Story
Maca · Lepidium meyenii
Last updated
In one pass Maca is a root vegetable from the high Andes of South America, a relative of broccoli, cabbage and radish.
Educational content, not medical advice — consult a clinician.
Story path
Chapter 1
What maca is
Maca is a root vegetable from the high Andes of South America, a relative of broccoli, cabbage and radish. Local people first ate it as food; only much later did anyone put it into capsules and sell it.
The part you eat is its swollen stem-root, buried in the soil. The plant stores a whole growing season's sugars and self-defense chemicals in this underground larder to get through the winter, so it is filling, and it also carries the pungent molecules typical of the cabbage family. As for the name Peruvian ginseng, it is pure marketing: true ginseng belongs to a different plant family and is not related to maca at all.
Every later judgment starts from this one sentence: maca has no known phytoestrogens and no established action on the androgen receptor. It is not a key that fits a hormone lock. How it works, if it works at all, is still unknown today.
The part you eat is its swollen stem-root, buried in the soil. The plant stores a whole growing season's sugars and self-defense chemicals in this underground larder to get through the winter, so it is filling, and it also carries the pungent molecules typical of the cabbage family. As for the name Peruvian ginseng, it is pure marketing: true ginseng belongs to a different plant family and is not related to maca at all.
Every later judgment starts from this one sentence: maca has no known phytoestrogens and no established action on the androgen receptor. It is not a key that fits a hormone lock. How it works, if it works at all, is still unknown today.
Background · Where it grows, and why it grew this way
Latin name Lepidium meyenii Walp., family Brassicaceae — same family as broccoli, cabbage, and radish. The edible part is the hypocotyl (a swollen root-stem), not a true root.Almost all of it is grown on the Junín plateau of central Peru, at 4000–4500 m — still able to grow under intense UV, cold, and poor soils. In the Andes it was a staple food first, a supplement later; traditional uses center on energy, stamina, fertility, and adapting to a harsh high-altitude environment.
Maca is not ginseng. Peruvian ginseng is a pure marketing tag: real ginseng is *Panax* (family Araliaceae) and is botanically unrelated to maca. The only things they share are growing underground and being sold as tonics.
Debunk one inference while we are here: the story ads love is that because it can survive such a harsh place, eating it will help you endure too. That inference is broken. A plant withstands intense UV with a chemical shield made inside its own cells; that shield protects the plant's own cells. After you chew it, soak it in stomach acid, let the gut wall pick through it, and let the liver remodel it, what reaches your blood is already not the original molecule — and it will not hand you the skill of tolerating a harsh environment as a side gift. How hardy a plant is in the wild, and what it can do once it is inside your body, are two unrelated questions. This rule applies to every plant sold as an adaptogen, not just maca.
Evidence · Black maca for the brain came from mice
The color phenotypes are real, but black maca for the brain, red maca for the prostate come mostly from rodent experiments:Yellow maca — most common, most commercialRed maca — reduced testosterone-induced prostate enlargement in a rat model (Gasco 2007)Black maca — improved memory in ovariectomized female mice (Rubio 2011). Note: the black maca boosts sperm line often attached to this citation is not from this study — Rubio 2011 used no male animals and measured no reproductive endpointThe honest caveat: these color differences come almost entirely from rodent experiments and are not established in humans. Be wary when black maca specifically for the brain is sold as a human fact
Why can't these results be moved onto people? Three breakpoints are worth keeping:
Those animals were not healthy to begin with. The red-maca arm first injected rats with testosterone to enlarge the prostate, then asked whether maca could press it back; the black-maca arm first removed the ovaries of mice to create a memory impairment, then asked whether maca could patch it. They answer whether an artificially created injury can be pulled back a little, not whether a person who was already fine would get better after eating itThe same word does not mean the same thing underneath. A mouse's memory improvement means it did better on a fixed set of tasks; the clearer head you talk about is a self-felt experience. They share a word; what they measure does not overlapDose and metabolism both fail to match. Animal doses converted by body weight are often far above a person's everyday food amount; and rodent metabolic rate is far from a human's — the concentration curve one gram of powder traces in two species is not the same curve
So the sentence red maca protects the prostate, at this stage, can at most be said as: it has been observed in rats whose enlargement was induced. The steps in the middle, no one has walked for you.
Mechanism · The three classes of molecule in it
Maca's chemical fingerprint is three classes of molecule, and each class maps straight onto a later argument:Macamides and macaenes — fatty-acid amides, largely unique to maca. A fatty-acid amide is a fat tail joined to a nitrogen-bearing head. Your own body makes molecules of this shape; the endocannabinoid anandamide is one of them, and it regulates the volume of mood, appetite, and pain. A macamide looks like it, so some people hypothesize that maca works from the neural end — but looking like is only looking like: even the most maca-favorable review goes no further than saying the two might act by similar mechanisms, based on experiments in cultured nerve cellsGlucosinolates and isothiocyanates — the cruciferous family signature. They sit quiet inside intact cells; the moment you crush them, the plant's own enzyme meets them and cuts them on the spot into pungent isothiocyanates. That sharp maca smell is the product of this cut, and the later discussion of whether raw maca presses on the thyroid also lands entirely on this classSterols + alkaloids (macaridine / β-carbolines / imidazole alkaloids, and others) — low content, almost no human studies, and so far no human evidence tying them to any specific effect
Look at the three together and one thing stands out: none of them looks like a sex hormone. Soy isoflavones are called phytoestrogens not because they are natural, but because their skeleton happens to fit the estrogen-receptor pocket, so the cell really does read estrogen was here. None of maca's three classes has that shape — that is the chemical reason the needles sit still when the hormone trials come up later.
