & Aging · Metabolic & Cardiometabolic
testosterone therapy suppresses sperm production
In plain terms: Does testosterone therapy affect fertility?
Strong support Longevity & Aging
Part of: 💊 Testosterone therapy (TRT)
RefutedContestedStrong support
consensus score 1.00
Yes — and this is important and often overlooked. Taking testosterone shuts down your body's own testosterone and sperm production, frequently causing very low or zero sperm counts (it was actually studied as a male contraceptive). It's usually reversible after stopping, but recovery can take a year or more. Men who want children should not be on standard testosterone without protective co-treatment.
📅 Last reviewed: 2026-07-14 ⓘ
Evidence ladder
How far up the ladder this claim has climbed. A high consensus on a low rung means "consistent so far," not "proven in people."
Top evidence so far: Human trials (RCT / n-of-1)
MechanismIn-vitroAnimalObservationalRCTMeta-analysis
How the studies fall
7 support 0 contradict 0 tested null 0 mixed · 7 sources, 7 independent groups
What the evidence shows
This is a definitively established and important harm: exogenous testosterone suppresses the body's own testosterone and sperm production, frequently causing severe low sperm counts or complete absence of sperm (azoospermia).
The evidence (15)
| Source | Grade | Stance | Quality | Finding |
|---|---|---|---|---|
| Swerdloff RS, Bagatell CJ, Wang C, Anawalt BD, Berman N, Steiner B, Bremner WJ 1998 · J Clin Endocrinol Metab | observational | supports | low | Single-arm, open-label induction-then-maintenance male-contraceptive study in 15 healthy fertile volunteers; NOT randomised and no control arm - every subject received the identical regimen: 'In this study, 15 healthy men (age 21-41 yr) with normal semen analyses were treated with T enanthate (TE) 100 mg im/week plus 10 mg Nal-Glu GnRH antagonist sc daily for 12 weeks to induce azoospermia or severe oligozoospermia.' The prior extract's headline number was doubly misstated. (1) It attributed the 14/15 to testosterone, but 14/15 is the COMBINED antagonist+TE induction result: 'At 12-16 weeks, 10 of 15 subjects had zero sperm counts, and 14 of 15 had sperm counts less than 3 x 10(6)/mL.' (2) It labelled the endpoint 'azoospermia or severe oligozoospermia', but this trial's cut is <3 x 10(6)/mL - three times looser than the <=1 x 10(6)/mL severe-oligozoospermia threshold used by the WHO/Liu/Gu trials this claim is otherwise built on. Testosterone alone is isolated only in the maintenance phase, where the number is 13 of 14, not 14 of 15: 'The 14 who were suppressed on combined treatment were maintained on TE alone (100 mg/week im) for an additional 20 weeks' and 'Thirteen of 14 subjects in the TE alone phase had sperm counts maintained at less than 3 x 10(6)/mL for 20 weeks.' Azoospermia proper: 'Ten remained persistently azoospermic or had sperm concentration of 0.1 x 10(6)/mL once during maintenance.' Gonadotropins fell to 'LH and FSH levels ... 0.4+/-0.2 IU/L and 0.5+/-0.2 IU/L in the induction phase, which was maintained in the maintenance phase.' Authors' conclusion: 'sperm counts suppressed with GnRH antagonist plus T can be maintained with relatively low dose TE treatment alone.' GRADE RCT -> observational at re-read: nothing was randomised, there is no comparator, and the design bearing on this claim is an uncontrolled pre-post series (same correction as trt-gu-2009). QUALITY moderate -> low: n=15/14, no control, abstract-only provenance, a suppression threshold looser than the claim's, healthy contraceptive volunteers rather than TRT patients, and a maintenance denominator conditioned on already having suppressed - so the TE-alone phase cannot separate testosterone's own effect from carry-over of the antagonist-induced suppression. Direction supports the claim. |
| McBride JA, Coward RM 2016 · Asian J Androl | observational | supports | low | Narrative review, no primary data: 'Both TRT and AAS use can lead to suppression of the hypothalamic-pituitary-gonadal (HPG) axis, resulting in a diminution of spermatogenesis and potential infertility.' The authors concede the direct evidence is missing — 'Data specifically describing the natural history of unassisted spermatogenesis recovery after long-term TRT are lacking, but such information can be extrapolated from the male contraceptive literature' — so the quantitative backbone is borrowed from Liu 2006 (Lancet) pooled male-contraceptive trials: suppression to <1 million/mL (severe oligozoospermia, NOT azoospermia alone) in a median 3.5 months, and recovery in '67%, 90%, 96%, and 100% of men at 6, 12, 16, and 24 months, respectively, after discontinuation of testosterone exposure.' Review-role echo of those trials, not an independent observation. |
