Oxiracetam

Water-soluble racetam

Oxiracetam (Water-soluble racetam) is a cognitive & mood research compound. Cholinergic and glutamatergic modulator — the 'logic and focus' racetam with mild stimulation.

How Oxiracetam works

Cholinergic and glutamatergic modulator — the 'logic and focus' racetam with mild stimulation.

Proposed benefits

Researched for focus, memory, neuroprotection, mood and stress resilience.

Where to get Oxiracetam

The evidence for Oxiracetam

Graded by what exists behind each claim.

✅ Clinically validated

• Studied in European clinical trials in dementia and cognitive impairment in the 1980s and 90s, with mixed results. Approved in Italy at one point; the trial base is dated and would not meet a modern regulatory bar.

📊 Correlative data

• Community use for focus and verbal fluency. Reported as the most 'stimulating' of the classical racetams without being a stimulant, and generally well tolerated.

🧪 Theoretical / extrapolated

• A racetam that modulates AMPA and NMDA receptor function and increases acetylcholine release in the hippocampus.
• The cholinergic component is why headache is the classic side effect and why a choline source is conventionally paired with it — a genuinely mechanistic reason rather than folklore.

How to read these tiers: they say how much human evidence exists, not how well something works — and ✗ flags harm, never a disappointing trial. How the evidence tiers work →

What Oxiracetam actually does

Oxiracetam is piracetam with one hydroxyl group, and that hydroxyl is the entire compound. The molecule is 4-hydroxy-2-oxo-1-pyrrolidineacetamide: the piracetam ring with an –OH hung on carbon 4. Two consequences follow, and only one of them is on any vendor page. The obvious one is solubility — an extra hydroxyl makes an already polar molecule more polar, which is why this is the water-soluble racetam and why it leaves the body the way it does. The consequence nobody states is that carbon 4 now carries four different substituents. It is a stereocentre. Oxiracetam exists as two non-superimposable mirror images, and what you buy is a 50:50 mixture of them.

That matters because the two halves are not equivalent, and the experiment separating them has been done. In rats with chronic cerebral hypoperfusion, (S)-oxiracetam and not (R)-oxiracetam rescued spatial learning and memory, reduced neuronal damage and white matter lesions, raised cerebral blood flow and suppressed astrocyte activation Li 2017. The authors state the position in their own title: oxiracetam is used as a racemic mixture whose effective ingredient had not been identified, and it is the S-enantiomer.

So say the arithmetic out loud, because it reorganizes every dose number attached to this compound. If (S) carries the activity, a 2,400 mg racemic dose is 1,200 mg of drug and 1,200 mg of ballast. Every trial ever run on racemic oxiracetam was, in the currency that matters, a half-dose trial — and the human data below behave exactly as that prediction says they should.

What the mechanism is not. This site's own card says oxiracetam modulates AMPA and NMDA receptor function and raises hippocampal acetylcholine. No binding affinity for either receptor has ever been published for this molecule — not a Ki, not an IC50, not a displacement curve. What has actually been measured is one step further out: using MALDI mass spectrometry imaging and LC-MS/MS on hypoperfused rat cortex, (S)-oxiracetam altered ATP metabolism, the glutamine–glutamate cycle and antioxidant levels Li 2017. That is a metabolic and hemodynamic signature, not a receptor one, and the honest version of this page is that oxiracetam has the best human data and the worst receptor pharmacology of anything in its class — the exact inverse of the usual arrangement.

Cell, rodent, human — and where it stops

There is no cell rung, and the chain is stronger without a pretend one. The mechanistic read-out for this compound was made in whole rat cortex by imaging mass spectrometry Li 2017, not in a dish. Nothing about oxiracetam has been established in a cell line.

Rodent. Rats, chronic cerebral hypoperfusion — the carotid-occlusion model of vascular cognitive impairment. Spatial learning and memory, neuron and white matter histology, cerebral blood flow, astrocyte activation. (S)-oxiracetam improved all of them; (R)-oxiracetam did not Li 2017.

Human, trial one, and it is the largest ever run on this drug: it found nothing. 500 people randomized, mean age 68.9, a median of 32 months after their stroke, 800 mg twice daily against placebo for 36 weeks Lim 2023, 457 completing. Change in MMSE was +0.13 ± 2.27 on oxiracetam against +0.27 ± 2.09 on placebo, p = 0.49; change in CDR sum-of-boxes was −0.14 ± 0.70 against −0.08 ± 0.80, p = 0.38. The authors' own conclusion is that oxiracetam did not prevent post-stroke cognitive impairment in high-risk patients, and that this supports South Korea's regulatory decision to suspend its use Lim 2025.