Chapter 2
Does it raise testosterone?
The loudest claim made for maca is natural testosterone booster, hormone balance. The awkward part is that maca's own most solid human trial (Gonzales 2003, 56 healthy men, 12 weeks) was designed specifically to test that sentence, and not one needle moved.
That blood draw measured the finished hormones (testosterone and ), the two orders the pituitary sends to the sex glands (, LH, and , FSH), and an intermediate on the synthesis route. The paper's conclusion reads, word for word: treatment with Maca does not affect serum reproductive hormone levels. Independent teams in other countries have since repeated the test in healthy men and in postmenopausal women, with the same result: self-rated scores moved, hormone readings did not.
One exception has to be written down: a few studies in postmenopausal women reported estradiol rising and FSH falling, but they cluster in one author's set of small trials and no independent team has reproduced them. By the same evidence ruler this story uses throughout, they are discounted. So the conclusion that stands is every trial run by an independent team found no hormone change, not the louder-sounding maca never touches any hormone.
That blood draw measured the finished hormones (testosterone and ), the two orders the pituitary sends to the sex glands (, LH, and , FSH), and an intermediate on the synthesis route. The paper's conclusion reads, word for word: treatment with Maca does not affect serum reproductive hormone levels. Independent teams in other countries have since repeated the test in healthy men and in postmenopausal women, with the same result: self-rated scores moved, hormone readings did not.
One exception has to be written down: a few studies in postmenopausal women reported estradiol rising and FSH falling, but they cluster in one author's set of small trials and no independent team has reproduced them. By the same evidence ruler this story uses throughout, they are discounted. So the conclusion that stands is every trial run by an independent team found no hormone change, not the louder-sounding maca never touches any hormone.
Mechanism · Only three roads raise testosterone
To judge whether something can boost testosterone, you first have to see how testosterone is made.The production line runs like this: deep in your brain the hypothalamus releases a signal on a rhythm; the pituitary receives it and puts (LH) into the blood; LH drifts with the bloodstream to the testes, docks on receptors on the surface of Leydig cells, and only then does the cell start the enzyme chain that remakes cholesterol, step by step, into testosterone. The testosterone that is made then rides the blood back to the brain and presses down the shouting from the hypothalamus and pituitary — a closed-loop machine that talks back to itself. A second order, (FSH), takes the neighboring line and runs the sperm shop.
To make a testosterone reading rise, there are only three paths:
Impersonate the hormone yourself — the molecule looks enough like it to dock straight into the receptor pocket (anabolic steroids and soy isoflavones take this one)Change a processing step — block the enzyme that turns testosterone into , or block the step that turns it into a stronger form, so the same output is read as a different allocationTurn up the upstream order — make the pituitary release more LH, and the testes work more
That is why the decisive trial measured so many items at once. It was not casting a net; it was blocking the roads one by one: testosterone and estradiol cover the finished product, LH and FSH cover the upstream order, and one more item measured an intermediate on the synthesis road — even whether half-finished product piled up or dropped.
The result: every cell sat where it started. Not only did the finished product not move — the upstream shout did not move either. That is much stronger than no effect was measured: it says the signal was never sent at any step. In other words, maca is not too little push; it is not plugged into this line at all.
Take this frame with you. The next time you see any natural hormone booster claim, you can check it yourself: which of the three roads does it take? Has anyone measured the markers on that road? If none of the three has evidence, the claim has nowhere to stand.
Evidence · The decisive trial and the pattern behind it
The decisive experiment:Gonzales 2003 (Journal of Endocrinology) — a 12-week double-blind, placebo-controlled designed specifically to answer whether maca moves hormones56 healthy men, 1.5 g or 3.0 g/dayMeasured testosterone (T) / (E2) / (LH) / (FSH) / prolactin / 17-OH-progesterone all at onceVerbatim conclusion: treatment with Maca does not affect serum reproductive hormone levels
This is not one lone study — it is a repeating pattern:
Gonzales 2002 (healthy men): self-rated desire rose after week 8, but testosterone and estradiol did not move — the paper's whole thesis is that the effect is unrelated to androgensBrooks 2008 (postmenopausal women): mood and sexual-function scores improved, but estradiol / FSH / LH / were all unchanged, and the authors state the effect is unrelated to estrogen or androgen contentStojanovska 2015 (postmenopausal women): depression score + diastolic BP dropped, but estradiol / FSH / SHBG / were unchangedMelnikovova 2015 (healthy men): hormones likewise unchanged
Notice what these four trials share: different populations, different sexes, different primary endpoints, and teams in Peru, Australia, and the Czech Republic — and only the hormone column is empty in every one. A negative result that keeps showing up across different designs and different labs is more persuasive than repeating the same design ten times.
