This is the oldest research programme in the palmitoylethanolamide literature and one of the strangest. Long before anyone marketed PEA for nerve pain, it was tested in Czechoslovak army units in the 1970s as a prophylactic against respiratory infection, under the name Impulsin.
Those trials reported large reductions in acute respiratory disease. Then the line of research effectively stopped for four decades, and was picked up again in 2023 by a modern randomized trial with 426 participants.
Both bodies of evidence are real and both come with limits that matter. The modern trial reports its headline episode count without a significance figure. The historical trials come from a single research programme, in young military men, and have never been replicated under modern methodology.
Key Takeaways
- A double-blind, randomised, placebo-controlled trial gave 426 participants either PEA 300 mg twice daily as Levagen+ or placebo for 12 weeks [1].
- The PEA group reported fewer infection episodes, 39 versus 64 [1]. The abstract reports this count difference without an accompanying significance figure.
- Significant reductions were reported for median symptom severity: scratchy throat 3 versus 7, and cough 2 versus 7 [1].
- Three serologically controlled double-blind field trials in army units in 1973, 1974 and 1975 reported acute respiratory disease rates of 22.7% vs 34.4%, 19.7% vs 40.7% and 10.6% vs 28.8%, all statistically significant [2].
- In the 1973 trial, during an influenza A2 outbreak, the manifestation rate among serologically infected subjects was 15.4% on the compound versus 44.9% on placebo [2]. Antibody formation was unaffected [2].
The Modern Trial
Upper respiratory tract infections are caused by bacteria or viruses, most commonly the common cold and influenza, and the sheer frequency of them keeps demand high for effective therapies with minimal side effects [1]. PEA is a signalling lipid that had previously been reported effective in improving the incidence of these infections, which is what prompted a modern trial [1].
The design was straightforward. 426 participants consumed either 300 mg of Levagen+ or a maltodextrin placebo twice daily for 12 weeks [1]. Participants completed the Wisconsin Upper Respiratory Symptom Survey 24 questionnaire daily from the moment symptoms began until they subsided [1].
That is a large trial by supplement standards, running across a full season, with symptom capture that starts when the participant notices something rather than at a scheduled visit.
What the Modern Trial Reported, Precisely
The PEA group reported fewer infection episodes than the placebo group, 39 versus 64 [1]. The PEA group also reported a significant reduction in the median severity score of symptoms for scratchy throat, 3 versus 7, and cough, 2 versus 7 [1]. The authors conclude the ingredient is safe and effective in reducing the incidence and symptoms associated with these infections [1].
Research note: A 1970s double-blind placebo-controlled trial in schoolchildren found fewer acute respiratory infections and sick days with PEA and a larger effect in younger children, per the review’s table [8].
Now read the two sentences against each other. Significance is stated explicitly for the severity scores. It is not stated for the episode count. A 39 versus 64 difference across 426 people looks substantial, and it may well be significant in the full paper, but the abstract does not say so, and the difference between “fewer episodes” and “significantly fewer episodes” is exactly the difference between an observation and a result.
The severity findings are the firmer part. A median scratchy throat score of 3 against 7 is a large relative difference on a symptom people can feel, and it is labelled significant.
Two structural caveats apply to the whole thing. Levagen+ is a commercially branded PEA ingredient, and this trial shares authorship with other trials of the same branded product, which is common in nutrition research and is a reason for independent replication rather than a reason to dismiss the work. And self-reported episode counts depend on participants deciding they are ill, which is not a blinded laboratory endpoint.
The 1970s Army Trials
The historical work is where the numbers get dramatic. Three serologically controlled double-blind field trials were run in army units and reported together in 1979 [2]. The compound was N-2-hydroxyethyl palmitamide, marketed as Impulsin, which is palmitoylethanolamide under an older name.
Evaluated on morbidity regardless of cause, the compound significantly reduced acute respiratory disease in all three trials [2]. In the 1973 trial with 901 volunteers, 22.7% of the Impulsin group developed acute respiratory disease against 34.4% on placebo, with P less than 0.0002 [2]. In 1974 with 610 volunteers the figures were 19.7% and 40.7%, P less than 0.002 [2]. In 1975 with 353 volunteers, 10.6% and 28.8%, P less than 0.004 [2].
The serological arm is what makes these trials unusual. By determining the actual infecting organism, the investigators could calculate a manifestation rate, meaning the proportion of people who actually became ill among all those with serologically proven infection [2]. In 1973, when influenza A2 England was prevalent, the manifestation rate was 15.4% in the Impulsin group against 44.9% on placebo, P less than 0.0002 [2]. Excluding people with high pre-infection antibody levels, the figures were 17.6% and 46.6% [2]. In 1974, with influenza B Hong Kong prevalent and negligible pre-existing antibodies, the manifestation rates were 14.3% and 57.1%, P less than 0.001 [2].
One further observation from those trials is mechanistically interesting: administration did not appear to influence the formation of antibodies [2]. Infection still happened and the immune system still responded. What changed was whether infected people became symptomatic.
