top of page
Search

Palmitoylethanolamide (PEA): What the Evidence Actually Shows


An endogenous lipid mediator for neuroinflammation and chronic pain — mechanism, trial data, and where it belongs in an integrative plan


Yoon Hang Kim, MD, MPH

Board-Certified in Preventive Medicine | Integrative & Functional Medicine Physician

Most supplements marketed for pain have a story but no trial data. Palmitoylethanolamide is unusual: it has both. It is a molecule the human body already makes, it has a defined nuclear receptor target, and it has now been through more than a dozen randomized controlled trials and four independent meta-analyses. That is a stronger evidence base than several agents I am routinely asked to comment on — and it is still thin enough that the honest answer involves real caveats.

This article walks through what PEA is, how it works, what the controlled trials show (including the ones that failed), and how I think about it when a client with chronic pain or suspected neuroinflammation sits in front of me.

What PEA Actually Is

Palmitoylethanolamide is an N-acylethanolamine — the amide of palmitic acid and ethanolamine — produced on demand by cells in response to injury and inflammatory stress. It is present in human tissue, in breast milk, and in ordinary foods such as egg yolk, soybeans, and peanuts. It was first studied in the 1950s as an anti-inflammatory fraction of egg yolk, and it has been available in Europe for years as a food for special medical purposes.

Two points matter for framing. First, PEA is not a cannabinoid in the way people assume. It is structurally related to endocannabinoids and is often lumped in with them, but it is not a meaningful direct agonist at CB1 or CB2, and it is not psychoactive. Second, PEA is not an analgesic in the NSAID or gabapentinoid sense. It does not block a pain signal acutely. It modulates the non-neuronal cells — mast cells, microglia, astrocytes — that keep a pain signal amplified. That distinction explains almost everything about how it behaves clinically, including why it takes weeks to work.

Mechanism: Resolution Rather Than Suppression

The clearest mechanistic finding came from Piomelli’s group in 2005, which identified the nuclear receptor PPAR-α as the molecular target responsible for PEA’s anti-inflammatory effects. PEA activated PPAR-α in vitro and reduced inflammation in wild-type mice, but had no effect in mice lacking PPAR-α — a clean genetic demonstration rather than a correlation.1 Later work confirmed that PPAR-α also mediates PEA’s acute effects on sensory neurons.2

Downstream of PPAR-α activation, PEA restrains NF-κB-driven transcription of pro-inflammatory mediators, stabilizes mast cells, and dampens microglial activation. Skaper and colleagues framed this as a strategy of resolution — supporting the body’s own mechanism for shutting an inflammatory response down — rather than blocking a single enzyme or receptor.3,4 This is also the basis of the older ALIA concept (autacoid local injury antagonism): a locally produced signal that rises after tissue insult and helps return the system to baseline.

A secondary, less firmly established mechanism is the so-called entourage effect: PEA appears to indirectly increase anandamide tone and to desensitize TRPV1 channels. This is biologically plausible and supported by preclinical work, but it should be presented as hypothesis rather than established human pharmacology.

If central sensitization and glial activation are driving a client’s pain, PEA is aimed at the right target. If the pain is purely mechanical and structural, the mechanism gives less reason to expect benefit.

The Clinical Evidence

The meta-analytic signal

Four meta-analyses have now been published, and they agree in direction while disagreeing in magnitude.

  • Artukoglu et al. (2017) published the first meta-analysis of PEA for pain, concluding that PEA reduced pain scores relative to control.5

  • Paladini et al. (2016) pooled raw data from controlled and open-label trials of micronized and ultra-micronized PEA. By day 60, 81% of PEA-treated participants had reached a pain score of 3 or below, compared with 40.9% of controls. The effect was independent of age, sex, and pain etiology.6

  • Lang-Illievich et al. (2023) restricted inclusion to double-blind RCTs — 11 trials, 774 participants — and found a pooled standardized mean difference of 1.68 (95% CI 1.05 to 2.31), with several trials also reporting improved function and quality of life and no significant adverse effects.7

  • A 2025 meta-analysis in Nutrition Reviews (18 studies, 1,196 participants) found significant pain reduction at 6 weeks, 8 weeks, and 24–26 weeks, with benefit across all three pain phenotypes: nociceptive, neuropathic, and nociplastic.8

An effect size near 1.0 or above is large for an analgesic — larger than most pain pharmacotherapy — and that should trigger appropriate skepticism. The included trials are small, heterogeneous in condition and outcome measure, and heavily weighted toward research groups and industry with a stake in the compound. The direction of effect is consistent and replicated; the precise magnitude is probably inflated.