But be honest: not every maca study that measured hormones came back negative
The 2024 review cited here (Ulloa), in its postmenopausal-women section, lists both kinds of result. The one that reported no change is Brooks 2008 (estradiol / FSH / LH / SHBG all unchanged); the ones that reported a change are another set — several papers by Meissner in 2005 and 2006, with estradiol rising and FSH falling.
So the sentence every trial that measured hormones found no movement goes too far. The version that actually stands is:
Every independent team (Brooks in Australia / Stojanovska in Australia + Hong Kong / Melnikovova in the Czech Republic / Gonzales in Peru) left the hormone column emptyThe only reports of hormone movement cluster in one author's set of small trials, and have never been replicated by any independent team
Why is writing it this way stronger? Because it uses the ruler this story comes back to later — when results in one direction cluster in a single source and have no independent replication, their credibility is discounted. That ruler has to be used here in the direction that hurts our own case; otherwise it is only a weapon you pull out when picking at an opponent.
One boundary: the judgment above on that set of studies (same author, small samples, unreplicated) comes from that review's retelling — the originals have not been checked here paper by paper. So the right stance is not to declare them wrong, but: until there is independent replication, they cannot carry the hormone-balance selling point.
Myth · The receptor is a lock, the molecule a key
Why would that be? — because there is no plausible molecular mechanism:Maca contains no phytoestrogen (unlike soy isoflavones)Maca does not agonize the androgen receptor (unlike anabolic steroids)It does not inhibit aromatase, and is not a 5α-reductase inhibitorThat a macamide looks like anandamide and may work from the neural end is still a hypothesis. Even the most maca-favorable 2024 review (Ulloa) uses only wording like might act by similar mechanisms. What it flatly labels unconfirmed (this hypothesis has not been confirmed) is an older idea — maca alkaloids balance hormones via the axis. Notice: what that review directly knocks down is the hormone-balance selling point itself
Think of the receptor as a lock and the molecule as a key, and the four lines above become easy.
Soy isoflavones are called phytoestrogens not because they are natural, but because their three-dimensional shape happens to fit the estrogen-receptor pocket and turn the lock a little — so the cell really does read estrogen was here. Anabolic steroids are more direct: they were remade from testosterone in the first place, and once they slot into the androgen receptor they can open a whole set of muscle-building genes. These two classes can move hormones because they have a matching key shape.
None of maca's three classes of molecule has either shape. It also does not touch the processing shop: it does not block aromatase (the enzyme that turns testosterone into ), and it is not a drug that blocks testosterone from being turned into a stronger form. Three locks, and not one matching key.
So hormone readings sitting still should not surprise you — it is exactly the result chemistry should have predicted. What actually needs explaining is the other half: if nothing happened chemically, why did some people's self-rated scores move? That half of the answer is in the chapter on what the human trials found, and it is more honest — and more interesting — than the marketing story.
Chapter 3
What human trials found
If hormones do not move, what is maca actually good for? Line up the human trials by who scores the endpoint, and a strikingly neat pattern jumps out.
Whenever you score it yourself (desire, mood, well-being), maca sometimes comes out a little ahead of placebo. Whenever an instrument reading decides (time-trial results, sperm, hormones), it has not beaten placebo. The cleanest example comes from the same group of cyclists: in one trial, self-rated desire beat placebo while time-trial performance did not.
Three top-level reviews say the same thing in unison: the evidence is limited, the quality too low to draw a firm conclusion.
Take one judgment tool with you first: the next time you see good news about any supplement, the first question is not how much it improved but who is scoring this endpoint, and what it is being compared with.
Whenever you score it yourself (desire, mood, well-being), maca sometimes comes out a little ahead of placebo. Whenever an instrument reading decides (time-trial results, sperm, hormones), it has not beaten placebo. The cleanest example comes from the same group of cyclists: in one trial, self-rated desire beat placebo while time-trial performance did not.
Three top-level reviews say the same thing in unison: the evidence is limited, the quality too low to draw a firm conclusion.
Take one judgment tool with you first: the next time you see good news about any supplement, the first question is not how much it improved but who is scoring this endpoint, and what it is being compared with.
