Piracetam: The Molecule That Named a Category

The word “nootropic” did not exist before this compound. Corneliu Giurgea coined it in 1972 to describe what he thought he had made [1].

Fifty years later the picture is stranger than the marketing suggests. A Cochrane review of the naming indication reached a flat conclusion. The evidence does not support using Piracetam in dementia or cognitive impairment [10].

Two further facts sit oddly beside the reputation. The pyrrolidone scaffold does have a high-affinity, stereoselective target in the brain, and Piracetam is the member of the family that does not bind it [9]. Where the drug does work, the dose is 24 grams a day [5].

Chemical identity

A small, very polar cyclic derivative of gamma-aminobutyric acid, and the parent of everything with “racetam” in its name.

Reading the structure

Two rings’ worth of chemistry in a molecule with almost nothing on it.

The core is a 2-pyrrolidinone, a five-membered lactam. An acetamide sits on the ring nitrogen. That is the entire molecule.

Note the logP of -1.3. Piracetam sits about as far from CoQ10 as an orally dosed molecule gets. Both need gram doses, for opposite reasons, and absorption is not the constraint here.

The pyrrolidone family

Every racetam shares the lactam core of Piracetam and differs in what hangs off it.

Aniracetam adds an anisoyl group. Oxiracetam adds a hydroxyl. Pramiracetam replaces the amide with a bulky diisopropylaminoethyl amine. Piracetam itself carries no substituent and no stereocentre.

That last point matters more than it looks. A molecule with no chiral centre cannot have an enantioselective target, and the family’s best-characterised target is enantioselective [9].

The word was invented for this molecule

Giurgea published the nootropic concept in 1972, in French, in a pharmacology annual [1]. He built the category around Piracetam, rather than assigning Piracetam to an existing category.

Giurgea’s criteria

The definition was a list of properties rather than a mechanism.

A nootropic should enhance learning and memory, and protect learned behaviour against disruption. It should protect the brain against physical or chemical injury and increase the efficacy of cortical control mechanisms. It should also lack the sedation, stimulation and toxicity of conventional psychotropics.

Read the list as a specification written from observations of one drug. Nothing in it names a receptor, a pathway or a measurable target.

What the criteria left out

A definition built from effects rather than mechanism cannot be falsified by pharmacology.

No criterion requires a binding site. None requires a dose-response relationship in a defined system. Nor does any of them separate an effect in an impaired animal from an effect in a healthy one, which turns out to be the distinction that matters most for Piracetam.

That structure is why the term travelled so far. It describes a hoped-for profile, and almost any compound can be claimed to fit it.

The mechanism problem

Sixty years of work has not produced an agreed molecular target, and the leading candidate is a physical property rather than a receptor.

Membrane fluidity

Müller and colleagues developed the best-supported account, and its most interesting feature is a negative result.

Preincubating brain membranes from aged mice with Piracetam at 0.1 to 1.0 mmol/L increased membrane fluidity (PMID 9037245). Aged rat and human brain membranes behaved the same way. Young mouse membranes did not change at all [2].

Chronic dosing reproduced the split. At 300 mg/kg daily, membrane fluidity rose in some brain regions of aged rats and not in young ones, and active avoidance learning improved in the aged animals only [2].

The same group later turned to human tissue. Hippocampal membranes from Alzheimer’s disease patients showed lower hydrocarbon core fluidity than elderly controls, and preincubation with Piracetam abolished the difference [8]. A mitochondrial extension followed, reporting enhanced membrane potential and ATP production in cell and animal models [13].

Note the concentration. Effects appear at 0.1 to 1.0 mmol/L in vitro, which is a demanding number to reach and hold in a human brain.

Müller and colleagues state the pattern plainly in their own review. Piracetam combines a non-specific physico-chemical mode of action with a clinical profile in which effects are always more pronounced when function is impaired [6].

That is an honest description and an awkward one. A drug that acts on a bulk physical property of membranes, and only where those membranes are already altered, has no reason to do anything in a healthy brain.

One laboratory, four papers

The membrane account is coherent, and it is also concentrated.

Four of the mechanistic papers cited here share an author, and three share a laboratory in Frankfurt [2][6][8][13]. Two of them appeared in the same 1999 journal supplement devoted to Piracetam [6][7].

None of that makes the work wrong. It does mean the leading mechanism has had less independent replication than its prominence suggests, and the concentration is visible from the author lists rather than inferred.

The target it does not hit

The pyrrolidone family does have a real, high-affinity, stereoselective binding site. It belongs to a different member.

Genton and Van Vleymen set the two side by side [9]. Levetiracetam is an S-enantiomer of a close relative. A brain-specific stereoselective binding site exists for which levetiracetam and other S-enantiomers have high affinity, and Piracetam does not.

