Gym, Protein Powder and Testosterone Before IVF: What to Stop and What to Keep

Evidence-based · Written by Dr. Lejla Fazlicic, D.Ac, L.Ac · All key claims cited to peer-reviewed research

The reason to stop the supplements has almost nothing to do with what is printed on the label. It is that 89.1% of FDA-flagged muscle-building supplements were found to contain synthetic steroids or steroid-like ingredients that were not declared — which means a man taking a "natural test booster" through his three-month window may be taking exogenous androgens without knowing it, and that is the one thing on this page that genuinely shuts sperm production down.

Working out what is in his routine, and what the timeline actually allows, is the part worth doing properly.

Book a free 15-minute call · See what working together looks like

One thing first. If he has been on testosterone and nobody told him what it does to fertility, that is an extremely common failure and it is not his. In one clinic series, 12% of men taking testosterone had been prescribed it to treat their infertility. The information gap here is at the prescriber level, not the patient level.

Does testosterone therapy stop you getting your partner pregnant?

Effectively, yes — and it is worth understanding why a normal blood test is no reassurance at all.

The testis does not run on blood testosterone. It runs on intratesticular testosterone, which is normally about 40 times higher than serum levels — measured at 822 nmol/L in the testis against 22.8 in blood. When men were placed on a hormonal regimen, LH fell by 98%, FSH by 97%, and intratesticular testosterone by 98% to 13.1 nmol/L — while serum testosterone was unchanged or higher at 28.7. Sperm counts fell from 65 million/mL to 1.3 (Coviello et al., 2004).

That is seven men, so hold the specific numbers loosely — but the direction is confirmed everywhere. In a male-infertility clinic database of 4,400 men, 88.4% of those taking testosterone were azoospermic (Samplaski et al., 2014).

The cleanest way to see it is that testosterone was studied as a contraceptive and it worked. In a WHO multicentre trial, 271 healthy fertile men on weekly testosterone enanthate produced one pregnancy across 1,486 months — a rate of 0.8 conceptions per 100 person-years (WHO Task Force, 1990). That is comparable to female hormonal contraception.

The AUA/ASRM male infertility guideline states it as a clinical principle: "For the male interested in current or future fertility, clinicians should not prescribe exogenous testosterone therapy."

This is not a rare situation. Testosterone use among US men aged 18 to 45 rose fourfold between 2003 and 2013, from 29.2 to 118.1 per 10,000 person-years — and the rise in young men was steeper than in older men.

How long does it take to recover?

The best data available is an integrated analysis of individual participant data from 30 studies covering 1,549 men, with sperm output measured monthly until recovery (Liu et al., 2006):

Recovery of sperm concentration after stopping hormonal suppression

Milestone	Time or probability
Median time back to 20 million/mL	3.4 months (95% CI 3.2–3.5)
Median time back to 10 million/mL	3.0 months
Recovered to 20 million/mL by 6 months	67%
By 12 months	90%
By 16 months	96%
By 24 months	100%
Back to a man's own baseline concentration (WHO trial)	6.7 months

Recovery is faster with shorter treatment duration, shorter-acting preparations, higher baseline sperm concentration and older age. A man on long-acting testosterone undecanoate for several years should expect the slow end.

Two honest caveats. These men were healthy volunteers on controlled, time-limited study regimens — not men who have self-administered supraphysiological doses for years, and the "100% by 24 months" figure should not be transferred to them. And a later trial found roughly 5% had still not recovered at twelve months.

The practical implication is the uncomfortable one: three months is the median, not a margin. If a retrieval is three months away and he stops testosterone today, he has roughly a two-in-three chance of being back above 20 million/mL by six months. This needs a reproductive urologist now. There are medical approaches to accelerating recovery — one retrospective series of 49 men using hCG-based therapy reported return of spermatogenesis in 95.9%, at a mean of 4.6 months — but that is a physician's decision, not a self-directed protocol, and the series had no comparison group.

Anabolic steroids

Same mechanism, worse prognosis, and the honest version includes the part that does not always resolve. In a case-control study of 37 current users, 33 former users a mean of 2.5 years since stopping, and 30 controls, 27.2% of former users still had total testosterone below the lower reference limit, against 0% of controls. Current users showed severely decreased inhibin B and AMH — direct markers of impaired spermatogenesis (Rasmussen et al., 2016).

That study is cross-sectional and cannot exclude that men with lower baseline androgens are more likely to use in the first place. But "recovery is usually complete" is more accurate than "recovery is always complete," and a man with a steroid history deserves a semen analysis early rather than an assumption.