An earlier report from the same Czechoslovak research programme was published in 1974 [3]. It has no abstract available in the PubMed record, so it is cited here for existence only and no figures are drawn from it.
Why the Historical Data Cannot Carry a Modern Claim
The 1970s results are strong on their face and still should not be treated as settled.
They come from one research programme, in one country, in one era, without independent replication anywhere since. The population was army volunteers, overwhelmingly young and male and living in close quarters, which is a setting with unusually high transmission and does not generalise cleanly to a family household. The 1975 trial’s serological results were unsatisfactory by the investigators’ own account, with antibodies against the circulating strain found in only 24.5% of respiratory disease cases, leaving that trial’s difference aetiologically unclarified even though it was statistically significant [2].
Methodology has also moved. Trial registration, pre-specified endpoints, published protocols and independent data monitoring did not exist in their current form in 1973. That does not make old results wrong. It does mean they are hard to audit.
What This Justifies Saying
The defensible summary is that PEA has a genuine and unusually long respiratory research history, with a modern randomized trial reporting significantly reduced symptom severity and a lower episode count, and historical trials reporting substantial reductions in illness rates among infected people.
What it does not justify is “PEA prevents flu”. The modern trial did not report a significance figure for episodes [1]. The historical trials are unreplicated and from a single programme [2]. And nothing here compares PEA with vaccination, which is the intervention with the actual evidence base for influenza.
Reviews describe PEA as well tolerated in humans, but most human studies were designed to test efficacy rather than safety and were short, so long-term safety data are thin [6][7]. Formulation remains a live issue, since poor oral bioavailability was the major obstacle in early research and is what modern delivery systems were built to address [5], while the pharmacokinetics, tissue distribution and excretion pathways remain poorly characterised [4].
Worth reading alongside this: the same branded ingredient, from the same research group, missed its primary endpoint in a seasonal allergy trial. This site covers that trial separately. Holding both results at once is the accurate picture of where this ingredient stands.
Frequently Asked Questions
Does PEA prevent colds?
A 426-person randomized trial reported fewer infection episodes on PEA, 39 versus 64, and significantly lower median severity for scratchy throat and cough [1]. The episode-count difference is reported without a significance figure in the abstract, so prevention is not an established claim.
What dose was used in the modern trial?
300 mg twice daily for 12 weeks [1].
What was Impulsin?
N-2-hydroxyethyl palmitamide, an older name for palmitoylethanolamide, tested prophylactically in three double-blind Czechoslovak army field trials between 1973 and 1975 [2].
Did the old trials stop people getting infected?
Not exactly. Infection still occurred and antibody formation was unaffected, but the proportion of serologically infected people who became symptomatic was far lower on the compound, for example 15.4% versus 44.9% during an influenza A2 outbreak [2].
Should this replace a flu vaccine?
No. Nothing in this literature compares PEA with vaccination, and the modern trial measured self-reported upper respiratory illness rather than laboratory-confirmed influenza [1].
References
- Rao A, Skinner J, Briskey D. The Efficacy of Palmitoylethanolamide (Levagen+) on the Incidence and Symptoms of Upper Respiratory Tract Infection: A Double Blind, Randomised, Placebo-Controlled Trial. Nutrients (2023). PMID 37892528
- Kahlich R, Klíma J, Cihla F, et al. Studies on prophylactic efficacy of N-2-hydroxyethyl palmitamide (Impulsin) in acute respiratory infections. Serologically controlled field trials. J Hyg Epidemiol Microbiol Immunol (1979). PMID 392005
- Masek K, Perlík F, Klíma J, Kahlich R. Prophylactic efficacy of N-2-hydroxyethyl palmitamide (impulsin) in acute respiratory tract infections. Eur J Clin Pharmacol (1974). PMID 4439865
- Rankin L, Fowler CJ. The Basal Pharmacology of Palmitoylethanolamide. Int J Mol Sci (2020). PMID 33114698
- Clayton P, Hill M, Bogoda N, et al. Palmitoylethanolamide: A Natural Compound for Health Management. Int J Mol Sci (2021). PMID 34069940
- Earle R. Nestmann Safety of micronized palmitoylethanolamide (microPEA): lack of toxicity and genotoxic potential. Food Science & Nutrition (2016). PMID 28265364
- Linda Rankin et al. Palmitoylethanolamide for the treatment of pain: pharmacokinetics, safety and efficacy. British Journal of Clinical Pharmacology (2016). PMID 27220803
- Vittori A et al. ALIAmides and pediatric health: a structured narrative review. Frontiers in pharmacology (2026). PMID 42812398
These statements have not been evaluated by the Food and Drug Administration. This information is not intended to diagnose, treat, cure, or prevent any disease. Content is for informational purposes only and is not medical advice; consult a qualified healthcare provider before starting any supplement. As an Amazon Associate we earn from qualifying purchases.