That site is synaptic vesicle protein 2A, and its importance is not theoretical. A homozygous SV2A mutation in a human produced intractable epilepsy, involuntary movements and microcephaly [14].

So the scaffold yielded a widely used antiepileptic once someone found a stereospecific site on it. An achiral molecule was never going to bind that site. The reason is structural rather than historical.

What that means for read-across

Results on one racetam do not transfer to another, and the family history shows why.

Levetiracetam and Piracetam differ by an ethyl group and a stereocentre. That changed the profile completely. Piracetam improves learning in preclinical models, and levetiracetam is far more active against seizures [9].

They also converge in one clinic in a way that cuts against simple potency arguments. Both show activity in cortical myoclonus [7][15][16], reaching it by different routes.

The Cochrane verdict

The naming indication has been formally assessed, and the assessment is the single most important document in this file.

Flicker and Grimley Evans searched the trial literature and went further. They obtained a comprehensive abstract list, including many unpublished studies, from the company that marketed most of the world’s supply [10]. That access makes the conclusion harder to dismiss as publication bias.

Global impression of change was the only outcome with a substantial pooled evidence base. The odds ratio favoured Piracetam at 3.55 under a fixed effects model and 3.47 under a random effects model. Heterogeneity between studies was significant [10].

The reviewers still concluded against use. Effects appeared on global impression of change and on nothing more specific. On that basis the available evidence did not support use in dementia or cognitive impairment [10].

The dose that does work

One indication produces unambiguous, dose-ordered improvement from Piracetam. The dose is far above anything sold for cognition.

Progressive myoclonus epilepsy

Koskiniemi and colleagues studied 20 patients with classical Unverricht-Lundborg disease (PMID 9527146). The design was randomised, double-blind and crossover, comparing three daily doses of Piracetam against placebo.

The 24 g daily arm produced significant improvement in the primary sum score (p = 0.005). Motor impairment, functional disability and global assessments by investigator and patient all improved [5].

Lower doses did less. Significant improvement in functional disability also appeared at a lower daily dose, and the primary outcome reached significance only at 24 g [5].

The dose gradient

Genton and colleagues reviewed the cortical myoclonus literature and added 12 patients of their own [7].

Requirements varied widely, from 7 to 24 g daily, and responses were dose-related. Their own patients received up to 45 g of Piracetam daily on top of existing antiepileptic treatment. Improvement held for up to seven years.

Set that against the nootropic dose range. Cognitive use is marketed at roughly 1.6 to 4.8 g daily. That is a fifth to a twentieth of the dose that demonstrably moves a clinical endpoint.

The stroke trial

A neuroprotection hypothesis reached a large trial, and the trial was null.

PASS

De Deyn and colleagues randomised 927 patients within 12 hours of acute ischaemic stroke (PMID 9412612). Piracetam was given as a 12 g intravenous bolus, then 12 g daily for four weeks, then 4.8 g daily for eight weeks.

Outcomes matched almost exactly. Mean Orgogozo scale at four weeks was 57.7 on treatment against 57.6 on placebo. Mean Barthel Index at twelve weeks was 55.8 against 53.1 [3].

Mortality ran the wrong way. Twelve-week mortality was 23.9% on treatment and 19.2% on placebo. That is a relative risk of 1.24, with a confidence interval of 0.97 to 1.59 and p = 0.15 [3]. It does not reach significance, and it is not a favourable trend either.

What happened afterwards

The trial protocol specified an early-treatment population, and post hoc work moved the goalposts twice.

The early-treatment group was first defined as treatment within 6 hours, then redefined as under 7 hours [3]. A published economic evaluation narrowed it again. Its population was patients treated within 6 hours 59 minutes who also had an initial Orgogozo score below 55 [4].

That analysis concluded Piracetam was cost-effective in moderate to severe stroke treated early [4]. It rests on a subgroup of a null trial, selected after the fact on two criteria.

The vascular strand

A separate literature treats Piracetam as a rheological agent rather than a cognitive one, and it fits the membrane account better.

Winnicka and colleagues summarise the case [12]. Piracetam alters plasma membrane physical properties, increases red cell deformability, and normalises aggregation of hyperactive platelets. Antithrombotic and microcirculatory effects follow from the same physical chemistry rather than from a receptor.

Winblad’s review lists documented efficacy in cognitive disorders and dementia, vertigo, cortical myoclonus, dyslexia and sickle cell anaemia [11]. The dementia entry is the one Cochrane had declined to support four years earlier [10]. Both documents are worth reading together.

