Antioxidants for Sperm Quality: What the Trials Found

No trial has shown that an antioxidant supplement makes a baby more likely. The two largest placebo-controlled trials in men found no improvement in live birth and no improvement in the semen measurements these products are sold to improve — one randomised 2,370 couples to folic acid with zinc, the other gave 174 men a seven-ingredient antioxidant.

That is not the same as proof that they do nothing. The trials are too small and too poorly reported to rule out a modest effect, and the review that reports one says so itself. What can be said is narrower and more useful: the claim on the label has not been demonstrated, and the specific studies quoted in support of it are weaker, and in four cases more compromised, than a product page will tell you.

After a semen report comes back abnormal, buying something is often the only step that feels like acting. That impulse is reasonable. It is worth knowing exactly what the bottle has and has not been shown to do before it becomes the plan.

The Folic Acid and Zinc Supplementation Trial is the biggest test of the commonest male fertility combination. Couples starting infertility treatment were randomised, and the male partner took folic acid with zinc, or a placebo, every day for six months. Roughly the same number of couples in each group had a baby. None of the semen measurements improved. Sperm DNA fragmentation was slightly higher in the men taking the supplement, and stomach upset was more common.

The Males, Antioxidants and Infertility trial tested a seven-ingredient combination — vitamin C, vitamin E, selenium, L-carnitine, zinc, folic acid and lycopene — in men who already had an abnormal semen result. After three months there was no difference from placebo in morphology, motility or DNA fragmentation.

Both trials were designed to find a benefit and were large enough to see a meaningful one. Neither did.

Cochrane's review is the largest summary of this question: 90 randomised trials, 10,303 subfertile men, 20 different oral antioxidants. Its headline is that antioxidants may increase live birth. Two things sit underneath that headline and change what it means.

The first is how thin the underlying data are. Of the 67 trials in the meta-analysis, only 20 reported whether anyone got pregnant and only 12 reported whether anyone had a baby. The live-birth result rests on 246 births across 1,283 couples in small and medium-sized trials. The review grades this very low certainty, which is its lowest rating.

The second is what happens when the weakest trials come out. The reviewers repeated the calculation with the trials at high risk of bias excluded, and the apparent benefit vanished — the confidence interval crossed the line where there is no effect at all. The review's own conclusion is that the current evidence is inconclusive because of serious risk of bias, poor reporting of randomisation, unclear or high dropout, and low event rates.

A separate 2025 review of 50 randomised trials, each running at least 12 weeks, reached the same place from a different direction: no effect on pregnancy and no effect on live birth, with individual nutrients moving individual semen numbers and most of the trials carrying a high risk of bias.

More of it than a product page will mention. Four papers that are routinely cited as evidence that a fertility antioxidant works now carry a formal flag from the journal that published them. Each can be checked in a few seconds by searching its PubMed identifier.

• The most-quoted trial of coenzyme Q10 in fertility — El Refaeey and colleagues, 2014, on coenzyme Q10 with clomiphene for polycystic ovary syndrome — has been retracted. PubMed lists it under the publication type Retracted Publication (PMID 24813752). An expression of concern was issued in September 2023 and the retraction notice followed in November 2023 (PMID 37683569).
• A trial of 300 mg coenzyme Q10 daily in 212 infertile men, which reported significant improvement in sperm density and motility, carries an expression of concern issued by the Journal of Urology on 10 January 2023 (trial PMID 19447425, notice PMID 36626312).
• A trial of selenium and N-acetyl-cysteine in 468 infertile men, which concluded by advocating their use for male infertility treatment, carries an expression of concern from the same journal on the same date (trial PMID 19091331, notice PMID 36626340).
• A myo-inositol trial in assisted reproduction carries an expression of concern issued in 2025 after a third party reported an undeclared conflict of interest — one author's role at a company marketing inositol products (trial PMID 21608442, notice PMID 40748343).

