What Does Inositol Really Do?

Ask what inositol does, and the answer must start with a correction about the name. The word does not describe one substance. Instead, it describes a family of nine closely related molecules. Two of them dominate human biology. They are chemically almost identical, yet they behave differently, and at times they pull in opposite directions.

That distinction sounds like a technicality. In practice, it decides which research applies to which product on a shop shelf. Interest has moved quickly. Clinicians began paying serious attention only over the past twenty years. Consequently, much of the clinical evidence is recent, and much of it remains unsettled.

Women with polycystic ovary syndrome (a common hormonal condition affecting the ovaries) put the subject on the clinical agenda themselves. They prioritised it internationally while the 2023 evidence-based guideline for the condition was being developed.

Many fertility clinicians now prescribe it before in vitro fertilisation, or to encourage ovulation. Even so, no high-quality evidence supports that practice. A review of off-label treatments found only limited support, because the trials were small and few included a placebo group.

Meanwhile, the compound is available for purchase online. That places it outside the usual route by which treatments reach patients. Many people also arrive at it looking for an alternative to metformin, whose gastrointestinal side effects some women cannot tolerate.

Here the subject becomes genuinely difficult. Pooled analyses frequently report favourable results. By contrast, the international guideline describes the evidence as indeterminate. Both conclusions are drawn from the same body of research. The reason is not carelessness on either side. Rather, different questions produce different answers. A trial can show that a hormone reading improved without showing that a life did.

Research on inositol in newborn medicine shows how far conclusions can travel. Trials in 1986 and 1992 suggested it improved survival in premature babies. A further review in 2015 agreed. Then a large multicentre trial found no benefit.

Part of the difficulty is reach. Laboratory work has reported anti-inflammatory and insulin-sensitising properties across a variety of experimental models. Findings on disturbed metabolism in insulin resistance then prompted investigation in fertility, egg development, tissue repair and fat metabolism.

A molecule involved in that many processes attracts questions from many directions at once. Some of those questions have good answers. Others have almost none, and the difference is rarely made obvious to the person reading a label.

The question is not whether it works. It is what it does, for whom, in which form, and measured against what. Improved laboratory readings and improved lives are not the same achievement. The distance between those two things is where most of the confusion lies.

What is inositol

A German chemist pulled it out of muscle tissue in 1850. He named it from the Greek word for sinew; to that, he added the endings used for sugars and alcohols. He also noted that it tasted sweet.

That origin story is a fair description of the molecule. It is a sugar alcohol: a six-sided carbon ring carrying six hydroxyl groups. Rearranging those groups produces different versions of the same compound, called isomers. Nine are possible, and eight of them occur in nature.

Even so, only three turn up as major forms in mammals. Myo-inositol is by far the most abundant. Scyllo-inositol comes second, and together the two account for more than 90% of the total in human cells. D-chiro-inositol makes up a small fraction, and the remaining forms appear only in traces.

Almost everything sold, studied or argued about as inositol concerns myo-inositol and D-chiro-inositol. Inositol reaches the body by two routes, and the smaller one is the one people think of first.

Food supplies roughly 1 g a day. Estimates based on phytate-rich foods put Western intakes lower still, at 500 to 700 mg. Figures run higher in parts of Africa and Asia. A typical 2500 kcal diet contains around 900 mg, with over half bound to fats rather than free.

Absorption is close to total. About 99.8% of what is swallowed crosses the gut wall. Once in the blood, it does not linger: the circulating pool turns over with a half-life of roughly 22 minutes.

Meanwhile, the larger route is manufacture. The body builds about 4 g a day from glucose, mostly in the kidneys. The liver, brain and reproductive organs contribute as well. In other words, your own tissues out-produce your diet by around four to one.

That single fact separates this compound from most things sold in a tub. It is not a nutrient the body waits passively to receive. Where it does appear in food, the pattern surprises people. Measuring 487 individual foods by gas-liquid chromatography showed that seeds carry the most: beans, grains and nuts. Cantaloupe melon and the citrus fruits, lemons excepted, are extraordinarily high.

Milk products sit low. Among vegetables, beans lead and leafy vegetables trail. Wholegrain breads carry more than refined ones, and oats and bran more than other cereals. Fresh produce, too, beats canned or frozen.

Brand matters too, which is an uncomfortable detail for anyone trying to eat to a target. Six brands of frozen orange juice ranged from 1.4 to 2.7 mg per gram. Across all 487 foods, repeated measurements varied by about 21%.

One large dietary source barely counts. Phytic acid is the storage form found in plant cell walls. After the free version, it is the most abundant form in food. However, its heavy negative charge means bacterial enzymes must break it down before anything can enter the bloodstream. Among animals, therefore, only ruminants manage that reliably.

D-chiro-inositol arrives differently again. Most of it comes as D-pinitol, a methylated version that the gut’s acidic conditions convert back. Legumes are the main source, and carob pods are unusually rich, at 10 to 80 g per kilogram.

None of this makes it a vitamin, although it spent decades being treated as one. The compound was long grouped with the B vitamins, and it is genuinely essential for cells to grow. For example, cells deprived of it cannot multiply in culture. That classification was abandoned once it became clear the body manufactures several grams a day.

A vitamin, by definition, is something you must obtain from outside. This does not qualify. Inositol is now classed as a nonvitamin micronutrient. It sits with carnitine and choline as conditionally essential. That means the body normally copes on its own, though it may not under strain. The label sometimes applied, vitamin B8, survives from the older view rather than the current one.

