THE EVIDENCE, ROW BY ROW
Fifty PubMed records exist for this molecule and its names; roughly fifteen are human clinical papers, every one of them Russian, mostly in the Korsakov Journal. This page is the ledger: the strongest design first, the largest datasets next, the negative signals in the same size type.
Seizure frequency and cognition improved in patients without epileptiform EEG abnormalities.
MIXED: the drug did not decrease the negative effects of standard antiepileptic therapy in 40% of patients — printed in the abstract.[16]
MFI-20 asthenia score fell a mean 16.3 points (95% CI 8.85–23.85) on 200 mg/day × 1 month; side effects 5.5%, transient.
All pooled studies small and single-country; journal of record for the manufacturer’s market; no placebo-pooled estimate in the abstract.[15]
Asthenia severity decreased significantly by month 1, sustained to month 3; greater than twofold symptom reduction.
No control group, no blinding, no randomisation.[17]
Neurologic function and daily-living recovery reported significantly better than controls (p<0.0001).
No randomisation or blinding stated in the abstract.[18]
Tolerability described as satisfactory.
TOLERANCE SIGNAL: the anxiolytic effect decreased after 4–8 weeks of continued treatment.[20]
Reported improvements in pareses, coordination, memory, attention, mood; EEG normalisation trend.
No control group stated.[19]
Reported seizure reduction, cognitive and quality-of-life improvement; no seizure aggravation.
Combination therapy with mexidol confounds attribution.[21]
Significant seizure decrease and EEG improvement versus antiepileptics alone.
No randomisation or blinding stated; small groups.[22]
A further set of Russian clinical reports — epilepsy, vascular encephalopathy, traumatic-brain-injury asthenia, vertebrobasilar insufficiency, dental premedication — is indexed on PubMed without abstracts; they are listed in the source library as title-only records and are not counted as efficacy evidence here.
Read as a whole, the corpus has a structure a reviewer must state out loud: every human efficacy study indexed in PubMed is Russian-language and single-country; with one exception, none is a blinded RCT; the two largest datasets lack control randomisation. In the broader evidence-based-medicine literature, effect sizes from such designs run high on average. No group outside Russia, and no group unaffiliated with the manufacturer’s market, has replicated any of it in the indexed record.
Family context, verbatim: for the parent compound piracetam, the Cochrane reviewers concluded that “the evidence available from the published literature does not support the use of piracetam in the treatment of people with dementia or cognitive impairment. Although effects were found on global impression of change, no benefit was shown by any of the more specific measures.” No Cochrane review of phenylpiracetam exists — which is itself a finding.[25]
The reviews of the molecule itself sit one tier below trials: the 2024 Gromova–Torshin narrative review surveys claimed anxiolytic, antiasthenic, antidepressant, anti-inflammatory and anticonvulsant effects; a 2023 review argues combination therapy is “pathogenetically justified” — a mechanism argument, not trial data.[23][24]
The honest summary: promising open signals from one country’s literature, one mixed double-blind trial, a tolerance question, and no human performance data at all. Clinical decisions belong with a qualified clinician; this site supplies the reading, not the prescription.