DSIP

Emideltide; DSIP; WAGGDASGE nonapeptide; Deltaran (Russian formulation), Delta Sleep-Inducing Peptide

Checked Oct 28, 2026

The short answer

Proven in humans for falling asleep faster and sleeping deeper

Grade A for sleep in people with insomnia, because three randomized placebo-controlled trials exist, but the result is mixed: one small trial reported better sleep, one found no difference from placebo, and the third found only a weak effect that its authors said could be partly a placebo-group change.

Approved?: Not approved for any use.
Safety: C, thinly characterized: FDA counted 209 people given DSIP by vein with no reports in its adverse-event database; in a 107-patient withdrawal series three had serious effects, two of them low blood pressure at the first injection, and nothing is known about long-term or subcutaneous use.

The full answer: what it is for, the dose, the units

What people take it for: People take it to fall asleep faster and sleep deeper. It is a nine-amino-acid peptide first isolated from the blood of sleeping rabbits in the 1970s.

Does it work for falling asleep faster and sleeping deeper? A, proven in humans. Result: mixed.

No approval for any use.

Safety: C, thinly characterized. FDA counted 209 people given DSIP by vein with no reports in its adverse-event database; in a 107-patient withdrawal series three had serious effects, two of them low blood pressure at the first injection, and nothing is known about long-term or subcutaneous use.

Do not use if: No formal contraindication list exists because DSIP is unapproved.

Could it be placebo? Of the three randomized trials, one reported better sleep than placebo in 6 people, one found no significant difference from placebo in 6 people, and one found a weak difference in 16 people that the authors partly credited to a change in the placebo group.

The dose people use, and the units to draw

Every human dose on record was given into a vein: 25 nmol per kg (about 21.3 mcg per kg) in the sleep trials, up to 150 nmol per kg across all studies FDA found. No named source gives a subcutaneous dose, and the microgram doses sold online have none.

Source	Dose	How often and how	Who it was for	Link
Schneider-Helmert and Schoenenberger 1981 (randomized, double-blind crossover)	25 nmol per kg (about 21.3 mcg per kg)	Intravenous over 4 minutes just before bedtime, on test nights versus placebo	6 middle-aged chronic insomniacs	abstract (dose); FDA briefing document for the randomized, double-blind crossover design
Monti et al. 1987 (randomized, double-blind crossover), as summarized by FDA	25 nmol per kg	Intravenous over 4 minutes, 4 nights of drug or placebo in random order, repeated after a 4-day washout	6 people with severe chronic insomnia	Section on clinical studies of effectiveness, Monti et al. 1987 paragraph
Bes et al. 1992 (randomized, double-blind, parallel groups)	25 nmol per kg	Intravenous on the afternoon before each of 3 nights, versus glucose placebo	16 chronic insomniac patients	abstract
Schneider-Helmert 1986 and 1987 (externally controlled), as summarized by FDA	30 nmol per kg	Intravenous nightly for 5 to 7 nights	14 and 18 chronic insomniacs; FDA treats the controls as an external healthy group	Section on clinical studies of effectiveness
FDA tally of published human studies	25 to 150 nmol per kg	All intravenous, for 1 to 15 days; FDA found no human study of the subcutaneous route	209 people across the published studies	Clinical Studies Assessing Safety paragraph

From 25 nmol per kg intravenously in the sleep trials (about 21.3 mcg per kg) up to 150 nmol per kg intravenously in the FDA's tally of published studies; none by subcutaneous injection.

Neither Huberman Lab episode searched (Apr 2024, Jun 2026) mentions DSIP, and Examine.com could not be opened (rate limited). The site's earlier subcutaneous ranges, starting dose and cycle patterns had no named source, and its earlier per-kg to flat-dose conversion was arithmetically wrong.

Reported by the named source, never prescribed by us.

How many units do I draw?

On a U-100 insulin syringe, units = 100 x dose (mg) / (vial mg / water mL). Enter your vial size, the water you add and the dose in the calculator and it does the arithmetic. Open the dose calculator

Why it might work

Assumed and still unproven. DSIP is amphiphilic and is hypothesized to modulate central sleep-regulating circuits, possibly via NMDA-receptor interaction and effects on neurotransmitter/neuroendocrine systems (lowering ACTH/cortisol, influencing GH/LH/somatostatin in animal data), rather than acting as a direct sedative. No specific DSIP receptor, gene, or precursor has ever been identified, so the mechanism remains theoretical. The persistence of next-day effects despite rapid blood clearance is taken to imply downstream/second-messenger action.

The trials, claim by claim

The letter counts randomized trials whether they worked or not; for DSIP the result tag is 'mixed' because the three sleep trials disagreed.