The two main forms are related by a one-way conversion, and that detail explains much of what follows. Insulin drives tissue-specific enzymes that turn myo-inositol into D-chiro-inositol. However, the reaction runs in only one direction. As a result, each organ maintains its own balance between the two, set by how much it converts. Glycogen-storing tissues such as liver and muscle hold relatively more D-chiro-inositol. Elsewhere, myo-inositol predominates.

In the bloodstream, the difference is stark. Myo-inositol circulates at roughly 26.8 to 43.0 micromolar. D-chiro-inositol normally sits below 100 nanomolar, hundreds of times lower.

Concentrations elsewhere are higher still, and not evenly spread. Tissues that burn a lot of glucose hold the most, particularly the brain, the heart and the ovary. Levels in the brain run 50 to 200 times those in blood plasma. In the fluid around the brain and spine, they run about seven times higher.

The compound is also built into the structure of cells rather than simply floating through them. It forms the backbone of phosphatidylinositol, a fat that makes up 10 to 15% of the membranes around mammalian cells. Cells pull it in through dedicated transporter proteins found in the kidney, brain, liver, pancreas, placenta, heart, and skeletal muscle.

Human breast milk is very rich in it, at more than 1200 millimoles per litre. That points to a real requirement in the weeks after birth. Likewise, it is concentrated in the umbilical artery of the developing embryo. It also forms part of surfactant, which keeps the air sacs of the lungs open.

Myo-inositol and D-chiro-inositol are chemically very similar. They are not, however, interchangeable, and in several settings they do opposing work.

How does it work in the body

Nothing in this section proves that supplements help anyone. It explains why researchers thought they might, which is a different and weaker claim. The compound does not act on the body from outside. Instead, it works as a messenger inside cells.

Hormones deliver their instructions to the cell surface. Something then has to carry that instruction inward. Myo-inositol built into the cell membrane is the raw material for inositol triphosphate. That molecule is one of the main couriers. It relays signals for several hormones, including insulin, thyroid-stimulating hormone and follicle-stimulating hormone.

That last one matters later, because follicle-stimulating hormone is the signal that ripens an egg.

On the metabolic side, both forms interact with insulin, though not identically. Myo-inositol moves GLUT4 glucose transporters to the cell surface, so the cell takes up more glucose. D-chiro-inositol stimulates glycogen synthase, the enzyme that stores glucose away, and boosts several relay proteins in the insulin chain.

Between them, they steer glucose in different directions. D-chiro-inositol pushes it towards storage. Myo-inositol pushes it towards being burned. Both, by reducing how hard insulin has to work, can behave as insulin sensitisers.

The evidence that they act as genuine insulin messengers is decades old. Insulin stimulation makes cell membranes release sugar-like molecules containing each form, and those molecules mimic insulin’s effects. When given to diabetic animals, they lowered blood sugar in a dose-dependent way.

Then, on the reproductive side, the two forms visibly diverge. In the ovary, myo-inositol acts as the second messenger for follicle-stimulating hormone. It supports the growth of the cells surrounding the egg. It also helps the egg itself mature.

D-chiro-inositol does something else. In theca cells, the hormone-producing tissue of the ovary, it behaves like insulin and stimulates testosterone production. The neighbouring granulosa cells tell a different story. There, dose by dose, it reduces the activity of the genes for aromatase. That enzyme converts male-type hormones into female-type ones.

So one form supports the signal that ripens an egg. The other, by contrast, suppresses the switch that makes oestrogen. It pushes production towards testosterone instead. A higher ratio of myo-inositol favours oestrogen; a lower one favours androgens.

Whether myo-inositol actively pushes that switch the opposite way is still a proposal rather than a finding. The reasoning is plausible, since it influences follicle-stimulating hormone signalling and that hormone stimulates aromatase. Direct evidence has not yet been produced, and it is worth saying so rather than assuming symmetry.

There is also a self-reinforcing loop worth understanding, because it explains why the compound keeps appearing in metabolic research.

Glucose and this compound compete for the same transporters, both in the gut and at the cell membrane. High blood sugar therefore blocks entry into cells. High blood sugar also drives the sorbitol pathway, which further depletes the internal supply.

At the same time, insulin resistance impairs the conversion between the two forms. In insulin-sensitive tissues of diabetic rats, conversion falls from a normal 20% to 30% down to around 5%. More is also lost in the urine, in both animals and people.

Depletion then worsens insulin resistance, which deepens the depletion. The relationship is close enough to be useful. The ratio between the two forms in urine has been proposed as a measure of insulin resistance.

That loop is the strongest argument for supplementation anywhere in the field. Where a shortage is driving a problem, restoring supply is a reasonable thing to try.

Beyond signalling, the molecule does structural and housekeeping work. It anchors proteins to cell surfaces and helps sort material taken in from outside. It also regulates osmotic balance, preventing cells from swelling or shrinking. It carries hydroxyl groups that can donate hydrogen to free radicals. For that reason, it has long been presumed to have antioxidant capacity.

In the brain the picture is different again, and it comes with a hard practical limit.

Inositol feeds the cycle, producing two of the brain’s key second messengers. Those messengers sit downstream of receptors for serotonin, noradrenaline and acetylcholine, which is why psychiatric researchers became interested. Lower levels have been found in the spinal fluid of people with depression. The anti-manic effect of lithium has also been linked to reducing them.