Evidence · Trials on sex, mood and exercise
Libido and sexual function (certainty of evidence: low — subjective endpoints and small samples):Gonzales 2002 (healthy men, 12 weeks): self-rated desire beat placebo after week 8. But the endpoint is subjective self-report, the sample is small, and it comes from a single labTop-tier review Shin 2010 (BMC Complement Altern Med): found only 4 , none reporting allocation concealment or adverse events, concluding limited evidence, insufficient for a firm conclusion — the most honest one-liner
Erectile dysfunction (ED; certainty of evidence: low — only mild ED was ever tested):
Zenico 2009: 50 men with mild ED, 2400 mg maca dry extract (note: an extract, not the 1.5–3.5 g of dry powder used elsewhere). Both maca and placebo improved IIEF-5, but maca's increment was 1.6 ± 1.1 points vs placebo 0.5 ± 0.6 points, P < 0.001 — the cleanest statistical separation in maca's entire evidence base; do not read it as a trial the placebo response swallowedIts real weaknesses sit elsewhere: only 50 men, only mild ED, and every endpoint is a subjective scale; and IIEF-5 runs to 25 points, so a 1.1-point gap is statistically clear, but how obvious it is in lived experience is another matterLee 2023 review: only 2 RCTs (n=79), one positive, one null; pooled, the result is significant (MD 1.13, 95% 0.64 to 1.61) — we do not hide that number, but read it with its base: two trials, 79 men, mild ED only, which is why the authors themselves still conclude limited evidence, insufficient for a firm conclusion. Moderate / severe ED has never been studied
Sexual dysfunction caused by and other antidepressants (certainty of evidence: low — a detail that is often misread):
Dording 2008: a dose-finding study with no placebo arm. The high-dose group improved within-group, but no control = cannot rule out placebo or regression to the meanDording 2015: this is the real placebo-controlled test (parallel design, ~42 women). Primary endpoints were largely null, with benefit only in a small postmenopausal subgroup — the honest read is a largely negative trial dressed in a subgroup-positive headline
Athletic performance (certainty of evidence: low — did not win the one placebo-controlled comparison):
Stone 2009: 8 cyclists; the 40 km time-trial beat baseline (P=0.01) but not placebo (P>0.05). The endurance claim lost its only direct contest (self-rated desire, again, did beat placebo)
Taken together, maca occasionally shows a weak signal on soft, subjective libido and mood endpoints; once the endpoint is objectively measured (endurance, hormones, sperm), it has not beaten placebo.
Evidence · Beating baseline is not beating placebo
The maca cycling trial hides a methods lesson worth learning on its own — the same data inside it produced two opposite headlines.Those eight riders, after taking maca, really were faster than before they started, and the difference still stood statistically. If the paper had stopped there, an ad could print measured endurance gain. But this study also carried a placebo arm: the same course, the same shape of powder, nothing inside. Against that arm, maca did not win.
Why can two comparisons split into opposite conclusions? Because faster than before already has a pile of forces pushing that have nothing to do with maca:
You got familiar with the course and the equipment — most people are faster the second time they take the same testRegression to the mean — the people willing to join a trial are often the ones whose current state is off (people doing well do not bother), and a person's state drifts back toward the middle on its ownYou know you are being watched — effort and pacing strategy both changeThe self-rated column is even more open — you paid, you put in weeks, you hoped it would work, and the score walks up
The whole point of a placebo arm is to soak up those forces in one pass. Both arms went through familiarization, both regressed to the mean, both were watched, both were expecting; the slice that is still higher after that is this thing's own effect.
So the next time you see after taking X, the marker improved, the first question is not how much it improved, but compared with whom. A claim that only reports better than baseline is not missing precision — it is missing the half of the information that can overturn the conclusion.
That also explains the shape that keeps showing up in maca's evidence: objective endpoints lose their edge once a control is added, while subjective endpoints, pushed by expectation, can sometimes hold a thin gap.
Evidence · Why the evidence is so soft
Maca's 'soft' evidence has two structural causes worth unpacking separately:① The most positive results cluster in one research group + a national industry interest
State it precisely first: maca's human trials are not almost all Peruvian. Zenico is Italian, Dording American, Stone British, Brooks and Stojanovska Australian, Melnikovova Czech, Shin and Lee's reviews Korean — independent teams are well representedWhat actually clusters are the most positive findings: libido, sperm, and the color phenotypes, where Gonzales / Gasco / Rubio recur, all from one Peruvian research groupAnd the independent teams are the ones that came back null or trend-level (Stone 2009 endurance, Dording 2015 , Melnikovova 2015 sperm). That distribution is itself the information — it says more than a blanket 'it's all Peruvian'Maca is a national export crop of Peru, with a clear industry / national interest attached. This doesn't mean fraud, but in evidence terms it's an independence concern: when positive results are concentrated in a few interested parties and not replicated independently, their credibility is discountedThis ruler has to cut both ways. The studies mentioned with the hormone trials that reported rising and falling also cluster in one author's hands with no independent replication — same ruler, same discount. A standard you only reach for when it favours you is not a standardWhat would actually be reassuring is a third-party, large-sample, long-term, pre-registered trial — maca has none
② On subjective libido / mood endpoints, the placebo effect is naturally huge
'Desire', 'well-being', and 'mood' are subjective experiences highly shaped by expectationA ready example is the Barton 2013 ginseng trial (in patients with cancer-related fatigue): its placebo arm improved by 8.2 fatigue points at 4 weeks and 10.3 at 8 weeks — half the participants took nothing and still got substantially betterDo not use Zenico 2009 for this point: there the placebo arm gained only 0.5 points against maca's 1.6, P < 0.001 — that is a trial the placebo response failed to swallowIf you believe in an ancient Andean superfood + paid money + expect to feel better → your self-rated scores rise. That improvement is a real experience, but it belongs to your expectation, not maca's pharmacologyThe only way to separate the two is placebo control + objective measurement — and maca is remarkably consistent at 'getting weaker / disappearing once a control is added' (Stone 2009 endurance, Dording 2015 SSRI, Melnikovova 2015 sperm)
Put the two together: maca's entire evidence base ≈ a dozen small trials of n=8–56, lasting up to 4 months, often single-lab, and three top-tier reviews (Shin 2010 / Lee 2011 / Lee 2023) say in unison: 'limited evidence, quality too low, can't conclude'. This doesn't mean 'it definitely does nothing' — it means anyone selling it as definitively effective is reaching beyond what the evidence allows.
lee-2023-maca-erectile-dysfunction-reviewdording-2008-maca-ssri-dose-findingdording-2015-maca-ssri-womenbarton-2013-jnci-ginseng-fatigue
Chapter 4
Menopause and fertility evidence
Two groups pin the most hope on maca: women having a hard time with menopause, and men who want to improve their fertility. These two sets of data are worth a close look: maca did give something, but the way it gave it shows once again that hormones are not involved.