A pattern worth naming

The same shape recurs across this compound’s evidence base and across others in this category.

A clinician’s global impression moves and the specific instruments do not [10]. Elsewhere a subjective scale separates while an objective marker does not, which is the shape of the statin and coenzyme Q10 literature. Then a subgroup rescues a null primary endpoint [3][4].

None of that proves an effect is absent. It does mean the endpoints most vulnerable to expectation are carrying the positive results. Piracetam is also a bitter powder taken by the gram, which makes blinding harder than it looks on paper.

The myoclonus work is the counterexample, and that is why it matters. It combined a rating scale, a dose-response gradient and a crossover design. The result held across separate doses in separate series [5][7].

How to read a Piracetam study

Four questions decide whether a result transfers.

Impaired or healthy?

The membrane effect appeared in aged animals and human tissue, and not in young animals [2]. A study in healthy young subjects tests a different system from the one where the mechanism operates.

Which dose?

Cognitive claims sit at 1.6 to 4.8 g. The myoclonus evidence sits at 24 to 45 g [5][7]. Those are not the same intervention.

Which endpoint?

Global impression and specific cognitive instruments diverged in the pooled analysis [10]. Any abstract reporting improvement should be checked for which of the two moved.

Which racetam?

The family shares a lactam core and little else [9]. Findings on levetiracetam belong to a stereoselective SV2A ligand and do not transfer to Piracetam.

Verifying research material

Piracetam is a small, achiral, highly water-soluble solid. That is the easiest class of compound to characterise, and there is no excuse for a thin certificate.

Identity

Melting point alone is informative here. Pure Piracetam melts sharply around 152 °C, so a depressed or broad range indicates impurity without any instrument at all.

Nuclear magnetic resonance settles the structure quickly. The molecule has six carbons and a simple proton spectrum, so a mismatch is obvious rather than subtle.

Mass spectrometry gives 142.07 for Piracetam. The plausible relatives differ by whole mass units or more: levetiracetam at 170.21, and the hydrolysis product 2-oxo-1-pyrrolidineacetic acid at 143.14.

Purity and related substances

Ask a supplier for the pharmacopoeial impurity profile.

The acid hydrolysis product is the impurity to expect. An amide in a moist environment slowly becomes a carboxylic acid. Melting point will not catch it and chromatography will.

Residual solvents and heavy metals belong on the certificate as usual. Chiral purity is not a question for Piracetam, which removes an ambiguity that complicates several of its relatives.

Handling

Store dry and sealed. Piracetam is hygroscopic and will pick up water from room air, which matters when weighing gram quantities.

Aqueous solubility is high, so preparing stock solutions is straightforward and no organic carrier is needed. Vehicle effects confound much of the cell literature on less soluble compounds, and they do not arise here.

Weighing accuracy is the practical issue at these masses. The gap between 1.6 g and 24 g separates two entirely different bodies of evidence. Recording the measured mass therefore matters more than usual.

Kimera publishes third-party certificates of analysis for every lot in its COA database. Laboratories source Piracetam as the reference member of the pyrrolidone series. It appears alongside aniracetam and oxiracetam, which vary the substituent on the same core. It also pairs with pramiracetam, where the amide itself is replaced. Related work appears in the nootropics category.

Common questions about Piracetam

Where does the word nootropic come from? Giurgea coined it in 1972 for this compound, defining the class by a list of desired properties rather than by a mechanism [1].

What is the proposed mechanism? Interaction with membrane phospholipids that raises fluidity. It is reported in aged and Alzheimer’s disease tissue and not in young animals [2][8].

Does it bind SV2A like levetiracetam? No. The stereoselective site takes S-enantiomers, and this molecule has no stereocentre [9].

What did the Cochrane review conclude? That available evidence does not support use in dementia or cognitive impairment, despite a positive pooled result on global impression of change [10].

Where is the evidence strongest? Cortical myoclonus. A crossover trial in 20 patients found the primary endpoint significant at 24 g daily [5].

What happened in the stroke trial? No difference from placebo across 927 patients, with mortality numerically higher on treatment [3].

Summary of the evidence

Identity: C₆H₁₀N₂O₂, 142.16 g/mol, CAS 7491-74-9. A 2-pyrrolidinone with an acetamide on the ring nitrogen, achiral, logP -1.3.

Origin: the term nootropic was written for this molecule in 1972, as a list of properties rather than a mechanism [1].

Mechanism: membrane fluidity effects at 0.1 to 1.0 mmol/L. Present in aged mouse, rat and human brain membranes, absent in young ones [2], with Alzheimer’s hippocampal membranes normalised in vitro [8].