That limit is the blood-brain barrier. The molecule is water-loving and cannot cross it at low concentrations. Getting meaningful amounts into the central nervous system requires doses above 12 g a day.

That figure explains something otherwise puzzling. It is why psychiatric trials used amounts several times larger than metabolic ones. Two cautions belong here rather than later, because they qualify everything above.

Much of the mechanism has been established in cells, in animals, or by inference. Adding the compound to a cell culture or an animal diet is hard to interpret. Those models do not capture the transformations it undergoes inside living cells. How free inositol participates in insulin signalling remains unknown. Synthetic versions of the insulin messengers, unlike natural ones, show no insulin-mimicking activity at all.

This plausibility justifies running the trials. It does not, however, stand in for their results.

Can inositol help with PCOS

Start with the finding that reframes the whole question. A healthy ovary holds myo-inositol and D-chiro-inositol at a ratio of roughly 100 to 1. In polycystic ovary syndrome (PCOS), a hormonal condition that can disrupt ovulation, periods and metabolism, that balance does not simply shift. It inverts, falling to about 0.2 to 1. The ovary now holds five times more of the second form than the first. Therefore, the ovary is not short of the compound overall. Instead, the mix has gone the wrong way round.

The mechanism behind that inversion is unusual, and it turns a familiar assumption upside down. Insulin resistance in this condition is tissue-selective. It affects muscle, fat and liver, but not the ovary, which stays fully sensitive to insulin.

Consequently, when insulin runs high, the ovary receives the full signal. This overstimulates the enzyme that converts myo-inositol into D-chiro-inositol. Theca cells from affected women were given identical insulin in the laboratory. That enzyme proved three times more active than in unaffected cells.

The ovary therefore manufactures its own imbalance. Rising D-chiro-inositol promotes androgen production. Meanwhile, the loss of myo-inositol degrades follicle-stimulating hormone signalling, the egg’s energy state, and egg quality.

Some researchers have reconsidered the condition entirely. They describe it less as insulin resistance and more as ovarian hypersensitivity to insulin. Insulin abnormalities affect only around 60% of patients. Yet young women without insulin resistance still respond to myo-inositol, which fits that reading.

It is an elegant explanation. Elegance is not evidence, and this is the point where the article has to slow down. Consider first what has clearly improved in trials, because a great deal has.

Against placebo or folic acid, pooled results show reductions in luteinising hormone and in several androgens. Total testosterone, free testosterone, androstenedione and dehydroepiandrosterone sulphate all fell. Sex hormone-binding globulin, meanwhile, rose substantially. Several of those reached moderate certainty with no inconsistency between studies. Insulin resistance measured by HOMA-IR (a measure of insulin resistance calculated from fasting glucose and insulin levels) improved by a mean difference of 1.09.

Fasting glucose did not improve. Nor did most body measurements. Every item on that list is a laboratory reading. None of them is a symptom, a period, an ovulation or a baby. Move to the outcomes people actually care about, and the picture changes character.

Ovulation rates rose against placebo or folic acid, with a risk ratio of 2.75 at moderate certainty. Pregnancy rates rose too, with an overall risk ratio of 1.29. Live births increased against placebo or folic acid at a risk ratio of 2.29. That figure rests on only two meta-analyses and was rated low certainty.

Individual trials look more striking still. Among 120 women given 4 g of myo-inositol with folic acid, 70% had cycles return. Ovulation returned in 60%, and 28.3% became pregnant. In an observational study of 3602 infertile women, ovulation returned in 70% and 15.1% conceived.

By contrast, the international guideline reached a different conclusion. It judged the evidence on inositol inadequate to support recommendations about clinical outcomes. Both statements are true. The reason lies in the quality of the underlying trials rather than the direction of their results.

The numbers behind the numbers are sobering. Thirty randomised trials of inositol were assessed, covering 2230 participants. Sample sizes ran from 8 to 195 per arm, averaging 36. Half were conducted in a single country too. Most outcomes were rated low to very low certainty and downgraded for risk of bias, inconsistency, and imprecision. Few trials were at low risk of performance or detection bias.

A separate assessment covered 13 meta-analyses and 85 outcome measures, searched to August 2025. Yet it found no high-quality evidence. Moreover, only a quarter of those reviews met high methodological standards.

The fertility evidence deserves particular care, because hope is strongest and the evidence thinnest.

Thirteen trials covering 1472 subfertile women have been pooled. Whether myo-inositol improves live birth rates before in vitro fertilisation remains uncertain, resting on two trials and 84 women. Whether it increases clinical pregnancy is likewise uncertain, with very low-quality evidence across four trials.

Two findings in particular show how fragile these estimates are. An apparent reduction in miscarriage vanished entirely when the single dominant trial was removed. The clinical pregnancy difference disappeared when one trial with unclear allocation concealment was excluded.

Not one of those trials used a placebo. All compared against standard treatment, which may have inflated the apparent effect. Eleven of the thirteen, further, came from Italy.

Reviews reaching cheerier conclusions did so on different foundations. One favourable analysis of ovulation rates drew on trials in women who were not trying to conceive. That is a meaningfully different question. Comparison with metformin produces a genuinely mixed verdict rather than a winner.

Across ten trials, the two performed alike on fasting glucose, insulin, HOMA-IR, cholesterol, weight and body mass index. Reproductive outcomes also did not differ, including menstrual cycle resumption, ovulation, and clinical pregnancy. Metformin was superior for excess hair growth and waist-to-hip ratio, both clinically important. Myo-inositol proved superior instead for sex hormone-binding globulin.