Human, trial two, and it found something — using the single enantiomer. A phase 3 across 51 Chinese hospitals, 2019 to 2024, 590 patients with traumatic brain injury and Glasgow Coma Scale 10–15, mean age 50.9, randomized 2:2:1 to L-oxiracetam 4 g/day, racemic oxiracetam 6 g/day, or placebo. Primary endpoint was the change in Loewenstein Occupational Therapy Cognitive Assessment score at 90 days. L-oxiracetam improved by a mean of 20.45 points, beating placebo by 8.97 points (Cohen's d = 0.48) and beating racemic oxiracetam by 4.54 points, with no between-group difference in serious adverse events Liu 2025.

The obstacles, and the first three are numbers rather than opinions. (1) The dose. The trial that worked used 4,000 mg of the active enantiomer a day; the trial that failed used 1,600 mg of a racemate, which is 800 mg of active enantiomer. This site's 800–2,400 mg sits at or below the failed arm. (2) The enantiomer showed up in a human endpoint. 4 g of L- beat 6 g of racemate by 4.54 points Liu 2025; on a pure dilution model, 6 g of racemate is 3 g of active and should have come close. It did not. (3) Both human trials ran in injured brains — chronic post-stroke and acute TBI. And the specific obstacle, the one this whole family shares and nobody writes: the rodent was hypoperfused. The mechanism with actual rodent evidence behind it is raising cerebral blood flow in a brain that does not have enough. A healthy person's cerebral perfusion is already normal, so the deficit the drug corrects does not exist in the reader. Set that beside the Cochrane verdict on the parent compound — no support for piracetam in dementia or cognitive impairment on any specific measure Flicker 2001 — and the family pattern is that these molecules rescue damage and do not add capacity.

Oxiracetam pharmacokinetics — how much of it actually gets in

This is the one compound in the family whose pharmacokinetics are genuinely settled, and the settled answer is unusual: almost nothing happens to it.

What degrades it — essentially no enzyme at all. After a single 800 mg oral dose, 84% of the dose was recovered unchanged in urine within 24 hours Perucca 1987. There is no cytochrome step worth naming, no ester for a carboxylesterase to open, no amide hydrolysis that matters to exposure. Oxiracetam is filtered out of the blood in the same chemical form it entered. On a site full of compounds whose real story is first-pass metabolism, this one has almost no metabolism to discuss — and that absence is the most useful thing about it, because it makes exposure predictable.

The numbers. Peak serum concentration about 25 micrograms per milliliter at 1–3 hours after 800 mg; elimination half-life 3–6 hours in very elderly women; 800 mg twice daily for 7 days produced no accumulation, with troughs of 4.6 and 17.0 mcg/mL on day 4 and 5.7 and 17.9 mcg/mL on day 7 Perucca 1987. In the direct comparison, 18 elderly patients against 6 healthy adults at the same 800 mg: half-life 12.3 hours in the elderly against 7.7 hours in healthy subjects, area under the curve doubled in the elderly, peak concentration almost unchanged and slightly delayed, volume of distribution unchanged, and linear kinetics between 800 and 2,000 mg in healthy subjects Lecaillon 1990.

The oral barrier is the smallest one on this site. A hydroxyl on a small, very polar amide is a molecule that dissolves and is absorbed; the card's ‘fasted: no’ is not a workaround for a solubility problem, because there is not one. There is no first-pass conversion step to survive.

The consequence that changes practice: your dosing variable is kidney function, not liver function. The elderly half-life rose and the AUC doubled because renal function falls with age, not because anything hepatic changed — volume of distribution was unchanged Lecaillon 1990. Anybody with reduced glomerular filtration — age, one kidney, chronic disease, chronic NSAID use — is sitting on the elderly curve at the same milligram dose: roughly double the exposure, at a 12-hour half-life rather than an 8-hour one. That is the practical reading of this site's ‘~8 hrs’, which is accurate for a healthy adult and wrong by half for a lot of the people taking it. Route: oral only, in every study cited on this page.

What would have to be true, and how you would know it was not

Four predictions. The first two argue against the product at the dose people actually take, and the first one has effectively already been run.