On the menopause side, a few small trials did see anxiety, depression and sexual-function scores move in the right direction; yet the authors all pointed out that estrogen, and the orders the pituitary sends out, did not move a tick. In other words, the body never received a signal that estrogen is back.
The male-fertility side is weaker: semen measures did rise in places, but either there was no placebo control or the change was too small to hold up statistically. It stops at a trend, short of proof.
Even in the settings where it is most likely to help, maca's benefit is subjective, at the level of symptoms, and uncoupled from hormones. It cannot fill a hormone gap, and there is no evidence that it treats infertility.
On the menopause side, a few small trials did see anxiety, depression and sexual-function scores move in the right direction; yet the authors all pointed out that estrogen, and the orders the pituitary sends out, did not move a tick. In other words, the body never received a signal that estrogen is back.
The male-fertility side is weaker: semen measures did rise in places, but either there was no placebo control or the change was too small to hold up statistically. It stops at a trend, short of proof.
Even in the settings where it is most likely to help, maca's benefit is subjective, at the level of symptoms, and uncoupled from hormones. It cannot fill a hormone gap, and there is no evidence that it treats infertility.
Clinical · What three menopause studies found
Menopause / perimenopause (weak signal on mood / symptoms, but not hormone replacement):Brooks 2008 (crossover , 14 postmenopausal women, 3.5 g/day): anxiety and depression scores + sexual dysfunction improved beyond placebo — but the authors specifically stress / / / all unchanged, the effect unrelated to estrogen or androgenStojanovska 2015 (crossover RCT, 29 postmenopausal women, 3.3 g/day): depression score + diastolic BP dropped, hormones again unchangedLee 2011 review (Maturitas): 4 RCTs were consistently favorable on the Kupperman / Greene menopausal scales, but limited evidence, methodological quality too low, safety not establishedHow to read it: this is a symptomatic / mood weak benefit, not hormone replacement therapy (/MHT). Someone with a genuine hormone gap and severe symptoms should be discussing menopausal hormone therapy (MHT) with a doctor, not using maca as a substitute
Mechanism · Which stretch of supply breaks in menopause
To see how clean the pairing scores moved, hormones did not really is, you first have to know which stretch actually broke in menopause.The number of follicles in the ovary is finite; once they are used up they no longer answer the upstream shout. So estrogen supply cuts off at the ovarian end; the hypothalamus and pituitary do not know the warehouse is empty — they only know the reply is gone — and they shout the order louder each time. That is the typical postmenopausal pair: low, high — one low and one high is itself the signature of this closed-loop machine spinning empty.
The next step is the key. If something really did fill the estrogen gap (even part of it), the pituitary would notice before you do: it got a reply, the shouting would drop, and the FSH reading would walk down with it. That is the fingerprint hormone replacement leaves on a lab slip.
And in the two trials run by independent teams — Brooks 2008 and Stojanovska 2015 — maca was taken for several weeks and FSH did not move a hair. This line is harder than the chemical inference it contains no phytoestrogen: the chemical inference says it should not work, while an unmoved FSH is something measured directly in living people — these two groups' brains never received even a sliver of estrogen signal.
A counterexample that has to be written down: another set of postmenopausal studies reported the opposite direction (FSH down, estradiol up). They cluster in one author's hands, the samples are all small, and they have never been independently replicated (details in the chapter on whether it raises testosterone). So the right reading of this stretch is not FSH can never move, but: in independent hands, it has not moved.
So what maca can do in menopause was, from the start, confined to another track: not splicing the broken stretch of supply back on, but making a marginal difference to how you feel about it. That also explains the shape of its report card — a weak signal on mood scales, a zero in the hormone column. As for symptoms driven directly by the estrogen gap, such as hot flashes, whether that small scale improvement is at the level of feeling or a real drop in episodes is something the existing small trials cannot separate.
A rule you can run yourself: anything sold as natural hormone balance can be tested on this closed-loop machine — if something really went in, the upstream shout must weaken; if the upstream reading does not move a hair, nothing went in.