Analysis was then restricted to the two trials at low risk of bias. Metformin won on fasting insulin, while myo-inositol won on regular menstruation. Metformin accordingly remains first-line treatment for metabolic and body-composition outcomes. Myo-inositol may be a reasonable alternative for women who cannot tolerate it. One further complication may explain some of the disagreement in the literature.

The condition is now divided into four phenotypes. Three involve excess androgens; the fourth does not. The proposed mechanism depends on correcting an androgen-driving imbalance. In that fourth group, therefore, it would have no compelling rationale. Trials that mix phenotypes together may therefore dilute a real effect, or manufacture an apparent one.

Meanwhile, a large meta-analysis of 26 trials reported better menstrual normalisation against placebo. It also found greater reductions in weight and body mass index. Another meta-analysis of 17 trials found no improvement in body measurements or metabolic outcomes.

Roughly a quarter of pooled outcomes carry significant heterogeneity that cannot be removed by dropping any single study. That points to systematic differences between trials in form, dose, duration and patient type, rather than to one rogue result.

Most trials also ran for 3 to 24 weeks with no follow-up beyond a year. Nothing can therefore be said about long-term cardiovascular risk or progression to diabetes.

What other benefits have been studied

Outside the ovary, the evidence separates into three very different tiers. Pregnancy has the most. Metabolic and neurological uses have some, although less. Mental health has surprisingly little, given how confidently it is discussed. Pregnancy first, because it is the strongest case in the whole field.

Six randomised trials covering 1140 pregnant women suggest myo-inositol may reduce gestational diabetes, with a risk ratio of 0.53. The effect appears dose-dependent. Daily amounts of 2 to 4 g are associated with lower rates of gestational diabetes and preterm delivery.

Individual trials, however, show more dramatic effects. Three trials gave 4 g daily from after the first trimester, and incidence fell in every group. Among women with a family history of diabetes, it was 6% against 15.3%. Among obese women, it was 14% against 33.6%. In 75 non-obese women with raised fasting glucose, prevalence fell from 71% on placebo to 6%.

Related outcomes moved too. Pooled results indicate reductions in hypertensive disorders of pregnancy, with a risk ratio of 0.34. Preterm birth also fell, to 0.35. Blood fats also improved substantially. Then the list of things that did not change, which is longer and less often quoted.

No difference emerged for caesarean section, birthweight, macrosomia or large-for-gestational-age infants. Shoulder dystocia, gestational age at birth and neonatal respiratory distress were unchanged. Nor was admission to neonatal intensive care. Among women who continued to term, the need for insulin therapy did not differ either.

Six of the trials came from a single country, and most enrolled white women. Applicability elsewhere is therefore unclear. No trial has reported postnatal depression, later maternal diabetes, perinatal mortality or any childhood outcome. Despite the encouraging headline figures, the evidence remains inadequate for routine inclusion in clinical guidelines.

A smaller and stranger body of work concerns neural tube defects, the birth malformations affecting the developing brain and spine.

There is a direct biological requirement here. Depriving rodent embryos of the compound produces skull defects. In a mouse strain whose defects folic acid cannot prevent, supplementation reduces spina bifida.

The human evidence is genuinely interesting and genuinely small. Twelve women had previous affected pregnancies that folic acid did not prevent. They took myo-inositol, and all their subsequent babies were unaffected. In a randomised trial among women with a previous affected pregnancy, 35 supplemented pregnancies produced no recurrence. A fully powered trial has still not been done.

Male fertility has attracted steady attention and rarely reaches the public conversation. Myo-inositol regulates sperm movement, capacitation and the acrosome reaction. In poor-quality samples, it dissolved the fibrous material coating the sperm. It also reduced damage to the mitochondria in the midpiece. Motility and fertilising ability both improved. It also protects sperm during freezing.

Given as a vaginal suppository, it raised total motility from 46.48% to 54.42%, against 46.21% with placebo. Pregnancy followed in 18.60% of treated couples.

On the metabolic side, results are modest but consistent.

In type 2 diabetes, it appears useful as an add-on rather than a replacement. Fasting glucose and HbA1c fell. Body mass index, blood fats and blood pressure did not change. In 20 patients, adding a myo-inositol and D-chiro-inositol combination to existing medication reduced both measures over three months.

Researchers have examined nerve function since the 1970s. Raising dietary intake from 772 mg to 1648 mg improved sensory nerve function in twenty patients with diabetic neuropathy. Even so, the hypothesis that disturbed metabolism directly causes that nerve damage has never been proven.

Now the tier where confidence is thinnest and public claims are loudest. The randomised evidence for inositol in depression consists of four short trials totalling 141 participants. Three came from one research group. All used 12 g a day, several times the metabolic dose, and all ran for four to six weeks.

Used alone against placebo, one trial produced a difference of 7.30 points on a depression rating scale. The confidence interval ran from 14.73 in favour to 0.13 against, so it did not exclude no effect. When added to a selective serotonin reuptake inhibitor, it produced no effect.

Pooling all four gave a standardised mean difference of 0.08, an estimate equally consistent with benefit and with none. Participants were also unusual cases, several having already failed conventional treatment. With evidence this thin, a single unpublished trial could change the answer.

Panic disorder is the one psychiatric use with a more encouraging signal. It reduced panic attacks against placebo over four weeks. In a crossover trial in 21 patients, it matched fluvoxamine for anxiety, agoraphobia and global impression scores.