Knee osteoarthritis

Steels and colleagues randomized 111 adults with mild-to-moderate knee osteoarthritis to 300 mg PEA, 600 mg PEA, or placebo daily for 8 weeks. Both active arms showed significant reductions in total WOMAC score (300 mg, p = 0.0372; 600 mg, p = 0.0012), with a dose-dependent pattern and improvements in secondary anxiety and sleep measures.9 This trial is important for a second reason discussed below: it used non-micronized PEA.

Diabetic peripheral neuropathy

Pickering and colleagues conducted a quadruple-blinded, placebo-controlled trial of 600 mg PEA daily for 8 weeks in 70 adults with type 1 or type 2 diabetes and painful peripheral neuropathy. They found significant reductions in Brief Pain Inventory total pain and pain interference and in NPSI total and sub-scores — with the notable exception of evoked pain (p = 0.09). Sleep quality improved, IL-6 and CRP fell, and 94% of participants completed the study.10

The evoked-pain result is the interesting one. It is consistent with the mechanistic story: PEA appears to act on the tonic, inflammation-driven component of neuropathic pain rather than on stimulus-evoked allodynia.

Where PEA failed

Any honest review has to include Andresen et al. (2016), published in Pain. Seventy-three individuals with spinal cord injury neuropathic pain received ultra-micronized PEA or placebo as add-on therapy for 12 weeks. There was no effect on the primary outcome.11

This is the most methodologically rigorous negative trial in the literature, and it is instructive rather than disqualifying. Established central neuropathic pain following long-standing cord injury is a different problem from active peripheral neuroinflammation — a point made directly in the published commentary on that trial. PEA is a resolution agent. It has the best chance where an inflammatory process is still running, and the least where the nervous system has already been structurally rewired.

The Formulation Question

Marketing in this space is emphatic that only micronized or ultra-micronized PEA works. The reasoning is real: PEA is lipophilic and poorly water soluble, and particle-size reduction improves dissolution and oral absorption. Preclinical work has shown micronized and ultra-micronized PEA to be superior to non-micronized PEA in a rat inflammatory pain model.12

The human data are less tidy. The Steels osteoarthritis trial and the Pickering neuropathy trial both used non-micronized or alternative-delivery formulations and both were positive, which the authors of the 2023 meta-analysis noted as an argument against micronization being strictly necessary in humans.7,9,10 Meanwhile, the negative spinal cord injury trial used the ultra-micronized product.11

My reading: particle size and delivery clearly influence bioavailability, but formulation is not the variable that separates responders from non-responders in the current human literature. Manufacturing quality, verified content, and third-party testing are the more defensible things to insist on.

Dosing and Time Course

Trial regimens have clustered narrowly. Most positive studies used 600 mg daily, given either as a single dose or as 300 mg twice daily, with some protocols using 1,200 mg daily during an initial loading phase before stepping down. Both 300 mg and 600 mg daily produced benefit in the osteoarthritis trial, with the higher dose performing better.9

Timeline matters more than dose, and it is the single most common reason for a false negative in practice. Meaningful separation from placebo generally appears between 4 and 8 weeks.8 A client who takes PEA for ten days, feels nothing, and stops has not actually tested it. I tell people at the outset to commit to a defined trial and to score their pain at baseline, so the decision to continue or stop is made on data rather than impression. A 2024 systematic review of extended treatment supports durability of benefit beyond the typical 8-week window.13

Safety and Tolerability

This is where PEA has its clearest advantage, and it is worth stating precisely rather than as “well tolerated.” Gabrielsson, Mattsson, and Fowler examined the safety literature critically — including the trials that reported it poorly — and concluded that for treatment durations up to 49 days, the available data argue against serious adverse drug reactions at an incidence of 1 in 200 or greater. For treatment beyond 60 days, the number of exposed participants is insufficient to rule out reactions at a rate below 1 in 100.14

That is a genuinely favorable profile with an honest boundary attached. Reported adverse effects are uncommon and mild, mostly gastrointestinal. No clinically significant drug–drug interactions have been documented, which is what makes PEA usable alongside conventional analgesics rather than in place of them. The realistic limitation is long-term safety data, not short-term risk.