Status

DSIP is a natural nonapeptide (sequence Trp-Ala-Gly-Gly-Asp-Ala-Ser-Gly-Glu) first isolated in the 1970s by Schoenenberger and Monnier from cerebral venous blood of rabbits in induced sleep.

Well-established and uncontested fact of biochemistry/history, documented across reviews and reference sources.

PubMed 6145137: Graf and Kastin, Delta-sleep-inducing peptide (DSIP): a review (1984); update at PMID 3550726 (1986)

Small human IV studies in the 1980s found modest improvements in sleep latency, sleep efficiency and slow-wave sleep, but effects were weak and partly attributable to placebo-group changes.

Result: mixed

Several small double-blind human trials exist (single-dose and repeated dosing, healthy volunteers and chronic insomniacs), but they were small, used IV dosing, and reported weak/inconsistent benefit; authors themselves doubted major therapeutic value.

PubMed 1299794: Bes et al., DSIP on sleep of chronic insomniac patients, double-blind, 16 patients (1992); also 7028502 (1981, 6 insomniacs)

Status

After ~50 years no DSIP gene, precursor, or specific receptor has been identified, and the link between DSIP and sleep has never been definitively characterized.

This is the central conclusion of the 2006 review explicitly titled 'DSIP: a still unresolved riddle.'

PubMed 16539679: Kovalzon and Strekalova, DSIP: a still unresolved riddle (J Neurochem 2006)

Reported benefits beyond sleep (analgesia, stress-limiting/neuroprotective, antioxidant, anti-tumor, anticonvulsant, opioid/alcohol withdrawal) rest mainly on animal and in-vitro work.

Result: not tested in humans

These claims derive overwhelmingly from rodent and cell studies plus a small number of older Eastern-European clinical reports; they are not established by Western controlled human trials.

PubMed 3550726: Graf and Kastin, DSIP: an update (Peptides 1986); withdrawal series PMID 6548969

Status

DSIP (Emideltide) is not FDA-approved and not lawfully compoundable; it was on the FDA 503A Category 2 bulk-substance list, and on July 24, 2026 the PCAC declined to recommend it for the 503A Bulks List.

Emideltide was the only one of the seven substances the committee did not recommend (6-7, one abstention). FDA's briefing document found insufficient evidence: a single narcolepsy case report, uncontrolled opioid-withdrawal studies, inconclusive insomnia studies, and human data collected by IV for a product proposed as a subcutaneous injection. No further FDA action found as of October 12, 2026.

FDA: July 23-24, 2026 PCAC meeting

Status

In FDA's 2026 review, 209 people had received IV DSIP in published studies; one 107-patient withdrawal series recorded three serious adverse effects, two of them low blood pressure at the first injection.

Primary FDA briefing document summarizing the original studies (PMID 6548969). Patients were in withdrawal, which confounds causality.

FDA briefing document, Emideltide, July 2026

Risks and the label warnings

Adverse events

FDA's 2026 review counted 209 people given IV DSIP and zero reports in its adverse-event database through March 2024. Insomnia studies reported it as well tolerated. In a 107-patient withdrawal series, nine had minor effects (sweating, headache, nausea, vertigo) and three had serious ones, including two cases of low blood pressure at the first injection; withdrawal itself confounds those reports.

Contraindications

No formal contraindication list exists because DSIP is unapproved. Prudent caution applies to pregnancy/breastfeeding (untested; DSIP-like material occurs in human breast milk), and to anyone on sedatives, sleep medications, or with unstable psychiatric or endocrine conditions, given the lack of interaction data and the hormone-modulating signals seen in animals.

Theoretical concerns

As an injectable peptide of gray-market origin, the FDA's stated concern is immunogenicity, the immune system reacting to the peptide or to manufacturing impurities, a class-level concern for compounded peptides rather than a DSIP-specific finding. Because DSIP touches stress-hormone (cortisol/ACTH), LH, GH and somatostatin pathways in animal data, theoretical endocrine perturbation cannot be excluded, though it has not been demonstrated as a clinical problem.

What nobody knows yet

The biggest unknowns are long-term safety (no chronic-use human data at all), product identity/purity from research-chemical suppliers, true effective dose and route (historic studies used IV, not the subcutaneous dosing sold today), and whether any subjective benefit exceeds placebo. With no identified receptor or gene, even the basic biology is unsettled.

What else could explain the result

Users almost never run DSIP alone for sleep: it is frequently stacked with melatonin, magnesium, glycine, GHK-Cu, GH-secretagogues (CJC-1295/ipamorelin), or taken alongside improved sleep hygiene and reduced alcohol/caffeine, any of which can drive the perceived improvement. Concurrent GH-peptides especially confound 'deeper sleep' claims.

Is the vial what it says?

