PEA 600-1200mg for Shift-Worker Neuropathy: NNT 1.5, and Why Most Pain Clinics Have Never Heard of It

The first time I noticed my feet had stopped screaming, I was three hours into a day shift, leaning against a nurse's station at 5:42 AM watching the night crew chart their last vitals. At 308 pounds, that lean used to be a survival posture. By hour eight on hospital tile, my plantar fascia felt like wet rope dragged across asphalt. I had built a whole micro-economy of compensations: shifting weight to the lateral edge of the right foot, locking the left knee for thirty-second relief windows, the slow shuffle from radiology back to the ED that I told myself was deliberate pacing.

Nine and a half months and 112 pounds later, at 196, the foot thing was gone. Not better. Not managed. Gone, like the weight itself had been the entire conversation.

What I could not stop noticing after that was the two night-shift nurses I was talking to. Neither was overweight. One was a former college soccer player. Both had been describing burning, tingling, electric-shock pain in their feet for months, and one had been on gabapentin since August and was tapering herself off because she could not finish a sentence at hour nine without losing the thread. Their problem was not load. Their problem was twelve-hour shifts on concrete subfloor, circadian disruption, and a peripheral nervous system that had started signaling pain in the absence of acute injury. That is neuropathy. The fitness industry has almost nothing to say about it, and the drug most of them get handed is mediocre.

This is the molecule I wish more of them knew about.

What PEA actually is

Palmitoylethanolamide is a fatty acid amide your body manufactures on demand at sites of tissue stress. Egg yolk has it. Peanut meal has it. It was first isolated in 1957, characterized for anti-inflammatory activity in the 1990s, and then ignored by the U.S. supplement market for thirty years because it is not patentable and does not get you high. Italy has been prescribing it as a medical food (Normast, PeaPure, Glialia) since 2008.

Mechanistically, the current model describes three converging pathways, all of which gabapentin misses entirely.

First, PEA is a high-affinity agonist at peroxisome proliferator-activated receptor alpha (PPAR-alpha), a nuclear receptor inside immune and neuronal cells that downregulates pro-inflammatory gene transcription, including the NF-kB cascade that drives chronic neuroinflammation. This is the same receptor family fenofibrate targets for lipid management, but PEA's binding profile is restricted to the inflammatory cascade rather than the metabolic one.

Second, PEA stabilizes mast cells. Mast cell degranulation is one of the underappreciated upstream events in peripheral neuropathic sensitization. When a sensory nerve is mechanically irritated for hours at a time, mast cells in the surrounding connective tissue release histamine, tryptase, TNF-alpha, and nerve growth factor. You see this elevated in CIDP, fibromyalgia, long-haul post-viral pain syndromes, and the kind of pressure-and-vibration neuropathy that builds in people who stand on hard floors for a decade.

Third, PEA potentiates the endocannabinoid system through what is called the entourage effect. It does not bind CB1 or CB2 directly. It occupies the FAAH enzyme that would otherwise degrade anandamide, lengthening the endogenous cannabinoid signal at the site of injury. No high. No tolerance curve. No dependency.

Gabapentin, by contrast, blocks the alpha-2-delta subunit of voltage-gated calcium channels and blunts excitatory neurotransmission across the entire central nervous system. The on-label adverse event profile includes somnolence in roughly 20 percent of users, dizziness in 17 percent, peripheral edema in 8 percent, weight gain, ataxia, and a cognitive slowing that is hard to quantify but consistently reported in patient surveys. Cochrane updated its peripheral neuropathy meta-analysis in 2017 and put gabapentin's NNT for 50 percent pain reduction at roughly 6 to 8.

The numbers that should have changed practice

Paladini and colleagues published a meta-analysis in Pain Physician in 2016 pooling 12 randomized controlled trials, 1,484 patients, across diabetic neuropathy, sciatica, post-herpetic neuralgia, carpal tunnel, and chemotherapy-induced peripheral neuropathy. The pooled NNT for 50 percent pain reduction was 1.5. That means treating between one and two patients to get one responder at the clinically meaningful threshold. That is roughly four times better than gabapentin on the same endpoint, with an adverse event profile statistically indistinguishable from placebo.

Cocito and colleagues, writing in CNS Neuroscience and Therapeutics in 2014, ran a 30-day trial in patients with sciatic nerve pain. The PEA arm dropped VAS scores by 50.1 percent. Placebo dropped 34.9 percent. The placebo effect in pain studies is famously enormous, and PEA still cleared it cleanly, with separation visible by day 10 and nearly half the active group crossing the 50 percent threshold that defines clinically meaningful relief.

Dosing in both data sets clusters around 600mg twice daily. Some of the diabetic neuropathy and CIDP-adjacent literature pushes to 1200mg twice daily for the first 30 to 60 days as a loading phase, then drops to 600mg twice daily for maintenance. Onset is slow. Most patients do not feel meaningful change until week three, and the curve keeps improving through week eight. This is not an analgesic you take for a flare. This is a signaling reset you take on a schedule.