The family target: SV2A takes S-enantiomers and Piracetam is achiral. The scaffold’s best-characterised binding site belongs to levetiracetam instead [9][14].

Dementia: Cochrane found a positive pooled global impression of change and no benefit on specific measures, and concluded against use [10].

Myoclonus: significant, dose-ordered improvement at 24 g daily in 20 patients [5]. A smaller series reported up to 45 g daily with improvement held for seven years [7].

Stroke: 927 patients, no difference in primary or secondary outcome, mortality 23.9% against 19.2% [3]. A post hoc subgroup carried the later economic case [4].

Limits: no agreed molecular target after sixty years, and an effect confined to impaired systems in the mechanistic work. A five to twentyfold gap separates the marketed dose from the demonstrated one.

Status: supplied for laboratory research use only.

References

1. Giurgea C. Pharmacology of integrative activity of the brain. Attempt at nootropic concept in psychopharmacology. Actual Pharmacol (Paris). 1972;25:115-156. PMID 4541214.
2. Müller WE, Koch S, Scheuer K, Rostock A, Bartsch R. Effects of piracetam on membrane fluidity in the aged mouse, rat, and human brain. Biochem Pharmacol. 1997;53(2):135-140. PMID 9037245. DOI
3. De Deyn PP, De Reuck J, Deberdt W, Vlietinck R, Orgogozo JM. Treatment of acute ischemic stroke with piracetam. Members of the Piracetam in Acute Stroke Study (PASS) Group. Stroke. 1997;28(12):2347-2352. PMID 9412612. DOI
4. Murphy N, Kazek MP, Van Vleymen B, Melac M, Souêtre E. Economic evaluation of Nootropil in the treatment of acute stroke in France. Pharmacol Res. 1997;36(5):373-380. PMID 9441728. DOI
5. Koskiniemi M, Van Vleymen B, Hakamies L, Lamusuo S, Taalas J. Piracetam relieves symptoms in progressive myoclonus epilepsy: a multicentre, randomised, double blind, crossover study comparing the efficacy and safety of three dosages of oral piracetam with placebo. J Neurol Neurosurg Psychiatry. 1998;64(3):344-348. PMID 9527146. DOI
6. Müller WE, Eckert GP, Eckert A. Piracetam: novelty in a unique mode of action. Pharmacopsychiatry. 1999;32 Suppl 1:2-9. PMID 10338102. DOI
7. Genton P, Guerrini R, Remy C. Piracetam in the treatment of cortical myoclonus. Pharmacopsychiatry. 1999;32 Suppl 1:49-53. PMID 10338109. DOI
8. Eckert GP, Cairns NJ, Müller WE. Piracetam reverses hippocampal membrane alterations in Alzheimer’s disease. J Neural Transm (Vienna). 1999;106(7-8):757-761. PMID 10907734. DOI
9. Genton P, Van Vleymen B. Piracetam and levetiracetam: close structural similarities but different pharmacological and clinical profiles. Epileptic Disord. 2000;2(2):99-105. PMID 10954241.
10. Flicker L, Grimley Evans G. Piracetam for dementia or cognitive impairment. Cochrane Database Syst Rev. 2001;(2):CD001011. PMID 11405971. DOI
11. Winblad B. Piracetam: a review of pharmacological properties and clinical uses. CNS Drug Rev. 2005;11(2):169-182. PMID 16007238. DOI
12. Winnicka K, Tomasiak M, Bielawska A. Piracetam, an old drug with novel properties? Acta Pol Pharm. 2005;62(5):405-409. PMID 16459490.
13. Leuner K, Kurz C, Guidetti G, Orgogozo JM, Müller WE. Improved mitochondrial function in brain aging and Alzheimer disease, the new mechanism of action of the old metabolic enhancer piracetam. Front Neurosci. 2010;4:44. PMID 20877425. DOI
14. Serajee FJ, Huq AM. Homozygous mutation in synaptic vesicle glycoprotein 2A gene results in intractable epilepsy, involuntary movements, microcephaly, and developmental and growth retardation. Pediatr Neurol. 2015;52(6):642-646. PMID 26002053. DOI
15. Frucht SJ, Louis ED, Chuang C, Fahn S. A pilot tolerability and efficacy study of levetiracetam in patients with chronic myoclonus. Neurology. 2001;57(6):1112-1114. PMID 11571347. DOI
16. Striano P, Manganelli F, Boccella P, Perretti A, Striano S. Levetiracetam in patients with cortical myoclonus: a clinical and electrophysiological study. Mov Disord. 2005;20(12):1610-1614. PMID 16078205. DOI

Piracetam is sold for laboratory research use only. Not for human consumption, nor medical, veterinary, or household uses.

Literature retrieved from PubMed.