| Oduwole OO et al. 2014 · FASEB J | animal | supports | moderate | OFF-SCOPE for this claim. LHR-/- knockout mice have no LH receptor, so exogenous T cannot withdraw a gonadotropin-driven intratesticular-T supply - the pathway the claim runs on is deleted in the model. What the paper measured runs the other way: T dose-dependently ACTIVATED spermatogenesis (elongated spermatids 0 per testis in untreated and 0.5-1.5 mg T mice, rising to 35.0 x10^6 at 5.0 mg T vs 32.2 in wild-type), and 'These mice also showed evidence of fertility, with 3/8 of mice treated with 5.0 mg T, and 5/8 of those treated with Silastic T implants siring litters.' The authors name the caveat: 'we assessed the dose response of T induced stimulation of spermatogenesis in hypogonadism rather than T induced suppression of spermatogenesis in eugonadism.' Its real target is the neighbouring question of the intratesticular-T threshold (full spermatogenesis at 5.4-17.7 nM vs 120-140 nM in wild-type), which is why T contraception fails to suppress uniformly. The previous abstract-grade extract asserted the opposite of the paper's own result. |
| Rabijewski M 2016 · Pol Merkur Lekarski | mechanism | supports | low | Single-author NARRATIVE REVIEW in Polish (English abstract only; MEDLINE PT: Review), no data of its own — not one number in the whole abstract. Rabijewski states: 'Exogenous testosterone and anabolic androgenic steroids can inhibit the hypothalamic-pituitary-gonadal axis leading to decreasing of endogenous testosterone synthesis and impaired spermatogenesis.' Note the modal — 'CAN inhibit', unquantified, no incidence and no endpoint. Therapeutic thesis: 'In hypogonadal men who are in reproduction age the goal of therapy should be not only replacement therapy but also achiving and/or maintaining of spermatogenesis', with hCG and SERMs said to 'reverse spermatogenesis disturbances and can to maintain elevated intratesticular testosterone levels necessary to optimal spermatogenesis' — asserted without a single supporting figure or named trial. Secondary summary of the male-contraceptive trial literature that also appraises this claim; not an independent vote. |
| Gu Y, Liang X, Wu W, Liu M, Song S, Cheng L, Bo L, Xiong C, Wang X, Liu X, Peng L, Yao K 2009 · J Clin Endocrinol Metab | observational | supports | moderate | Single-arm multicenter phase III male-contraceptive efficacy trial, NOT randomized and with no control arm - 'This was a multicenter, phase III, contraceptive efficacy clinical trial', in which all participants received one regimen: 'A total of 1045 healthy fertile Chinese men were recruited throughout China into the study' and 'Injections of 500 mg TU were administered monthly for 30 months.' Suppression was near-complete: 'Forty-three participants (4.8%) did not achieve azoospermia or severe oligozoospermia within the 6-month suppression phase' (threshold <=1 x 10^6/ml), i.e. ~95% of assessed men suppressed - note the 4.8% is over the ~896 men assessed, not over the 1045 recruited. The disconfirming half is stated by the paper too: 'The combined method failure rate was 6.1%, comprising 4.8% with inadequate suppression and 1.3% with postsuppression sperm rebound.' Reversibility: 'Spermatogenesis returned to the normal fertile reference range in all but two participants.' Grade corrected RCT -> observational (single-arm, no randomization or masking described anywhere in the readable text); quality high -> moderate on abstract-only provenance. Chinese men only - azoospermia rates on androgen-alone regimens run higher in East Asian than Caucasian cohorts, so the ~95% figure is not a general one. |
| Page ST et al. 2006 · J Clin Endocrinol Metab | observational | supports | moderate | Randomized two-arm trial (JCEM 2006, n=44) in healthy male-contraception volunteers, but the randomization tests the ADD-ON, not testosterone: 'Forty-four healthy men, ages 18-55 yr, were randomized to T gel (100 mg daily)+DMPA (300 mg/3 months) or acyline (300 microg/kg.2 wk x 12 wk)+T gel+DMPA.' Every man received testosterone, so for this claim there is no testosterone-free comparator and the suppression signal is uncontrolled pre-post. 'Thirty-eight men completed the 24-wk treatment protocol.' The endpoint is severe oligospermia, not azoospermia — the stated objective was 'to determine the rate of severe oligospermia (< or = 1 million sperm/ml) using T gel+DMPA' and the result was 'All men had dramatic suppression of spermatogenesis; 90% of the subjects became severely oligospermic, a rate comparable to implantable and injectable T+progestin combinations.' The abstract never says whether that 90% is over the 44 randomized or the 38 completers, and reports no azoospermia rate. Testosterone was co-administered with DMPA, an independently potent gonadotropin suppressant, in both arms, so the paper cannot apportion the effect to testosterone alone. The paper's own randomized question was null: 'The addition of acyline did not significantly accelerate spermatogenic suppression or improve rates of severe oligospermia.' Direction supports the claim. GRADE RCT -> observational at re-read #460 under CONVENTIONS §5 addendum 3: the randomised variable was acyline, not testosterone, and every arm received testosterone, so for THIS claim there is no randomised comparison and RCT weight 5 would be unearned. The trial keeps its RCT standing for the acyline question it actually randomised, which is banked as a tested-null candidate. Quality moderate kept (n=44, 14% attrition, DMPA co-intervention, healthy volunteers rather than TRT patients, abstract-only provenance). |