Clinical · What male-fertility claims still lack
Male fertility / sperm (only at the trend level, not the proven level):Gonzales 2001 (n=9, no placebo control, 4 months): semen volume, sperm count, and motility rose, hormones unchanged — but open-label + single-lab + tiny sample = hypothesis-generating onlyMelnikovova 2015 (placebo-controlled, n=20, 12 weeks): sperm concentration and motility showed only non-significant upward trends, hormones unchanged — underpowered to prove a fertility benefitHow to read it: there is a signal, but no adequate controlled trial confirms it. Infertility is a real problem for reproductive medicine; maca cannot replace proper evaluation and treatment
Why is the sperm endpoint especially good at handing out numbers that look good and do not stand? Three reasons:
The sperm you measure today was not made today. A spermatogonium has to walk the whole production line — divide, change shape, grow a tail, then learn to swim in the epididymis — before it becomes a mature sperm that can be counted, and that line itself takes several months. In a trial that only runs a dozen-plus weeks, a substantial share of what you measure is still old inventory that was already in the pipe before the trial startedThe same person sending two samples will already differ a lot. Semen markers swing hard with days of abstinence, a recent fever, season, and whether the sample is complete; that natural noise often covers whatever effect a supplement might have. So on this endpoint, a before-and-after with no control arm carries almost no informationWhen the sample is small, a negative is not 'no effect' — it is 'cannot see'. The non-significant trend in that placebo-controlled trial is exactly this state — it did not have the power to answer the question, rather than answering no
Put the three together: the existing data let you say there is a signal worth checking again; they do not let you say it raised fertility. What sits between those two sentences is a trial that is long enough, large enough, and controlled — and it has not been run.
Chapter 5
Is it safe, and should you take it
The good news: eaten as food, maca is fairly safe in the short term. The leading database on drug-induced liver injury puts its likelihood of causing liver damage in the lowest tier, with only one reported case worldwide so far; in short-term trials, side effects were mostly mild, passing stomach upset and headache.
The two things that really call for care are not about whether it is toxic but about which jar is in your hand:
Raw, in large amounts, in someone already short of iodine: the pungent molecules of the cabbage family go and compete with iodine for the thyroid's entrance. The Andean habit of cooking maca first, and the gelatinized form of many commercial powders, both take this risk apartWhere it was grown: soils in some Andean mining areas carry cadmium and lead, and what you eat is exactly the storage organ that sits in that soil
Pregnancy and breastfeeding: there are not enough human data, so the usual practice is to avoid it.
The two things that really call for care are not about whether it is toxic but about which jar is in your hand:
Raw, in large amounts, in someone already short of iodine: the pungent molecules of the cabbage family go and compete with iodine for the thyroid's entrance. The Andean habit of cooking maca first, and the gelatinized form of many commercial powders, both take this risk apartWhere it was grown: soils in some Andean mining areas carry cadmium and lead, and what you eat is exactly the storage organ that sits in that soil
Pregnancy and breastfeeding: there are not enough human data, so the usual practice is to avoid it.
Safety · Short-term data and the traditional dose
LiverTox (NIH's authoritative drug-induced liver injury database) rates maca's hepatotoxicity likelihood as E (unlikely), with only 1 case reported worldwide (and that was a maca medicinal liquor, 2017)Short-term (≤12 weeks) adverse effects are mostly mild, passing stomach and gut upset and headacheTraditional dietary dose 1.5–3.5 g/day (dried powder)Read those three lines together and they say something quite specific: at a food-level amount and a timescale of a few months, maca has left no signal worth worrying over. That is also the real difference between it and many extract-type supplements — it is a crop people have eaten as food for a long time, not a single compound concentrated out of a plant.
But short-term safety is not long-term safety; those two sentences are not the same thing in the evidence: the first is held up by a set of trials lasting weeks to a dozen-plus weeks, and the second has no data at all on maca. So the two remaining risks, one (thyroid) that may only show if the time is long enough and one (heavy metals) that only settles the bill after years of accumulation, will not appear on a dozen-week adverse-event table.
Mechanism · why raw maca crowds the thyroid
To make hormone, the thyroid's first step is to move iodine from the blood into thyroid cells. The cell membrane has a dedicated door (the sodium–iodide symporter) that recognizes an iodide ion by size and charge.Here is the problem: after a cruciferous glucosinolate is crushed and cut by the enzyme, one class of product is a small ion called thiocyanate, whose size and charge happen to sit close to iodine. So it goes and stands in that door's queue and crowds iodine out. Less iodine gets in → not enough raw material to make thyroid hormone → thyroid hormone in blood walks down → the pituitary notices and shouts louder (on the lab slip that is a rise in ) → the thyroid is shouted into growing. That is where the word goiter comes from.
Notice this is a seat-stealing mechanism, not a toxicity mechanism. The result of stealing a seat depends on how much is on each side: when iodine is plentiful, thiocyanate cannot out-queue it; when iodine is already scarce, the crowd-out can actually move something. So the real risk condition is three things holding at once: eaten raw (the plant's own enzyme still alive) + high dose + already iodine-deficient. Drop any one and the chain does not connect.
That also lets two practices that look like they are only about taste and mouthfeel show their safety meaning:
Andean tradition boils maca first: heat knocks that enzyme out and also destroys a share of the glucosinolate outright. The common claim that boiling and discarding the water removes most of it comes from other cruciferous vegetables, not from measurements on maca. The conservative statement is that boiling and discarding the water removes a good share; how much, there are no data on maca itselfCommercial forms are often gelatinized — also heat-treated, and digestibility improves as a side effect
The honest boundary: there is no maca-specific long-term thyroid-outcome trial, so this is a precaution, not a documented harm. The broader cruciferous evidence is reassuring — cooked is basically fine, and ordinary servings of broccoli or cabbage have never been the cause of a thyroid problem.