Elsewhere, the answers are largely negative. Three trials in schizophrenia showed no improvement in symptom scores. Post-traumatic stress disorder showed no change in core symptoms. Trichotillomania showed nothing across 38 patients. Autism showed nothing. Obsessive-compulsive disorder produced discordant results, some small studies positive and others not.

That is a fair summary of the mental-health case. One promising signal, several clear negatives, and a depression literature too small to settle anything. Two further areas are worth knowing about, because both illustrate how evidence behaves over time.

In Alzheimer’s disease, laboratory work is genuinely intriguing. Scyllo-inositol and related forms prevent beta-amyloid protein from forming the fibrils characteristic of the disease. In mouse models, treatment prevented cognitive decline. In a phase II trial in people, 250 mg over 78 weeks produced slight improvement in neurological performance.

Higher doses in that trial caused deaths, and the safety concerns have kept them out of clinical use.

The preterm infant story is the field’s sharpest lesson. Trials in 1986 and 1992 reported improved survival and less severe retinopathy of prematurity. That is the retinal disorder which can blind premature babies. A 2015 review concluded the compound probably reduced preterm death, severe retinopathy and severe bleeding into the brain.

Pooled evidence on inositol across six trials and 1194 infants now shows no effect on severe retinopathy. Measured efficacy fell as publication year advanced, a relationship that reached statistical significance.

Two explanations have been offered. The early trials predate the routine use of antenatal steroids and artificial surfactant. They may therefore have provided a benefit that modern care now provides. The definition of severe retinopathy has also changed over thirty years.

Nothing was wrong with the early trials. The context around them changed, and the finding did not survive it.

Which form and dose matter

Everything established so far converges on a practical problem. Two forms, several ratios, a range of doses and an unregulated market produce a great many possible products. Most of them have never been tested. The doses actually used in research are narrower than the marketplace suggests.

In polycystic ovary syndrome, myo-inositol has been given at 1 to 4 g daily, most commonly 4 g. Folic acid is usually added, at 200 to 400 micrograms. D-chiro-inositol alone has been given at 600 mg to 1.2 g. Pregnancy trials have ranged from 200 mg to 4 g. The most common regimen is 4 g split into two doses.

Psychiatric trials sit in a different world entirely. Schizophrenia trials used 6 g, and depression and post-traumatic stress disorder trials used 12 g. Eating disorder trials went up to 18 g.

Timing and formulation change how much reaches the blood. Absorption is best when 2 g is taken twice daily away from meals. A soft gel capsule is absorbed better than powder, so a dose a third lower achieves the same blood level. Glucose competes for the same transporters, so what is eaten alongside it matters.

The maximum permitted daily intake of inositol is currently 4 g. Splitting that into two daily doses appears to provide better coverage throughout the day. Improvements in hormonal and metabolic markers generally need at least 12 weeks before they stabilise.

Which brings us to the number printed on so many tubs. The 40:1 ratio looks arbitrary. It is not. It approximates the proportion between the two forms found in the blood of healthy women. Supplementing that way aims to restore a physiological balance rather than impose an arbitrary one.

Direct comparisons support it, up to a point. Researchers compared seven ratios directly in 55 women over three months, at a total daily dose of 4 g. The 40-to-1 combination produced the greatest improvements in restoring ovulation. In mice, the same ratio came closest to restoring normal ovarian architecture, while others performed less well.

The case is not closed, and the honest summary is more mixed than the marketing.

Neither trial comparing ratios found any difference for HOMA-IR, the critical measure of insulin resistance. In one study, a combination at roughly 3 to 1 beat the standard formulation. Pregnancy and live birth rates were both higher. Another study found that roughly 2 to 1 improved pregnancy rates and egg-quality measures. The physiological range in the body may run anywhere from 10 to 1 up to 100 to 1.

Large-scale trials and pharmacokinetic studies are still needed before the ratio can be called settled. There is also a mechanical reason the second form is not simply better.

D-chiro-inositol competes with myo-inositol for the same intestinal transporters. Taken together at doses of 1 g or more, it significantly reduces how much myo-inositol is absorbed. The balance in the blood shifts accordingly. At the small amounts used in a 40:1 product, that competition is negligible.

On top of that, doses above 300 mg daily may impair egg quality by altering the ratio inside the ovary. Combining the two forms has not consistently outperformed myo-inositol used alone across the range of outcomes measured.

Head-to-head, myo-inositol has generally done better. In 84 women preparing for in vitro fertilisation, it produced an odds ratio for clinical pregnancy of 3.86 against D-chiro-inositol. In women at high risk of gestational diabetes, 2 g twice daily worked better. It reduced incidence more than D-chiro-inositol alone or a combination.

The most practical finding in this section has nothing to do with ratios. Between 25% and 75% of people treated with myo-inositol do not respond to it. They are described as resistant, and the reason is not fully understood.

Responders differ from non-responders in measurable ways. Those who ovulated or conceived had lower testosterone, at 2.3 against 3.4 nmol/L. They also had higher sex hormone-binding globulin and a lower free androgen index. Most non-responders were obese. Compared with a woman of normal weight, an obese woman had roughly half the probability of ovulating. Her chance of conceiving was a quarter.

Non-responders also failed to show a rise in blood levels, which points to absorption rather than to the tissue response. That suspicion has been tested directly. Fourteen resistant women were given myo-inositol with 50 mg of the milk protein alpha-lactalbumin. They took it twice daily for three months. Twelve of them ovulated, and blood levels rose from 17.0 to 35.0 micromoles per litre.