| Ledesma BR et al. 2023 · Fertility and Sterility | observational | supports | low | Retrospective single-centre case series (University of Miami, 2018-2022) of 45 men presenting with infertility after prior anabolic-steroid/testosterone use — 'A retrospective analysis of a cohort of men with a prior history of anabolic steroid use and infertility complaints (between 2018 and 2022) was conducted.' Median prior T use 4 years; 'The median initial sperm concentration was 0 million/cc (IQR 0-1.15), and 23 (51.1%) men initially presented with azoospermia.' After stopping T plus clomiphene+hCG: 'In initially azoospermic men (N: 23), 5 were lost to follow-up, 6 (33.3%) progressed to severe oligospermia (<5 million/cc), 6 (33.3%) to oligospermia (<15 million/cc), 1 (5.6%) to normozoospermia (>15 million/cc), and 5 (27.8%) remained azoospermic after medical treatment for 6 months.' Quality low: the cohort was selected on the very outcome asserted (men referred for infertility), there is no unexposed or untreated comparator, 5 of 23 were lost to follow-up in the key subgroup, and prescribed TRT is not separated from non-prescription steroid use (34.8% oral). Supports the claim's direction and its incomplete-reversibility caveat; cannot estimate how often suppression occurs. |
| Hochu G, Geyer-Kim I, Kim E 2025 · Translational Andrology and Urology | observational | supports | moderate | Narrative (non-systematic) review, no new data — self-declared: 'As this review was narrative in nature rather than a systematic analysis, the potential for selection bias should be acknowledged.' It restates the suppression mechanism — 'Conventional testosterone replacement therapies (TRT) restore serum testosterone but suppress gonadotropins, resulting in impaired spermatogenesis' and 'This suppression leads to decreased levels of luteinizing hormone (LH) and follicle-stimulating hormone (FSH), which in turn reduce intratesticular testosterone (ITT) and impair Sertoli cell function-critical elements of spermatogenesis' — and cites the 2024 AUA Grade-A recommendation that 'Exogenous testosterone therapy should not be prescribed to men who are currently trying to conceive', plus recovery data ('In one study, 67% percent of men recovered sperm concentration of 20 million/mL within 6 months after cessation of TRT; however, some men required up to 2 years'). Direction is right and the claim is well supported, but the supporting cohorts (Coviello 2005, Hsieh 2013, Kim 2016, Ramasamy 2020, Stocks 2025) are other groups' — counting this paper again would double-count by construction, so it is retained for transparency at zero weight. |
| Al Hashimi M et al. 2025 · Asian J Androl | observational | supports | low | Clinical-guide review: exogenous testosterone reduces intratesticular testosterone and impairs spermatogenesis, producing azoospermia. |
| Whitaker BM et al. 2021 · Expert Rev Endocrinol Metab | mechanism | supports | low | Narrative review, no new data: anabolic-steroid use is 'an uncommon but important cause of male infertility'; recovery is hedged twice — SERMs/hCG/AIs/rFSH 'may lead to spermatogenesis recovery' 'although additional studies are necessary'. Subject arm is supraphysiologic AAS/SARM misuse, not therapeutic TRT. |
| Moshe S et al 2026 · AME Case Rep | observational | supports | low | Case report (n=1) of a 33-y-o man with congenital HH from PNPLA6/Oliver-McFarlane syndrome, on testosterone replacement from age 12.5: 'Despite androgen-induced secondary sexual characteristics, as expected testicular volume remained prepubertal and azoospermia persisted'; at 29 y 'markedly atrophic testes (~4 mL)', suppressed LH/FSH, and the authors' reading is that 'prolonged testosterone use led to severe testicular involution and azoospermia'. Directionally consistent with suppression but causally confounded — the underlying gonadotropin deficiency alone produces azoospermia, gonadotropins were still suppressed 2 y after testosterone was stopped, and after ~2 y of hCG+rFSH 'repeated semen analyses over >18 months remained azoospermic', sperm obtained only by micro-TESE. Weight stays minimal (observational, low). |
| Handelsman DJ 1995 · Aust N Z J Med | mechanism | supports | low | Narrative review, no original data: 'Landmark WHO contraceptive efficacy studies have established that hormonally-induced azoospermia provides highly effective and reversible contraception for at least 12 months with minimal short-term side effects.' Also: 'Even among the small subgroup of men who remain oligozoospermic during hormonal suppression, good contraceptive efficacy is achieved.' The entire empirical basis is the two WHO multicentre trials, so this vote is review-layer and not independent of them. |