An expression of concern is not a finding of misconduct. It is the journal telling readers not to rely on the paper while it looks into a problem. A retraction is stronger: the paper has been withdrawn from the record. Either way, those four are among the largest and most-cited single studies in this field, and two of them are the specific trials behind the claim that coenzyme Q10 or selenium improves semen quality.

This matters practically. If you are weighing a product because a website said a nutrient was clinically proven, it is worth asking which study, and then checking whether that study still stands.

Not on the strength of a web page, and not without telling them. A prescriber may have a reason that does not appear in the trial literature — a documented deficiency, a specific result they are acting on, or a deliberate decision to try something cheap and low-risk while other steps are arranged. Stopping a prescribed treatment quietly also removes the chance to find out what they were watching for.

Two questions are reasonable to ask at the next appointment. What result would tell us this is working, and when are we going to check it. If both have answers, the supplement is part of a plan. If neither does, that is worth knowing too, and it is a conversation rather than a decision to make alone.

About three months, not three weeks. A cycle of sperm production was measured directly in 11 men who drank deuterium-labelled water: labelled sperm appeared in the ejaculate after a mean of 64 days, with a spread of roughly eight days either side and a full range of 42 to 76 days. Anything you start today acts on sperm that have not entered the pipeline yet.

Two other things make early retesting misleading. The same man produces measurably different samples from week to week, so a single change can be normal variation rather than an effect. And a repeat at a different laboratory, using different methods, is not comparable — use the same lab if you can.

If you decide to try a supplement, the practical version is: take it for three months, repeat the analysis at the same lab, and if nothing has changed by six months you have your answer.

The steps below are not alternatives to a supplement so much as the things that change what happens next, because they can identify a cause. An antioxidant cannot.

• Repeat the semen analysis after about three months, at the same laboratory. One abnormal result is not a diagnosis.
• Ask for a physical examination. A varicocele, a small testicular volume or an absent vas deferens are found by examination, not by a capsule.
• Ask about FSH and testosterone. Together with testicular volume they separate a production problem from an obstruction, which changes the treatment entirely.
• Declare any testosterone, anabolic steroid or over-the-counter muscle product. Exogenous testosterone suppresses sperm production, and no supplement counteracts it.
• Stop smoking, reduce heavy drinking, address weight, and treat a genital tract infection if one is found. These act on the same oxidative pathway the supplements are aimed at, and they remove the cause rather than adding a counterweight to it.

Sometimes the honest answer is that the next step is testing, timing or a simpler treatment rather than anything you take. A varicocele, an obstruction, a hormonal problem or exogenous testosterone will not respond to an antioxidant, and finding one of them changes the whole plan.

Being clear about the gaps is part of the answer here, because most of the confidence in this field is borrowed from studies that cannot support it.

• It does not establish that antioxidants increase live birth. It also does not establish that they do not. The trials are too small, and too many failed to report whether anyone had a baby, to rule out a modest effect either way.
• It does not establish a dose, a duration, or which combination. Cochrane compared 20 antioxidants and could not draw conclusions from antioxidant-against-antioxidant comparisons, because too few trials compared the same two things.
• It does not identify which men benefit, if any do. No retrieved trial selected men on a measured oxidative-stress result and then treated them. The prevalence Cochrane itself reports for oxidative stress as a cause of male subfertility spans 25% to 87% of cases — a range that wide is an admission the contribution is not measured in the individual man.
• It does not establish safety at high dose over long periods. Cochrane found more mild gastrointestinal discomfort on antioxidants than placebo across 16 trials and 1,355 men, and no difference in miscarriage across 6 trials and 664 men, both on very low to low certainty evidence. The folic acid and zinc trial found more abdominal pain, nausea and vomiting, and slightly higher sperm DNA fragmentation.
• It does not establish that what is in the capsule matches the label. Cochrane states that most antioxidants are uncontrolled by regulation.

• Do foods or supplements actually improve sperm quality?
• Supplements sold to couples: what the guidelines say
• L-carnitine and sperm: what the trials found
• How to read your semen analysis report
• Does alcohol reduce sperm count?
• New research in male fertility