That is a striking result from a very small group, and it has not been replicated at scale. Finally, this is the part of the subject that receives almost no attention and probably deserves the most.

Four myo-inositol products sold for polycystic ovary syndrome were bought at random from pharmacies in Milan. A certified laboratory analysed them. Only one contained more than 95% of the amount stated on its label. One contained less than 75%.

Measured individually, a product declaring 2000 mg contained 1950 mg. A second declaring 2000 mg contained 1782 mg. A third declaring 2000 mg per tablet contained 1485 mg.

Correcting the prices for what was actually in the packet completely overturned the cost ranking. The product that looked cheapest on its label became more expensive than a rival once measured.

Some products also showed peaks in analysis classified as impurities. One contained maca extract, which raises testosterone, in a product intended for women whose testosterone is already high.

Inositol is not a prescription medicine. It is sold as a dietary supplement. That category is largely unregulated by the European Medicines Agency and the US Food and Drug Administration. European and Italian rules set no strict requirements for manufacturing quality. The raw material source is often poorly controlled, so consistency between batches of the same product is not guaranteed either.

A trial result belongs to the preparation that was tested. Published benefits in polycystic ovary syndrome came from one product on the market since 2004. It contained only myo-inositol and folic acid. Results with newer products were lower. In one group of 40 women, cycles returned in 30% and ovulation in 20%. The established preparation produced 70% and 60%.

What are the side effects of inositol

For most people taking ordinary amounts, the answer is a mildly upset stomach or nothing at all. Doses across trials ranged from 4 to 60 g a day, with exposure lasting one to twelve months. The only adverse events reported were nausea, wind and diarrhoea. Those appeared only above 12 g a day. Their severity did not increase as the dose rose further.

The commonly used 4 g daily has been essentially free of side effects. At that dose for twelve months in 30 women, there were no dropouts and no adverse events. At 6 g for a month in patients with Alzheimer’s disease, psoriasis and bipolar disorder, none were reported either.

Higher amounts produce more. At 12 g for a month in 13 patients with depression, two events occurred. One was nausea, and one was wind. At 18 g in 25 patients, three reported nausea, tiredness, headache and dizziness. Powder causes more gastrointestinal trouble than the soft gel form, which is partly why that form was developed.

Compared with metformin, the difference is clear. A meta-analysis of six trials covering 355 women found a risk ratio of 5.17 for side effects with metformin. Gastrointestinal events were less common with myo-inositol, and typically mild and self-limiting.

That comparison is one of the strongest practical arguments for it, and it deserves a caveat. Of 29 trials in polycystic ovary syndrome, 23 did not report adverse events at all.

Supplementation with inositol is safe and well tolerated in the general population. That holds across the observational studies and clinical trials conducted to date. The US Food and Drug Administration defines it as generally recognised as safe. That permits its use even in infants. Preclinical work indicates no toxicity affecting kidney function, cognition or cancer development.

So far, so reassuring. The rest of this section is where the caution actually lies, and none of it concerns myo-inositol. D-chiro-inositol behaves differently at higher doses, and the evidence for that is uncomfortable.

Normal female mice were given a dose equivalent to about 1200 mg a day in humans. It produced ovarian changes resembling polycystic ovary syndrome. Testosterone rose several times above untreated animals, and aromatase in the ovary fell. The uteruses resembled those of animals no longer cycling normally. At higher equivalent doses, ovarian organisation was disrupted further.

Human data point the same way. At 1200 mg a day, it lowered testosterone and improved ovulation. At 2400 mg, those gains disappeared, and testosterone rose instead by 14.7% for total and 29% for free testosterone.

This is the clearest example in the whole subject of more producing less. D-chiro-inositol suppresses aromatase and promotes androgen production. Used alone at high doses, it risks worsening the very problem it is often bought to address. Higher amounts have also been linked to more immature, lower-quality eggs.

It is not a harmful compound. It is a compound whose effect depends on how much is given, and to whom. A woman who already has too much androgen is the wrong recipient for a high dose. Drug interactions are documented for two medicines and unstudied for the rest.

Lithium and sodium valproate both deplete the body’s myo-inositol, by different routes. That depletion may account for side effects such as excessive thirst and urination, underactive thyroid, and weight gain. When supplementation is used alongside them, the dose must be set carefully. Brain levels must not rise and interfere with the medicine’s intended action.

That makes it a decision for a doctor rather than for self-treatment. Beyond those two drugs, the evidence reviewed here says nothing about interactions. An absence of data is not the same as an absence of risk.

Two more gaps deserve stating plainly rather than smoothing over. Safety data in pregnancy are encouraging but limited. Of seven randomised trials, five measured adverse effects and all five reported none. Even so, longer-term safety for mother and child remains unknown. No trial has followed children into later life.

Breastfeeding is simply not covered. The compound is abundant in human breast milk. That is a statement about what a baby needs. It says nothing about what a nursing mother can safely take. Anyone in that position needs a doctor, not a website.

There is also one setting where the evidence turned actively negative, and it belongs here rather than buried.

In preterm infants, the largest and most recent randomised trial reported a significant increase in deaths. Across six trials, 22.13% of infants given the compound died, compared with 15.48% given placebo. Removing the single trial that predated artificial surfactant made that increase statistically significant. Current evidence does not support routine supplementation in preterm infants.

Taken together, this produces a fair summary. The evidence supports a narrower range than the marketing, and a broader range than the sceptics allow.