Are testosterone boosters worth taking?

Two separate answers: the labelled ingredients mostly do nothing, and what is not on the label is the actual risk.

The labelled ingredients

• Tribulus terrestris. A randomised placebo-controlled trial in 21 healthy men found no significant difference in testosterone, androstenedione or LH at either dose. The authors concluded tribulus saponins "possess neither direct nor indirect androgen-increasing properties."
• D-aspartic acid. Worse than useless in one trial: at 6 g/day for 14 days in resistance-trained men, total testosterone fell significantly and free testosterone fell versus placebo. A 12-week randomised trial by the same group found no change in testosterone and a 16% reduction in estradiol, concluding DAA is "ineffective at changing testosterone levels, or positively affecting training outcomes."
• Ashwagandha — does it increase sperm count? The frequently cited trial reported a 167% rise in sperm count in 21 treated oligospermic men — a pilot, single centre, reported as within-group change rather than a between-group effect, with an implausibly large effect size for a 90-day botanical, published in a journal with documented quality problems, and not independently replicated. A second commonly cited study had no placebo arm at all.
• Maca. A double-blind randomised pilot in 20 healthy men found "rising trends" in concentration and motility — meaning not statistically significant — and no hormone change.
• Fenugreek. No verified human trial with semen parameters as an endpoint exists.

What is actually in the tub

This is the argument that should change behaviour, and almost nobody hears it.

An analysis of the FDA's database of tainted products marketed as dietary supplements, covering 2007 to 2016, identified 776 adulterated supplements from 146 companies. Broken down by category, 45.5% were sexual enhancement products, 40.9% weight loss, and 11.9% muscle building. Among the muscle-building products, 89.1% contained synthetic steroids or steroid-like ingredients. And of 28 products named in multiple FDA warnings more than six months apart, 67.9% contained new unapproved ingredients when re-tested (Tucker et al., 2018).

Independently, a laboratory analysis of 634 ostensibly non-hormonal supplements bought from 215 suppliers across 13 countries found 14.8% contained anabolic androgenic steroids not declared on the label, with the US among the highest positive rates at 18.8% (Geyer et al., 2004) — though those products were bought in 2000 and 2001, so treat that as a historical rather than a current market estimate.

Put those together with the first half of this article. Undeclared androgens suppress spermatogenesis exactly as prescribed testosterone does. A man taking an unregulated muscle-building supplement through his preparation window is running a real chance of doing the single most harmful thing on this list while believing he is doing something helpful.

The FDA-flagged sample is enforcement-driven, so it cannot tell you what percentage of the whole market is adulterated. It tells you the category is not clean, and that the muscle-building shelf is the worst of it.

Do antioxidant supplements improve sperm?

Less than the marketing, and the honest answer has moved in the last few years.

The Cochrane review covering 61 studies and 6,264 subfertile men — the 2019 version, since updated in 2022 — found live birth odds ratio 1.79 but graded the evidence low quality, based on only 124 live births across 750 couples. The number that matters most: when studies at high risk of bias were removed, there was no evidence of increased live birth — odds ratio 1.38, confidence interval 0.89 to 2.16 (Smits et al., 2019). The authors' own conclusion describes the current evidence as "inconclusive based on serious risk of bias." Gastrointestinal upset was more common on antioxidants.

Then a properly designed trial tested it. MOXI was multicentre, double-blind, randomised and placebo-controlled across nine US fertility centres in 174 men with male-factor findings, using a combination of vitamins C and E, selenium, L-carnitine, zinc, folic acid and lycopene for three to six months. At three months, the change in sperm concentration was numerically worse on antioxidants. No significant difference in morphology, motility or DNA fragmentation. Cumulative live birth at six months was 15% on antioxidants versus 24% on placebo — a difference the trial was not powered to test, so the correct reading is "no benefit," not "antioxidants reduced live birth" (Steiner et al., 2020).

On CoQ10 specifically: the most-cited individual trial now carries an editorial expression of concern, and it reported no pregnancy or live birth outcome. Neither Cochrane nor MOXI supports a live-birth benefit.

AUA/ASRM: "Clinicians should counsel patients that the benefits of supplements (e.g., antioxidants, vitamins) are of questionable clinical utility in treating male infertility." The European Association of Urology puts it similarly — antioxidant regimens "may reduce the risk of DNA fragmentation and improve sperm quality. These data have not been supported by RCTs."