A step you can push further yourself: the same spoonful of powder is almost no issue for someone whose iodine intake is adequate; the risk only becomes discussable for someone whose iodine is already tight and who swallows large amounts of raw powder every day. Dose, form, and your own iodine status decide the result together — a more useful question than is it toxic.
Safety · Why heavy metals favor a root-stem
Mendoza 2021 (Toxicology Reports): across three mining-affected districts of Junín, Peru, maca measured cadmium 0.32 ± 0.23 mg/kg + lead 0.20 ± 0.12 mg/kg, both above the FAO/WHO limits. Modeled cancer risk for arsenic and cadmium exceeded the 1×10⁻⁶ tolerable line in both children and adults; in the Ondores district, children's arsenic risk also exceeded 1×10⁻⁴.The same paper's other half belongs here too: it calculated a bioconcentration factor below 1, an estimated daily intake of each metal below the oral reference dose, and both the hazard quotient (HQ) and the hazard index (HI) below 1 — from which the authors judge non-cancer harm unlikely.
Both halves together are the full picture, and they actually make this section's real point sharper: the near-term non-cancer risk is low; what the model pushes over the threshold is the long-run cancer risk — because cadmium is a cumulative toxin, and its bill is not settled today. The next stretch explains why.
Why this crop in particular? Two things landed together:
What you eat is the storage organ buried in the soil. A leafy vegetable's edible part grows in air; maca's hypocotyl sits entirely in the soil solution, and is exactly where the plant hoards things — whatever is in the soil, it has a chance to hoardThe root's transport channels cannot tell cadmium apart. Plants need metals such as zinc, so they fit dedicated transporters on root cells; cadmium's chemistry is too close to theirs, and those channels bring it in as a hitchhiker, then send it up with the water stream into storage tissue
Why does cadmium's bill take many years to settle? Because it does not strike on the spot the way an acute toxin does. Cadmium that enters the body is mainly held by the liver and kidney; after binding a small metal-bundling protein it rides the blood to the kidney; the tubule takes the whole complex back into the cell, the protein is taken apart, and the cadmium stays. It leaves extremely slowly, so it accumulates year after year in the renal cortex. By the time tubular cells are damaged, the sign is small proteins that should have been reclaimed starting to leak into urine — and by then the accumulation is already done.
So the countermeasure is not dose, it is provenance. This is a product-quality and origin issue, not a pharmacology one: two jars both labeled maca powder can differ widely in cadmium. Choose a product with third-party heavy-metal testing (cadmium, lead, and arsenic not detected or far below limits) + a stated origin + a reputable brand.
This rule reaches far beyond maca. Any crop whose underground part is eaten — roots, tubers, swollen stems — is more exposed to soil contamination than one whose above-ground part is eaten; and anything eaten every day, for many years, is worth a little extra work to confirm the source. The longer you eat it, the more provenance matters than dose.
In practice · Should you take it
Should you take it:Want to boost testosterone or balance hormones → maca cannot (as the chapter on the hormone trials shows); do not buy it for this reasonMild menopausal symptoms + want a gentle, low-risk adjunct → a 4–8 week trial is reasonable, but first get sleep, exercise, alcohol limits and stress management in place; with severe symptoms, discuss menopausal hormone therapy () with a doctorInfertility / moderate-severe ED → see reproductive medicine / urology; maca is not a treatmentIf you decide to try: 1.5–3.5 g/day gelatinized powder, pick a third-party heavy-metal-tested product, evaluate subjectively at 4–8 weeks, stop if you feel nothing — do not do faith-based daily use forever
Why give yourself a stop date? Because the trials have already made the shape of maca's effect clear: it lands on the subjective feeling end, and subjective feeling is the easiest thing for expectation to push. You paid, you waited weeks, and at the moment of scoring the brain will automatically read those in. Set a date, set one concrete feeling you care about (not a fuzzy phrase like overall state), and on that date answer honestly once — that is the closest thing to a placebo control you can build for yourself.
In the end, maca is an Andean, basically safe food with a weak, possibly real, hormone-independent subjective effect on libido and mood. It is not a hormone booster, not a proven fertility or ED treatment, and not a supplement that improves athletic performance. Anyone selling it as natural testosterone, hormone balance is refuted by maca's own foundational trials, and that is the rule to keep for every adaptogen: ask what the evidence says before you ask what the marketing says.