1. Against: MoCA will not move in an intact adult at the label dose. Five hundred people with a genuine reason to improve took 800 mg twice a day for 36 weeks and their MMSE moved 0.13 points, against 0.27 on placebo Lim 2025. A healthy adult has less room to gain, not more. Draw the line concretely: MoCA at baseline and at 12 weeks, using alternate forms so you are not measuring practice. A gain inside the instrument's own test-retest error is not a result, and for MoCA that error is a point or two.

2. Against, and this one is arithmetic: 2,400 mg of racemate is 1,200 mg of active drug. The phase 3 that produced a real effect used 4,000 mg of L-oxiracetam Liu 2025. Prediction: somebody running the top of this site's range is taking under a third of the only regimen that has ever beaten placebo, of the only preparation that has ever beaten placebo. If that person reports a clear effect, the explanation to reach for first is expectancy.

3. Cystatin-C predicts your exposure, and nobody has ever framed it that way. Because 84% of the dose leaves unchanged in urine Perucca 1987 and the elderly AUC doubled on a renal mechanism Lecaillon 1990, exposure at a fixed milligram dose should scale inversely with filtration rate. Draw cystatin-C before you start. If it is high — and it catches reduced filtration earlier than creatinine does in muscular people, which describes this site's readers — predict roughly the elderly curve: twice the exposure, a 12-hour half-life, and no reason to be taking it twice a day.

4. The prediction that could go the compound's way, and its marker. The only mechanism with rodent evidence is restoring flow and metabolism in a hypoperfused brain Li 2017. So the population where this should work is one with genuinely compromised cerebral perfusion, and the practical proxy a reader can actually order is vascular: hs-CRP and a lipid panel, plus documented hypertension. Prediction: in somebody with normal vascular markers there is no deficit for this drug to correct, and the null in trial one is what a whole population of such people looks like.

What nobody has tested yet

Five experiments. The first costs one afternoon and would tell every buyer something no certificate of analysis currently says.

1. Nobody has published the enantiomeric ratio of a commercial oxiracetam powder. Chiral HPLC, five vendors, one run. Everyone assumes 50:50 because a non-stereoselective synthesis usually gives 50:50, and ‘usually’ is not a specification. Given that one enantiomer does the work Li 2017, the ratio in the tub is the single most decision-relevant number about the product and it has never been measured in public.

2. Nobody has given (S)-oxiracetam to a healthy adult. Both arms of the phase 3 were in traumatic brain injury Liu 2025; the 500-person trial was post-stroke Lim 2025. A four-arm dose-ranging study in healthy adults — 1,000, 2,000 and 4,000 mg of the single enantiomer against placebo, sustained attention as the endpoint — is a completely ordinary trial that has never been run for the only racetam with a phase 3 win.

3. Nobody knows whether (R)-oxiracetam is inert or antagonistic. Li's design showed that it does not work Li 2017; it did not test whether it interferes. That distinction is worth a real amount: if (R) merely sits there, 6 g of racemate should behave like 3 g of (S), and in the phase 3 it did not — it lost to 4 g of L- by 4.54 points Liu 2025. A three-arm rodent study, (S) alone against (S) plus (R) against racemate at matched (S) content, would settle it.

4. There is no exposure study in renal impairment. The elderly data are a proxy for reduced filtration, not a measurement of it Lecaillon 1990. For a drug that is 84% renally excreted unchanged Perucca 1987, the absence of a formal renal-impairment pharmacokinetic study is the gap that would matter first in any modern approval.

5. Nobody has explained the instrument gap. LOTCA moved 20 points in traumatic brain injury Liu 2025 while MMSE moved 0.13 points in chronic post-stroke Lim 2025. Same drug class, same year, opposite answers. Whether that is the instrument's sensitivity, the acuteness of the injury, or the enantiomer has never been analyzed, and it is the question that decides whether this compound has a use at all.

Oxiracetam — its own safety story, not its class's

The class block worries about cholinergic headache and theoretical excitotoxicity. Oxiracetam's own risk story is neither. It is regulatory, renal, and about what is in the tub.

1. A national regulator suspended it, and that is in the trial report rather than a rumor. The 500-person randomized trial concluded that its result supports South Korea's decision to suspend the use of oxiracetam Lim 2025. That is a suspension for lack of efficacy, not for harm, and the distinction is worth holding carefully: this drug's problem is that it did not work in the population it was licensed for, not that it hurt anybody.

2. The measured safety record is better than most things on this site. 590 patients across 51 hospitals with no between-group difference in serious adverse events Liu 2025, and 457 people completing 36 weeks of continuous dosing Lim 2025. That is a real human safety database, and this compound is one of very few in this catalog that has one.