None of which means a prescribed supplement is wrong for a specific man. It means the decision belongs with his physician rather than with a shelf.

Should he stop lifting before IVF?

No. This is the part where restriction costs something and buys nothing.

The largest and best-designed study is in exactly the right population: 2,261 men contributing 4,565 fresh semen samples, enrolled before their first IVF cycle. Its finding, verbatim: "none of the semen parameters were materially associated with regular exercise." The single exception was cycling at five or more hours a week, associated with low sperm concentration at odds ratio 1.92 (1.03–3.56) and low total motile sperm count at 2.05 (1.19–3.56) (Wise et al., 2011).

And the direction of the other evidence is favourable. Among 189 young men, those in the top quartile of moderate-to-vigorous activity — 15 or more hours a week — had 73% higher sperm concentration than the bottom quartile, while men watching more than 20 hours of television weekly had 44% lower (Gaskins et al., 2015). A cross-sectional comparison found physically active men had better motility, better morphology and higher testosterone than sedentary men.

One important correction, because it cuts against the reassuring story and you should know it: the strongest-sounding randomised trial claiming that resistance training improves semen quality has been retracted. So the defensible statement is "lifting does not appear to harm semen quality and may modestly help" — not "lifting improves fertility."

Can too much exercise lower sperm count?

The threshold appears to sit at elite endurance volumes, not gym volumes.

The male overtraining evidence is thinner than its reputation. The most-cited study measured 12 high-level triathletes preparing for a national championship. Their hormones and conventional semen parameters were all within normal ranges; the only abnormality was sperm DNA fragmentation at 20.4 ± 6.1% (Vaamonde et al., 2018). Twelve men, no control group, cross-sectional, multiple correlations. It is hypothesis-generating and should be described that way.

The frequently quoted "training to exhaustion doubles fertility problems" finding is real, but it is a study of women — 3,887 participants in a Norwegian health study — not men. There is no equivalent male cohort.

A randomised trial of three exercise modalities in 280 healthy men found all three reduced seminal oxidative stress and inflammatory markers, with moderate-intensity continuous training superior to high-intensity. I would hold that loosely: a closely related trial by the same authors has been retracted, so I would not build a recommendation on that group's work alone.

The proportionate reading: gym training is fine. Marathon and ultra-endurance blocks during the preparation window are a reasonable thing to moderate, and that is a training-load conversation rather than a fertility prohibition.

Are creatine and protein powder bad for sperm?

There is no human evidence that either harms semen parameters, fertility-relevant hormones or fertility outcomes. Not weak evidence. None.

Creatine. The first human trial examining creatine and semen opens with the line: "No studies have investigated whether dietary creatine, administered alone or in combination with other nutrients, can induce changes in the physical-chemical and microscopic characteristics of human semen." That trial, in 15 volunteers, found a non-significant trend toward higher sperm concentration. It is a pilot with declared author conflicts of interest, so it is not evidence of benefit either — but it is the documented statement that, until 2025, nobody had looked.

The entire basis for "creatine messes with your hormones" is a single unreplicated crossover in 20 rugby players that found DHT rose 56% after a loading phase — with testosterone unchanged, all values within physiological range, no semen parameters measured, and a published critical comment attached.

Whey protein. A structured literature search returns no human trial of whey protein supplementation with semen parameters or fertility as an outcome. On the heavy-metal question, a quantitative risk assessment modelling arsenic, cadmium, mercury and lead in protein powders found a hazard index below 1 for all products at typical intake, with mass-gainer products highest and whey powders the lowest-risk category — an industry-affiliated analysis, so treat it as such. Note also that it assessed general toxicity, not reproductive endpoints: nobody has measured protein-powder heavy metals against semen quality at all.

So the honest answer to "should he stop his protein shake" is that there is nothing to stop it for.

What the evidence does not show

• That any supplement improves live birth. Cochrane's benefit disappears when high-risk-of-bias studies are removed; MOXI found nothing.
• That test boosters raise testosterone. Tribulus: no effect. D-aspartic acid at 6 g/day: testosterone went down.
• That creatine or whey harm fertility. No human study has found harm, because until recently no human study had asked.
• That resistance training improves semen quality. The trial most often cited for it has been retracted.
• That normal gym training harms anything. The largest study in men before IVF found no association with any semen parameter.
• That there is a male "training to exhaustion" study. The famous one is in women. The male evidence is 12 triathletes with no control group.
• That three months off testosterone is enough. It is the median, and one in three men will not be back above 20 million/mL at six months.