| Wallace EM, Gow SM, Wu FC 1993 · J Clin Endocrinol Metab | observational | supports | moderate | Edinburgh single-arm contraceptive trial, 28 healthy fertile men on 200 mg testosterone enanthate i.m. weekly for 6 months: 'Seventeen of the 28 men achieved azoospermia; the other 11 remained oligozoospermic (sperm density 3.3-4.7 x 10(6)/mL)'. 'Both LH and FSH were consistently suppressed to below the limits of detection.' Note the threshold: no <=1 or <3 x 10^6/mL cut was used, and the 11 non-azoospermic men were above both - so this is 61% azoospermia on a denominator of 28 dosed men, not 28/28 reaching a contraceptive endpoint. Sperm density recovered to baseline or >20 x 10^6/mL after stopping, more slowly in the azoospermic group. |
| Rajmil O, Moreno-Sepulveda J 2023 · Actas Urol Esp (Engl Ed) | observational | supports | low | OFF-SCOPE for a testosterone-therapy claim, on both subject and outcome. The exposure is illicit anabolic-steroid abuse, not therapeutic testosterone - 'This systematic review aims to evaluate the optimal treatment for male infertility resulting from Anabolic Androgenic Steroids (AAS) abuse.' - and the measured outcome is the efficacy of RECOVERY protocols, not suppression: 'Studies that compared different protocols for the recovery of spermatogenesis in patients after AAS use were included.' '13 studies investigating different protocols to restore spermatogenesis in patients with AAS abuse met the inclusion criteria.' Suppression of spermatogenesis appears only as a background exhortation - 'Clinicians need to be aware of the detrimental effects of AAS on spermatogenesis.' - with no suppression rate, no participant total and no pooled estimate reported anywhere in the paper. What the review actually synthesises is treatment: 'The available agents that showed restoration of spermatogenesis include injectable gonadotropins, selective estrogen receptor modulators, and aromatase inhibitors, but their use is still poorly described in the literature.' and 'AAS-associated infertility may be reversible, but sperm production may take over a year to normalize.' It is in any case a PRISMA systematic review reporting no new data (CONVENTIONS Sec.5 review) - 'A systematic review was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement.' - and PubMed types it 'Systematic Review', not 'Meta-Analysis', so the stored grade meta-analysis overstates it. Quality low: abstract-only (no PMCID, Europe PMC isOpenAccess=N), no registration seen, and the authors' own verdict on their evidence base is 'Further understanding of male reproductive endocrinology and high-quality data on the field of restoration of spermatogenesis after AAS abuse are warranted.' The prior extract also embellished the paper - it said 'testosterone-based AAS' (the paper says AAS generally; MeSH is Anabolic Agents / Testosterone Congeners), 'reversible in most' (the paper says infertility 'may be reversible'), and dropped aromatase inhibitors from the agent list. Subject-wise this paper belongs on claim-anabolic-steroids-cause-testicular-atrophy-and-infertility, still as a zero-weight review. GRADE meta-analysis -> observational at calibration #29b: two blind readers independently found it pools nothing and reports no effect size or participant total (Europe PMC types it 'Systematic Review'), so weight 8 was unearned; the row is zero-weighted by evidence_role in any case. Source study_type corrected meta-analysis -> review. |
| Anderson RA et al. 1997 · Hum Reprod | RCT | supports | moderate | Single-arm male-contraceptive trial (n=33, no control arm): 'Thirty-three healthy Caucasian men were given weekly i.m. injections of 200 mg testosterone oenanthate (TE), 18 became azoospermic, while 15 remained oligozoospermic' — i.e. every man's sperm production was suppressed, but 45% did not reach azoospermia. The <10% figure is testicular steroidogenesis, not sperm count: 'Urinary excretion of epitestosterone, a specific testicular product, was reduced to <10% of pretreatment values'. Same Edinburgh trial as trt-wallace-1993 (paper I of the series). |
Disagree, or know a study we missed?
We grade by evidence, not opinions. The way to weigh in is to point us to a study we haven't cited (check the evidence table above first), or to flag a problem with one we have. Every submission is reviewed; if it holds up, the grade updates and shows in Science Changes Its Mind.
Opens a short form. You'll sign in with Google so submissions are tied to a real account — we don't display your identity, and we only accept a link we can verify (PubMed, DOI, ClinicalTrials.gov).
Educational only, not medical advice. Grades and scores reflect published evidence weighted by study design and quality; see the methodology.