Myo-inositol at 2 to 4 g a day is well tolerated and cheap relative to its alternatives. It reliably improves a set of hormonal and metabolic readings in polycystic ovary syndrome. It probably improves ovulation and pregnancy rates, based on low- to moderate-quality evidence. It shows real promise for preventing gestational diabetes in higher-risk women. It performs comparably to metformin on most measures, worse on some, with markedly fewer side effects.

What the evidence does not establish is equally definite. It does not show improved live birth rates with any confidence. It does not support a specific form, ratio or dose as clinically superior. It has not been tested beyond a year, so long-term outcomes remain unknown. It has produced little of value in depression, and nothing in schizophrenia, autism or post-traumatic stress disorder.

Inositol is not a substitute for metformin, for fertility treatment, for antidepressants or for any other prescribed care. It has been examined as a companion to those things. Occasionally it is an option for people who cannot tolerate them.

The marketing here is loud, and the guidelines are quiet. Between them sit three questions worth carrying away. Which form, at what dose, and measured against which outcome? Those three separate almost every claim worth taking seriously from almost every claim that is not.

Sources

• Abdali D, Samson SE, Grover AK. How effective are antioxidant supplements in obesity and diabetes? Med Princ Pract. 2015;24(3):201-215.
• Baillargeon J.-P., Nestler J.E. Polycystic Ovary Syndrome: A Syndrome of Ovarian Hypersensitivity to Insulin? Clin. Endocrinol. Metab. 2006;91:22–24.
• Beemster P, Groenen P, Steegers-Theunissen R. Involvement of inositol in reproduction. Nutrition Reviews 2002;60(3):80-7.
• Bevilacqua A., Dragotto J., Lucarelli M., Di Emidio G., Monastra G., Tatone C. High Doses of D-Chiro-Inositol Alone Induce a PCO-Like Syndrome and Other Alterations in Mouse Ovaries. Int. J. Mol. Sci. 2021;22:5691.
• Cabrera-Cruz H., Oróstica L., Plaza-Parrochia F., Torres-Pinto I., Romero C., Vega M. The insulin-sensitizing mechanism of myo-inositol is associated with AMPK activation and GLUT-4 expression in human endometrial cells exposed to a PCOS environment. Am. J. Physiol. Endocrinol. Metab. 2020;318:E237–e248.
• Carlomagno G., De Grazia S., Unfer V., Manna F. Myo-inositol in a new pharmaceutical form: A step forward to a broader clinical use. Expert Opin. Drug Deliv. 2012;9:267–271.
• Carlomagno G., Unfer V. Inositol safety: Clinical evidences. Eur. Rev. Med. Pharmacol. Sci. 2011;15:931–936.
• Carlomagno G., Unfer V., Roseff S. The D-chiro-inositol paradox in the ovary. Fertil. Steril. 2011;95:2515–2516.
• Cavalli C, Teng C, Battaglia FC, Bevilacqua G. Free sugar and sugar alcohol concentrations in human breast milk. J Pediatr Gastroenterol Nutr. 2006;42(2):215–221.
• Clements R.S. Jr., Darnell B. Myo-inositol content of common foods: development of a high-myo-inositol diet. Am. J. Clin. Nutr. 1980;33(9):1954-67.
• Condorelli R.A., La Vignera S., Bellanca S., Vicari E., Calogero A.E. Myoinositol: Does it improve sperm mitochondrial function and sperm motility? 2012;79:1290–1295.
• Facchinetti F., Bizzarri M., Benvenga S., et al. Results from the international consensus conference on myo-inositol and d-chiro-inositol in obstetrics and gynecology: the link between metabolic syndrome and PCOS. European Journal of Obstetrics & Gynecology and Reproductive Biology. 2015;195:72–76.
• Facchinetti F., Orrù B., Grandi G., Unfer V. Short-term effects of metformin and myo-inositol in women with polycystic ovarian syndrome (PCOS): A meta-analysis of randomized clinical trials. Gynecol. Endocrinol. 2019;35:198–206.
• Garzon S., Laganà A.S., Monastra G. Risk of reduced intestinal absorption of myo-inositol caused by D-chiro-inositol or by glucose transporter inhibitors. Expert Opin. Drug Metab. Toxicol. 2019;15:697–703.
• Greene N.D., Leung K.Y., Gay V., Burren K., Mills K., Chitty L.S., Copp A.J. Inositol for the prevention of neural tube defects: A pilot randomised controlled trial. Br. J. Nutr. 2016;115:974–983.
• Greff D, Juhász AE, Váncsa S, et al. Inositol is an effective and safe treatment in polycystic ovary syndrome: a systematic review and meta-analysis of randomized controlled trials. Reprod Biol Endocrinol. 2023;21(1):10.
• Heimark D., McAllister J., Larner J. Decreased myo-inositol to chiro-inositol (M/C) ratios and increased M/C epimerase activity in PCOS theca cells demonstrate increased insulin sensitivity compared to controls. Endocr. J. 2014;61:111–117.
• Howlett A, Ohlsson A, Plakkal N. Inositol in preterm infants at risk for or having respiratory distress syndrome. Cochrane Database Syst Rev. 2015;2:D366.
• Isabella R., Raffone E. Does ovary need D-chiro-inositol? Ovarian Res. 2012;5:14.