Why this matters for shift workers and CIDP-adjacent populations

Standing at the unit handoff at 0700 you see the same neurological pattern on repeat: night-shift nurses, six to eighteen months into a rotation, developing burning feet, stocking-distribution numbness, and the specific gait change that says the medial plantar nerve is irritated. Twelve hours of standing on concrete under fluorescent light, with circadian disruption suppressing the cortisol rhythm that normally regulates inflammatory tone, produces a low-grade peripheral neuropathy that is mechanistically distinct from diabetic neuropathy but presents identically.

The CIDP-adjacent population is a separate concern. Chronic inflammatory demyelinating polyneuropathy patients, and the broader cohort of people with autoimmune-driven small fiber neuropathy, are managed primarily with IVIG, subcutaneous immunoglobulin, or corticosteroids. Pain is treated as a secondary symptom, usually with the gabapentin-pregabalin-duloxetine ladder. Several open-label studies, including a 2014 Italian observational cohort in carpal tunnel patients and a 2017 paper on diabetic peripheral neuropathy in the Journal of Pain Research, suggest PEA functions as a useful adjunct in patients who cannot tolerate the cognitive cost of central nervous system depressants.

Ultramicronized versus standard: this part actually matters

PEA is fat soluble and poorly absorbed in its native crystalline form. Particle size determines how much of an oral dose reaches plasma. Standard PEA powder runs 100 to 700 micrometers. Micronized PEA sits around 6 micrometers. Ultramicronized PEA, the form used in essentially every positive Italian RCT including Cocito's, runs roughly 0.8 to 2 micrometers.

Petrosino and Di Marzo published a 2017 review in the British Journal of Pharmacology that quantified the difference. Ultramicronized PEA achieves roughly 1.75 times the plasma concentration of standard PEA at equivalent oral doses, with a substantially flatter absorption curve and measurably better penetration into nervous tissue. Impellizzeri 2014 in the Journal of Neuroinflammation showed similar separation at the tissue level.

If you buy bulk PEA powder off a U.S. supplement aggregator at $18 for 250 grams, you are almost certainly getting standard particle size, and you are eating two to three times the dose to get equivalent tissue exposure. The brands that publish particle size and have third-party testing cost more per gram and are worth it. Look for the words "ultramicronized" or "um-PEA" on the label, plus a particle size spec. If the label does not specify, assume the absorption is roughly half of what the trial data describes. This is the single most common reason a PEA protocol underperforms.

Where this fits in a shift-worker stack

The protocol that aligns with the published trial data is 600mg ultramicronized PEA twice daily for the first 30 days, taken with a fat-containing meal. For CIDP-adjacent or severe diabetic neuropathy presentations, push to 1200mg twice daily for the loading phase, then drop to 600mg twice daily for maintenance. Most responders feel meaningful relief between days 14 and 21, with continued improvement through day 60.

For the night-shift nurse already on gabapentin who wants out, the approach with the most clinical traction is to add ultramicronized PEA at 600mg twice daily for 60 days while holding gabapentin steady, then taper gabapentin by 25 percent every two weeks.

Pair it with the nutrient floor that supports peripheral nerve function:

• Alpha-lipoic acid 600mg in the morning. Ziegler 2011 in Diabetes Care put symptomatic diabetic neuropathy NNT around 4.
• Methylcobalamin 1mg sublingual, not cyanocobalamin. The cyano group has to be cleaved before the methyl donor is bioavailable.
• Methylfolate 400 to 800mcg for one-carbon metabolism.
• Magnesium glycinate 300 to 400mg at night for NMDA tone.
• Vitamin D3 5000 IU with K2 MK-7 100mcg at the largest fat-containing meal, dosed to 25-OH serum above 50 ng/mL. Vitamin D below 30 ng/mL independently worsens small fiber neuropathy and is functionally universal in night-shift workers who never see the sun.

None of this fixes the root problem, which is twelve hours on concrete during a circadian window when your nervous system was built to be repairing itself. PEA is not a fix for compressive neuropathy from a herniated disc, and it is not a substitute for losing the mechanical load if the load is the variable. But for a nurse three years into night shifts who cannot change the schedule, or a CIDP patient whose pain breaks through immunoglobulin therapy, it gives the nervous system enough of a chemical floor to stop misfiring without the cognitive tax that makes the standard-of-care drug feel like trading one disability for another.

The molecule has been sitting in the literature for three decades. The meta-analysis has been sitting there since 2016. It is not exotic. It is not a peptide. It is a fatty acid your liver makes when you are stressed, sold over the counter, with an NNT that should embarrass the prescribing guidelines. Sixty days at full ultramicronized dosing before you judge the response. Start there.