Where PEA Fits in an Integrative Plan

I consider PEA when one or more of the following is true:

  • NSAIDs are unavailable or unwise — chronic kidney disease, GI bleeding risk, anticoagulation, cardiovascular risk, or simply a client who has been on daily NSAIDs for years and needs an exit strategy.

  • The clinical picture suggests neuroinflammation or central sensitization — pain out of proportion to imaging, widespread rather than localized symptoms, poor sleep, cognitive fog, flares tracking with immune activation.

  • Mast cell activation is part of the picture. PEA’s mast-cell-stabilizing mechanism makes it mechanistically attractive here, though I want to be clear that MCAS-specific controlled trial data for PEA do not yet exist. This is mechanism-based reasoning, not proven efficacy.

  • An additive, low-risk layer is needed alongside an existing regimen — including alongside low-dose naltrexone, where the glial-modulating rationale overlaps but the targets differ.

What PEA is not: a rescue analgesic, a replacement for addressing the driver of the pain, or a reason to skip a structural workup. The most common clinical error I see is adding PEA to an unexamined problem and calling that a treatment plan.

What We Still Do Not Know

  • Trial size and independence. Most studies are small and single-center, and a substantial share come from groups connected to manufacturers. Large, independent, multicenter replication has not happened.

  • Who responds. No validated predictor identifies likely responders in advance. Response appears to be roughly even odds, with a subset reporting substantial relief.

  • Beyond pain. Signals for mood, sleep, cognition, and post-viral symptoms are mechanistically coherent and preliminary. They should not be presented to clients at the same confidence level as the chronic pain data.

  • Long-term use. Reassuring short-term safety does not establish multi-year safety.

The Clinical Bottom Line

PEA is one of the better-supported non-pharmaceutical options for chronic and neuropathic pain: a defined mechanism, replicated positive RCTs across four meta-analyses, a favorable short-term safety profile, and no meaningful interaction burden. It is also modest in effect for many people, unproven in several of the conditions it is marketed for, and useless if stopped at two weeks.

Given that risk–benefit ratio, an 8-week structured trial at 600 mg daily with baseline and follow-up pain scoring is a reasonable thing to consider for the right client — as one layer of a plan, not as the plan.


Important: This article is educational and is not medical advice. It does not establish a physician–client relationship and is not a substitute for individualized evaluation. Supplements can interact with medications and are not appropriate for everyone, including during pregnancy and lactation. Do not start, stop, or change any treatment based on this article. Please discuss any therapy described here with your own licensed clinician before making a decision.


References

  1. Lo Verme J, Fu J, Astarita G, et al. The nuclear receptor peroxisome proliferator-activated receptor-alpha mediates the anti-inflammatory actions of palmitoylethanolamide. Mol Pharmacol. 2005;67(1):15-19. doi:10.1124/mol.104.006353. PMID: 15465922.

  2. Khasabova IA, Xiong Y, Coicou LG, et al. Peroxisome proliferator-activated receptor α mediates acute effects of palmitoylethanolamide on sensory neurons. J Neurosci. 2012;32(37):12735-12743.

  3. Skaper SD, Facci L, Giusti P. N-palmitoylethanolamine and neuroinflammation: a novel therapeutic strategy of resolution. Mol Neurobiol. 2015;52(2):1034-1042.

  4. Skaper SD, Facci L, Fusco M, et al. Palmitoylethanolamide, a naturally-occurring disease-modifying agent in neuropathic pain. Inflammopharmacology. 2014;22(2):79-94.

  5. Artukoglu BB, Beyer C, Zuloff-Shani A, Brener E, Bloch MH. Efficacy of palmitoylethanolamide for pain: a meta-analysis. Pain Physician. 2017;20(5):353-362. PMID: 28727699.

