DSIP: A 50-Year-Old Sleep Peptide Science Still Isn’t Sure About
DSIP was isolated from the cerebral venous blood of rabbits in 1977 by a Basel research group studying electrically induced delta-wave (deep) sleep, and named for the effect it appeared to reproduce when transferred to other rabbits. Nearly 50 years and several small human trials later, the peptide's own receptor has never been identified, and a 2006 scientific review found the evidence for DSIP itself, as opposed to related "DSIP-like" analogs, actually causing that effect to be genuinely weak. The human trials that exist are real, but they don't agree with each other, and one directly contradicts a claim commonly made about it today.
Reviewed by the PepGuard team · Last reviewed Jul 28, 2026
Synthetic linear nonapeptide: Trp-Ala-Gly-Gly-Asp-Ala-Ser-Gly-Glu · also referred to as Delta Sleep-Inducing Peptide, Emideltide
Photo: Jp Valery / UnsplashFour trials. They don’t agree with each other.
DSIP was isolated from rabbit blood in 1977 by researchers who had electrically induced delta-wave sleep and were chasing whatever circulated in the bloodstream during it. The name describes the experiment, not a confirmed mechanism. Nearly 50 years later, DSIP’s own receptor has never been identified, and a 2006 review in the Journal of Neurochemistry called the sleep-factor hypothesis behind it “extremely poorly documented and still weak,” noting that in some animal studies it was related “DSIP-like” analogs, not DSIP itself, that showed sleep-promoting activity. If that holds up, decades of safety and efficacy assumptions about synthetic DSIP may not even describe the molecule actually doing anything.
The four published human trials don’t resolve it. One small acute-dose study found real benefits, longer sleep, fewer interruptions, no next-day grogginess. The largest controlled trial, using the same dose nightly for four nights, concluded the improvement over placebo was of little clinical significance. A separate trial built specifically to test whether DSIP blunts the body’s stress-hormone response found no effect at all, directly contradicting a claim you’ll see made about it. None of this makes DSIP a scam. It makes it a genuinely unresolved case, which is different from the confident dosing charts most pages present it with.
What the trials used, none of it a nightly routine.
| What's cited | Range |
|---|---|
| Acute insomnia trial (Schneider-Helmert & Schoenenberger 1981, n=6, intravenous, single dose) | 25 nmol/kg, one injection |
| Repeat-dose crossover trial (Monti et al. 1987, chronic insomniacs, intravenous) | 25 nmol/kg nightly, for 4 nights |
| Open-label repeat-dose trial (Kaeser 1984, n=7, route not specified in the abstract) | 10 injections total, administered as a series |
| Hormone-response study dose (Späth-Schwalbe et al. 1995, n=15 combined, intravenous infusion) | 3-4 mg total, infused around a CRH challenge |
Four published human trials used four different protocols, and none of them map onto a "nightly sleep aid" routine the way most vendor pages imply. The two acute studies used a single intravenous 25 nmol/kg dose; the repeat-dose crossover trial used the same 25 nmol/kg nightly for four nights and still concluded the sleep benefit was of little clinical significance; the open-label study used a series of 10 injections over an unspecified interval, in a small uncontrolled group with a noted history of drug dependence in at least some patients. None of these trials tested, or endorses, an ongoing self-administered protocol, and none used a subcutaneous route, the one most vendors sell it for. We're not converting these intravenous research doses into a self-injection schedule, because none of the underlying trials did either.
Short trials, so mostly “unknown,” honestly.
| Signal | Frequency | Context |
|---|---|---|
| Daytime sedation | rarely reported | One acute-dose trial specifically noted its absence, along with a mild stimulating effect in the first hour post-injection, unlike many conventional sleep aids. |
| Adverse events across the published trials | rarely reported | None of the four human trials we found reported a significant adverse event, but none were designed or powered as dedicated safety studies, and the largest ran only 4 nights. |
| Effects of long-term or repeated self-administered use | unknown / not studied | The longest published trial was a 10-injection open-label series with no control group. No published trial has tested ongoing, indefinite self-administration the way it's commonly used today. |
| Whether the marketed peptide is the molecule responsible for any of the above | unknown / not studied | A 2006 review found that related "DSIP-like" analogs, not DSIP itself, showed sleep-promoting activity in some animal studies, and that DSIP's own receptor has never been identified. If that holds up, safety and efficacy data on synthetic DSIP may not even describe the right compound. |
Not FDA-approved. FDA staff recommended against adding it to the compounding-bulks list, and unlike six other peptides reviewed the same week, the agency's advisory panel agreed.
DSIP, referred to by the FDA under its more formal name emideltide, has never been approved by the FDA for any use. It was nominated for the agency's 503A Bulks List, the pathway that lets compounding pharmacies legally prepare a substance, by LDT Health Solutions (on behalf of the International Peptide Society) and separately by Wells Pharmacy Network, for chronic insomnia, opioid withdrawal, and narcolepsy. Both nominations were later withdrawn by the nominators, but FDA elected to review the substance anyway at its Pharmacy Compounding Advisory Committee meeting on July 23-24, 2026. FDA's own briefing document for that meeting states plainly: "FDA is proposing that Emideltide (free base) NOT be included on the 503A Bulks List" and the same for Emideltide acetate. Unlike the other six peptides FDA staff reviewed that same week (BPC-157, MOTS-c, KPV, TB-500, Epitalon, and Semax, all of which the advisory panel voted to recommend anyway, against staff's advice), Emideltide/DSIP was, per independent reporting from NPR and CBS News, the one compound the panel actually voted down, the only one where the panel agreed with FDA staff's recommendation. That's still an advisory vote, not an FDA decision, and the agency hasn't yet acted on any of the panel's July 2026 recommendations.
Every claim above, traced back.
Four published human trials, a 2006 scientific review questioning the molecule itself, and the FDA’s own briefing document from its most recent review. Click through and check them yourself.
DSIP questions, answered.
The evidence is genuinely mixed. A small acute-dose trial found real improvements, longer sleep, fewer interruptions, no daytime sedation, in 6 chronic insomniacs. But the largest controlled trial, a crossover study in chronic insomniacs given the same dose nightly for 4 nights, concluded the sleep improvement was "of little clinical significance." A 2006 scientific review still called the case for DSIP as a sleep factor "extremely poorly documented and still weak," almost 30 years after its discovery.
A dedicated 1995 trial tested exactly this, infusing DSIP intravenously around a CRH (stress-hormone) challenge in healthy men, and found no effect: ACTH and cortisol responses were almost identical between DSIP and placebo. That directly contradicts a stress-hormone-lowering claim you'll see made about it.
No. It was nominated for the FDA's 503A compounding-bulks list under the name emideltide, for insomnia, opioid withdrawal, and narcolepsy. FDA's own briefing document for its July 23-24, 2026 review recommended against including it. We haven't independently verified the committee's final vote, so check the current public record.
Probably the same synthesized sequence, but whether that sequence is actually responsible for any sleep-promoting effect is exactly what a 2006 scientific review questioned. It found that related "DSIP-like" analogs, not DSIP itself, produced sleep-promoting activity in some studies, and that DSIP's own receptor has never been identified, decades after its discovery.
Keep researching.
This is a preview of the DSIP report.
Create your account to check any peptide against the full research base.