3. The person to be careful with is the one with kidneys that do not filter well. Because 84% of the dose leaves unchanged in urine Perucca 1987 and reduced filtration doubled the exposure Lecaillon 1990, chronic kidney disease, a single kidney, or habitual high-dose NSAID use are the conditions that turn a label dose into a double dose. No renal dose adjustment has ever been published, because no renal-impairment study has been run, so there is nothing to look up — which is precisely why it belongs on this page rather than in a footnote.

4. Anti-doping: this one is not the problem child of the family. Oxiracetam is not a stimulant and is not named on the WADA Prohibited List as this page is written — unlike phenylpiracetam, which is. That is a statement about a list that changes annually, so check the current Prohibited List before you compete rather than trusting this paragraph.

5. The supply risk here is chirality, not quantity. Analysis of cognitive-enhancement supplements found 75% of declared quantities inaccurate Cohen 2021. For a gram-dosed compound with a 36-week human exposure record, a quantity error is survivable. The error nobody can even detect is enantiomeric: no certificate of analysis reports the (S):(R) ratio, mass spectrometry cannot separate mirror images, and the difference between a 50:50 powder and a 30:70 powder is the difference between a half-dose and a third-dose of the only fraction that works.

Sources read for this page

• Li W, Liu H, Jiang H, Wang C, Guo Y, Sun Y, Zhao X, Xiong X, Zhang X, Zhang K, Nie Z, Pu X. (S)-Oxiracetam is the Active Ingredient in Oxiracetam that Alleviates the Cognitive Impairment Induced by Chronic Cerebral Hypoperfusion in Rats. Scientific Reports 2017;7(1):10052 · PMID 28855592
• Liu T, Wang J, Zhao Z, Jiang W, Zhang M, Yu Y, Liu Y, Liu M, Chen L, Zhang H, et al. Efficacy and safety of L-oxiracetam on cognitive function in patients with traumatic brain injury: a multicentre, randomised, double-blind, phase 3 clinical trial. Signal Transduction and Targeted Therapy 2025 · PMID 41381424
• Lim JS, Rha JH, Park JH, Lee K, Chang DI, Heo SH, Lee YB, Kwon JH, Kim EG, Choi JC, et al. Oxiracetam and physical activity in preventing cognitive decline after stroke: A multicenter, randomized controlled trial. European Stroke Journal 2025 · PMID 40882961
• Lim JS, Lee J, Kang Y, Park HT, Kim DE, Cha JK, Park TH, Heo JH, Lee KB, Park JM, et al. Efficacy and safety of oxiracetam in patients with vascular cognitive impairment: A multicenter, randomized, double-blinded, placebo-controlled, phase IV clinical trial. Contemporary Clinical Trials 2023;126:107108 · PMID 36724841
• Lecaillon JB, Dubois JP, Coppens H, Darragon T, Theobald W, Reumond G, Beck H. Pharmacokinetics of oxiracetam in elderly patients after 800 mg oral doses, comparison with non-geriatric healthy subjects. European Journal of Drug Metabolism and Pharmacokinetics 1990;15(3):223-230 · PMID 2253653
• Perucca E, Parini J, Albrici A, Visconti M, Ferrero E. Oxiracetam pharmacokinetics following single and multiple dose administration in the elderly. European Journal of Drug Metabolism and Pharmacokinetics 1987;12(2):145-148 · PMID 3691580
• Flicker L, Grimley Evans G. Piracetam for dementia or cognitive impairment. Cochrane Database of Systematic Reviews 2001;(2):CD001011 · PMID 11405971
• Cohen PA, Avula B, Wang YH, Zakharevich I, Khan I. Five Unapproved Drugs Found in Cognitive Enhancement Supplements. Neurology: Clinical Practice 2021;11(3):e303-e307 · PMID 34484905

Oxiracetam — safety, predicted from mechanism

Oxiracetam — interference & stacking

Bloodwork to run alongside Oxiracetam

Baseline first, then again at 8–12 weeks.

The Brain Fog & Cognition panel covers these in one order — 12 markers, $233.06 with the discount applied.

Check results you already have → · All 103 markers A–Z

Oxiracetam — frequently asked questions

What Oxiracetam is used for

Oxiracetam appears under 1 goal in the goal router.

Related Cognitive & Mood compounds

← Explore the full Protocol Vault

↑ Back to on this page