  6. Paladini A, Fusco M, Cenacchi T, Schievano C, Piroli A, Varrassi G. Palmitoylethanolamide, a special food for medical purposes, in the treatment of chronic pain: a pooled data meta-analysis. Pain Physician. 2016;19(2):11-24. PMID: 26815246.

  7. Lang-Illievich K, Klivinyi C, Lasser C, Brenna CTA, Szilagyi IS, Bornemann-Cimenti H. Palmitoylethanolamide in the treatment of chronic pain: a systematic review and meta-analysis of double-blind randomized controlled trials. Nutrients. 2023;15(6):1350. doi:10.3390/nu15061350. PMID: 36986081.

  8. Meta-analysis of palmitoylethanolamide in pain management: addressing literature gaps and enhancing understanding. Nutr Rev. 2025;83(7):e1604. PMID: 39798151. [Author list to be confirmed before publication.]

  9. Steels E, Venkatesh R, Steels E, Vitetta G, Vitetta L. A double-blind randomized placebo controlled study assessing safety, tolerability and efficacy of palmitoylethanolamide for symptoms of knee osteoarthritis. Inflammopharmacology. 2019;27(3):475-485. doi:10.1007/s10787-019-00582-9.

  10. Pickering E, Steels EL, Steadman KJ, Rao A, Vitetta L. A randomized controlled trial assessing the safety and efficacy of palmitoylethanolamide for treating diabetic-related peripheral neuropathic pain. Inflammopharmacology. 2022;30(6):2063-2077. doi:10.1007/s10787-022-01033-8. PMID: 36057884.

  11. Andresen SR, Bing J, Hansen RM, et al. Ultramicronized palmitoylethanolamide in spinal cord injury neuropathic pain: a randomized, double-blind, placebo-controlled trial. Pain. 2016;157(9):2097-2103. doi:10.1097/j.pain.0000000000000623. PMID: 27227691.

  12. Impellizzeri D, Bruschetta G, Cordaro M, et al. Micronized/ultramicronized palmitoylethanolamide displays superior oral efficacy compared to nonmicronized palmitoylethanolamide in a rat model of inflammatory pain. J Neuroinflammation. 2014;11:136. doi:10.1186/s12974-014-0136-0. PMID: 25164769.

  13. Schweiger V, Schievano C, Martini A, et al. Extended treatment with micron-size oral palmitoylethanolamide (PEA) in chronic pain: a systematic review and meta-analysis. Nutrients. 2024;16(11):1653. doi:10.3390/nu16111653. PMID: 38892586.

  14. Gabrielsson L, Mattsson S, Fowler CJ. Palmitoylethanolamide for the treatment of pain: pharmacokinetics, safety and efficacy. Br J Clin Pharmacol. 2016;82(4):932-942. doi:10.1111/bcp.13020. PMID: 27220803.

  15. D’Amico R, Impellizzeri D, Cuzzocrea S, Di Paola R. ALIAmides update: palmitoylethanolamide and its formulations on management of peripheral neuropathic pain. Int J Mol Sci. 2020;21(15):5330. doi:10.3390/ijms21155330. PMID: 32727084.

  16. Polati E, Martini A, Schweiger V. Ultramicronized palmitoylethanolamide treatment in central neuropathic pain following longstanding spinal cord injury: try to extinguish the fire after everything was burned. Pain. 2017;158(4):763-764. PMID: 28301405.

About Dr. Kim

Yoon Hang Kim, MD, MPH is a board-certified physician with more than 20 years of clinical experience. He completed a fellowship at the University of Arizona under Dr. Andrew Weil and holds board certification in Preventive Medicine along with certifications in medical acupuncture and integrative and holistic medicine. His clinical focus includes low dose naltrexone (LDN), autoimmune conditions, chronic pain, integrative oncology, fibromyalgia, chronic fatigue syndrome, mast cell activation syndrome (MCAS), and mold toxicity.

He is the author of four books and more than 20 peer-reviewed articles, including MCAS: Epidemic in Plain Sight and LDN Primer, both available on Amazon. He is also the founder of the LDN Support Group.

Yoon Hang Kim, MD, MPH | Board-Certified in Preventive Medicine | Integrative & Functional Medicine Physician

 
 
 

Comments

bottom of page