References · 15
- Ulloa del Carpio, N., Alvarado-Corella, D., Quiñones-Laveriano, D. M., Araya-Sibaja, A., Vega-Baudrit, J., Monagas-Juan, M., Navarro-Hoyos, M., & Villar-López, M. (2024). Exploring the chemical and pharmacological variability of Lepidium meyenii: A comprehensive review of the effects of maca. Frontiers in Pharmacology, 15, 1360422. 10.3389/fphar.2024.1360422
- Gasco, M., Villegas, L., Yucra, S., Rubio, J., & Gonzales, G. F. (2007). Dose–response effect of red maca (Lepidium meyenii) on benign prostatic hyperplasia induced by testosterone enanthate. Phytomedicine, 14(7–8), 460–464. 10.1016/j.phymed.2006.12.003
- Rubio, J., Qiong, W., Liu, X., Jiang, Z., Dang, H., Chen, S.-L., & Gonzales, G. F. (2011). Aqueous extract of black maca (Lepidium meyenii) on memory impairment induced by ovariectomy in mice. Evidence-Based Complementary and Alternative Medicine, 2011, 253958. 10.1093/ecam/nen063
- National Institute of Diabetes and Digestive and Kidney Diseases. (2019). Maca. In LiverTox: Clinical and Research Information on Drug-Induced Liver Injury. Bethesda, MD: NIDDK. Bookshelf ID NBK548552. www.ncbi.nlm.nih.gov/books/NBK548552
- Gonzales, G. F., Córdova, A., Vega, K., Chung, A., Villena, A., & Góñez, C. (2003). Effect of Lepidium meyenii (Maca), a root with aphrodisiac and fertility-enhancing properties, on serum reproductive hormone levels in adult healthy men. Journal of Endocrinology, 176(1), 163–168. 12-week double-blind, placebo-controlled, randomized parallel trial in healthy men aged 21-56: gelatinized maca 1.5 g or 3.0 g/day vs placebo; LH, FSH, prolactin, 17-alpha hydroxyprogesterone, testosterone and 17-beta estradiol measured at 0, 2, 4, 8 and 12 weeks. Compared with placebo, maca had no effect on any hormone, and none changed over time. The abstract gives no sample size (abstract, PMID 12525260). 10.1677/joe.0.1760163
- Gonzales, G. F., Córdova, A., Vega, K., Chung, A., Villena, A., Góñez, C., & Castillo, S. (2002). Effect of Lepidium meyenii (MACA) on sexual desire and its absent relationship with serum testosterone levels in adult healthy men. Andrologia, 34(6), 367–372. 10.1046/j.1439-0272.2002.00519.x
- Shin, B.-C., Lee, M. S., Yang, E. J., Lim, H.-S., & Ernst, E. (2010). Maca (L. meyenii) for improving sexual function: A systematic review. BMC Complementary and Alternative Medicine, 10, 44. 10.1186/1472-6882-10-44
- Zenico, T., Cicero, A. F. G., Valmorri, L., Mercuriali, M., & Bercovich, E. (2009). Subjective effects of Lepidium meyenii (Maca) extract on well-being and sexual performances in patients with mild erectile dysfunction: A randomised, double-blind clinical trial. Andrologia, 41(2), 95–99. 10.1111/j.1439-0272.2008.00892.x
- Stone, M., Ibarra, A., Roller, M., Zangara, A., & Stevenson, E. (2009). A pilot investigation into the effect of maca supplementation on physical activity and sexual desire in sportsmen. Journal of Ethnopharmacology, 126(3), 574–576. Pilot: 8 trained male cyclists, randomized crossover, 14 days each of maca extract and placebo, with a 40-km time trial before and after. Maca improved 40-km time vs its own baseline (P = 0.01) but NOT vs placebo after supplementation (P > 0.05); self-rated sexual desire improved vs baseline (P = 0.01) and vs placebo (P = 0.03) (abstract, PMID 19781622). 10.1016/j.jep.2009.09.012
- Brooks, N. A., Wilcox, G., Walker, K. Z., Ashton, J. F., Cox, M. B., & Stojanovska, L. (2008). Beneficial effects of Lepidium meyenii (Maca) on psychological symptoms and measures of sexual dysfunction in postmenopausal women are not related to estrogen or androgen content. Menopause, 15(6), 1157–1162. 10.1097/gme.0b013e3181732953
- Stojanovska, L., Law, C., Lai, B., Chung, T., Nelson, K., Day, S., Apostolopoulos, V., & Haines, C. (2015). Maca reduces blood pressure and depression, in a pilot study in postmenopausal women. Climacteric, 18(1), 69–78. 10.3109/13697137.2014.929649
- Lee, M. S., Shin, B.-C., Yang, E. J., Lim, H.-J., & Ernst, E. (2011). Maca (Lepidium meyenii) for treatment of menopausal symptoms: A systematic review. Maturitas, 70(3), 227–233. 10.1016/j.maturitas.2011.07.017
- Gonzales, G. F., Córdova, A., Gonzales, C., Chung, A., Vega, K., & Villena, A. (2001). Lepidium meyenii (Maca) improved semen parameters in adult men. Asian Journal of Andrology, 3(4), 301–303. pubmed.ncbi.nlm.nih.gov/11753476
- Melnikovova, I., Fait, T., Kolarova, M., Fernandez, E. C., & Milella, L. (2015). Effect of Lepidium meyenii Walp. on semen parameters and serum hormone levels in healthy adult men: A double-blind, randomized, placebo-controlled pilot study. Evidence-Based Complementary and Alternative Medicine, 2015, 324369. 10.1155/2015/324369
- Orellana Mendoza, E., Cuadrado, W., Yallico, L., Zárate, R., Quispe-Melgar, H. R., Limaymanta, C. H., Sarapura, V., & Bao-Cóndor, D. (2021). Heavy metals in soils and edible tissues of Lepidium meyenii (maca) and health risk assessment in areas influenced by mining activity in the Central region of Peru. Toxicology Reports, 8, 1461–1470. 10.1016/j.toxrep.2021.07.016