• Joardar S, Duarah P, Purkait MK. Recent advances in myo-inositol recovery and purification from agricultural sources as potential dietary supplements: A review. Sustain Chem Pharm. 2023;36:101331.
• Kamenov Z., Gateva A. Inositols in PCOS. Molecules. 2020;25:5566.
• Larner J., Craig J.W. Urinary myo-Inositol-to-chiro-Inositol Ratios and Insulin Resistance. Diabetes Care. 1996;19:76–78.
• Leung K.Y., Mills K., Burren K.A., Copp A.J., Greene N.D. Quantitative analysis of myo-inositol in urine, blood and nutritional supplements by high-performance liquid chromatography tandem mass spectrometry. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 2011;879:2759–2763.
• Levine J., Barak Y., Gonzalves M., Szor H., Elizur A., Kofman O., Belmaker R.H. Double-blind, controlled trial of inositol treatment of depression. Am. J. Psychiatry. 1995;152:792–794.
• Majunder A., Biswas B., editors. Subcellular Biochemistry. Springer; New York, NY, USA: 2006. Biology of inositols and phosphoinositides.
• Monastra G., Sambuy Y., Ferruzza S., Ferrari D., Ranaldi G. Alpha-lactalbumin Effect on Myo-inositol Intestinal Absorption: In vivo and In vitro. Curr. Drug Deliv. 2018;15:1305–1311.
• Morley LC, Tang T, Yasmin E, Norman RJ, Balen AH. Insulin-sensitising drugs (metformin, rosiglitazone, pioglitazone, D-chiro-inositol) for women with polycystic ovary syndrome, oligo amenorrhoea and subfertility. Cochrane Database of Systematic Reviews 2017, Issue 11.
• Murru A., Popovic D., Pacchiarotti I., Hidalgo D., León-Caballero J., Vieta E. Management of adverse effects of mood stabilizers. Curr. Psychiatry Rep. 2015;17:603.
• Murthy P.P. Structure and nomenclature of inositol phosphates, phosphoinositides, and glycosylphosphatidylinositols. Subcell Biochem. 2006;39:1–19.
• Nestler J.E., Jakubowicz D.J., de Vargas A.F., Brik C., Quintero N., Medina F. Insulin Stimulates Testosterone Biosynthesis by Human Thecal Cells from Women with Polycystic Ovary Syndrome by Activating Its Own Receptor and Using Inositolglycan Mediators as the Signal Transduction System. J. Clin. Endocrinol. Metab. 1998;83:2001–2005.
• Nordio M., Basciani S., Camajani E. The 40:1 myo-inositol/D-chiro-inositol plasma ratio is able to restore ovulation in PCOS patients: Comparison with other ratios. Eur. Rev. Med. Pharmacol. Sci. 2019;23:5512–5521.
• Palatnik A., Frolov K., Fux M., Benjamin J. Double-blind, controlled, crossover trial of inositol versus fluvoxamine for the treatment of panic disorder. J. Clin. Psychopharmacol. 2001;21:335–339.
• Papaleo E, Molgora M, Quaranta L, Pellegrino M, Michele F. Myo-inositol products in polycystic ovary syndrome (PCOS) treatment: quality, labeling accuracy, and cost comparison. European Review for Medical & Pharmacological Sciences 2011;15(2):165-74.
• Phelps DL, Watterberg KL, Nolen TL, et al. Effects of Myo-inositol on type 1 retinopathy of prematurity among preterm infants <28 Weeks’ gestational age. JAMA. 2018;320(16):1649.
• Pintaudi B, Di Vieste G, Bonomo M. The effectiveness of Myo-inositol and d-chiro inositol treatment in type 2 diabetes. Int J Endocrinol. 2016;2016:1-5.
• Pundir J, Psaroudakis D, Savnur P, Bhide P, Sabatini L, Teede H, et al. Inositol treatment of anovulation in women with polycystic ovary syndrome: a meta-analysis of randomised trials. BJOG 2018;125(3):299-308.
• Ravanos K., Monastra G., Pavlidou T., Goudakou M., Prapas N. Can high levels of D-chiro-inositol in follicular fluid exert detrimental effects on blastocyst quality? Rev. Med. Pharmacol. Sci. 2017;21:5491–5498.
• Sacchi S., Marinaro F., Tondelli D., Lui J., Xella S., Marsella T., Tagliasacchi D., Argento C., Tirelli A., Giulini S., et al. Modulation of gonadotrophin induced steroidogenic enzymes in granulosa cells by d-chiroinositol. Reprod. Biol. Endocrinol. 2016;14:52.
• Ueber eine neue, aus dem Muskelfleische gewonnene Zuckerart. Justus Liebigs Annalen Chemie. 1850;73:322–328.
• Spector R. Myo-inositol transport through the blood-brain barrier. Neurochem. Res. 1988;13:785–787.
• Teede HJ, Misso ML, Costello MF, et al. Recommendations from the international evidence-based guideline for the assessment and management of polycystic ovary syndrome. Hum Reprod. 2018;33(9):1602-1618.
• Teede H, Tay CT, Laven J, et al. International Evidence-based Guideline for the Assessment and Management of Polycystic Ovary Syndrome. 2023. Monash University.
• Vitagliano A., Saccone G., Cosmi E., Visentin S., Dessole F., Ambrosini G., Berghella V. Inositol for the prevention of gestational diabetes: A systematic review and meta-analysis of randomized controlled trials. Arch. Gynecol. Obstet. 2019;299:55–68.
• Vitek W, Alur S, Hoeger KM. Off-label drug use in the treatment of polycystic ovary syndrome. Fertility and Sterility 2015;103(3):605-11.