What we do with this

1. If he is on testosterone in any form, call a reproductive urologist this week. Not next month. The recovery clock is measured in months and the median is longer than most IVF runways.
2. Clear the muscle-building shelf. Not because tribulus does anything, but because of what has repeatedly been found alongside it.
3. Do not start an antioxidant stack on your own. If a physician has prescribed one for a specific finding, that is different. Off the shelf, the evidence does not support it and MOXI is a clean negative trial.
4. Keep lifting. The largest study in exactly this population found nothing, and active men look better than sedentary men on every parameter measured.
5. Keep the creatine and the whey. There is no evidence to stop them.
6. Moderate ultra-endurance blocks during the window — marathon and triathlon builds, not gym sessions — and treat it as a training-load choice rather than a rule.
7. Skip the post-gym sauna for the window. That one does have evidence, and it is covered in the heat article.

Same window. Same clock. Same team.

Related reading

• What does he actually have to give up before IVF?
• Hot tubs, saunas, cycling and laptops: which heat actually matters?
• Exercise in the 90 days before IVF
• The male fertility diet
• Lowering sperm DNA fragmentation
• What nobody can tell you yet about preparing for IVF
• The research library

Frequently asked questions

Does testosterone therapy cause infertility?

It suppresses sperm production, and a normal blood testosterone gives no reassurance. Intratesticular testosterone normally runs about 40 times higher than serum, and exogenous testosterone reduces it by around 98% while serum levels appear normal or high. Among men taking testosterone who presented to a male-infertility clinic, 88.4% were azoospermic. In a WHO trial, testosterone produced one pregnancy across 1,486 months of use — a contraceptive-level effect. AUA/ASRM state that clinicians should not prescribe it to men interested in fertility.

How long after stopping testosterone can he father a child?

The median time back to a sperm concentration of 20 million/mL is 3.4 months, based on monthly monitoring in 1,549 men. But only 67% have recovered by six months, 90% by twelve and 96% by sixteen, and the WHO trial put the return to a man's own baseline concentration at 6.7 months. Roughly 5% had not recovered at one year in a later trial. Recovery is faster after shorter treatment and with shorter-acting preparations. This needs a reproductive urologist, not a countdown.

Do testosterone boosters work, and are they safe before IVF?

The labelled ingredients largely do nothing — a randomised trial of tribulus found no effect on testosterone, and D-aspartic acid at 6 g/day significantly reduced total and free testosterone. The real risk is undeclared: 89.1% of FDA-flagged muscle-building supplements contained synthetic steroids or steroid-like ingredients, and 14.8% of 634 ostensibly non-hormonal supplements in an international analysis contained undeclared anabolic steroids. Those suppress sperm production exactly as prescribed testosterone does.

Should he stop going to the gym before IVF?

No. In 2,261 men contributing 4,565 semen samples before their first IVF cycle, none of the semen parameters were materially associated with regular exercise. Among 189 young men, the top quartile of moderate-to-vigorous activity at 15 or more hours a week had 73% higher sperm concentration than the bottom quartile. The only exercise signal was cycling at five or more hours a week. The concern threshold is elite endurance volume, not gym volume.

Is creatine or whey protein bad for sperm?

There is no human evidence of harm from either. The first human trial of creatine and semen quality, published in 2025, stated in its opening line that no prior study had examined the question, and found a non-significant trend toward higher concentration. The entire basis for hormone concerns is one unreplicated crossover in 20 rugby players in which DHT rose while testosterone was unchanged. No human trial of whey protein with semen or fertility endpoints exists.

Do antioxidant supplements improve sperm quality?

The evidence is weaker than the marketing. The Cochrane review of 61 studies and 6,264 men graded the evidence low to very low quality, and when studies at high risk of bias were removed there was no evidence of increased live birth. The randomised, placebo-controlled MOXI trial found no improvement in sperm concentration, motility, morphology or DNA fragmentation, with a cumulative live birth rate of 15% on antioxidants versus 24% on placebo. AUA/ASRM describe supplement benefits as of questionable clinical utility.

---

This article is educational and reflects associations reported in the scientific literature. It is not medical advice, not a diagnosis, and not a way to interpret your own results. As a Doctor of Acupuncture I offer lifestyle guidance that works alongside your physician and fertility clinic — I do not diagnose, interpret labs, or prescribe. Testosterone therapy should never be started, stopped or altered without a physician: that conversation belongs with a urologist or the original prescriber.