Cross-checked against PubMed-indexed primary human trials and reviews (including the explicitly skeptical 2006 'unresolved riddle' review and the 2001 safety editorial), FDA 503A bulk-substance/PCAC regulatory materials, and community/vendor sources (clearly down-weighted). Where sources conflicted (e.g., half-life 7-8 min vs 15-30 min; immunogenicity emphasis), the disagreement is flagged rather than resolved falsely.

Half-life and storage

Half-life: Very short. In-vitro and plasma estimates cluster around ~7-15 minutes (aminopeptidase degradation), with some sources citing up to ~30 minutes after subcutaneous injection. Notably, reported pharmacodynamic effects on sleep/stress are said to persist 6-8 hours, far outlasting the peptide's measured time in blood, a disconnect that is itself part of the 'riddle.'

Once mixed: Sold as a lyophilized (freeze-dried) powder reconstituted with bacteriostatic water and gently swirled (not shaken) to avoid degrading the peptide. As a small unmodified peptide it is considered fragile: keep lyophilized powder cold, store reconstituted solution refrigerated at ~2-8 C (39 F), minimize light and heat exposure, and use within a few weeks. Freeze-thaw cycling and rough handling should be avoided.

What I noticed, and what people report

What people report

Volume of reports: mixed. Reports exist but contradict each other.

Moderate. DSIP is a fairly well-known 'sleep peptide' in biohacker/peptide communities with steady forum discussion and many vendor writeups, but it is a niche relative to mainstream sleep aids, and a large share of available content is commercial rather than independent user reports.

Low/inconsistent. The recurring community verdict is that DSIP is 'hit or miss': some report genuinely deeper sleep without grogginess, many report little or nothing, and a subset report vivid dreams or next-morning grogginess (often dose/timing dependent). This mirrors the mixed clinical literature.

Low-to-moderate. Much of the favorable signal originates from vendors and affiliate sites selling DSIP, which were down-weighted. Independent user reports exist but are subjective, uncontrolled, and frequently confounded by stacked compounds, so they constitute anecdote rather than evidence.

Commonly reported (anecdote): deeper, more 'solid' sleep and feeling more rested, with users emphasizing the absence of the hangover/grogginess they get from antihistamines or prescription hypnotics.
Just as commonly reported: 'it did nothing', a large share of users feel no clear effect, which is why the community broadly calls DSIP hit-or-miss and very individual.
Vivid or intense dreams are a frequent theme, described as interesting by some and sleep-disrupting by others, and said anecdotally to be more likely at higher doses.
A minority report mild next-morning grogginess, headache, or a 'heavy' feeling.

Reports are anecdote, not proof.

Can you tell if it is working?

Partly trackable

The only honest objective tracker is a sleep wearable (Oura/WHOOP) tracked against your own baseline, and even then the human signal is weak, largely unreplicated, and placebo-prone.

Bloodwork

No blood proxy

Nothing on a routine panel confirms this one. Anyone who says otherwise is guessing.

DSIP is an endogenous nonapeptide that acts centrally to modulate sleep architecture and the stress/HPA response. It has a very short plasma half-life, is not itself an analyte on any routine panel, and controlled human infusion studies show it does not reliably shift cortisol, ACTH, or other pituitary hormones, so it leaves no measurable fingerprint on standard bloodwork.

Read this before you spend on labs

Routine bloodwork cannot confirm DSIP is working. Its effect is on sleep and subjective calm/stress, a central-nervous-system, experiential endpoint that is captured only by a sleep study (polysomnography), wearable/actigraphy sleep metrics, or a validated sleep questionnaire, not by any LabCorp/Quest panel. The one plausible endocrine proxy people reach for, cortisol/ACTH, was specifically tested: controlled human DSIP infusion studies found no reliable effect on the HPA axis, so there is no hormonal backdoor either. Track efficacy with a sleep diary or wearable sleep-stage data, not labs. (Note: DSIP-immunoreactivity assays exist but are research-only and would only show the peptide is present, not that it is working.)

Other ways to track it

Sleep / wearableIndirect / noisy

Oura/WHOOP (or Apple Watch), sleep-onset latency, sleep efficiency, deep-sleep (slow-wave) minutes, nocturnal HRV, resting HR

If DSIP does anything in you, the plausible wearable signals are shorter sleep-onset latency and higher sleep efficiency versus your own baseline, possibly with higher nocturnal HRV and lower resting HR on nights it 'works.' Do NOT expect a reliable deep-sleep change: the human PSG trials were equivocal-to-negative on slow-wave/delta sleep despite the peptide's name, and consumer wearables stage deep sleep poorly, so that specific metric is the least trustworthy readout. A single-arm improvement cannot rule out a placebo effect or regression to the mean until a blinded on/off design separates them.

Sleep is the ONLY domain DSIP has ever been studied for in humans, and small 1970s-80s IV-infusion polysomnography trials are the entire evidence base. They were mixed: some reported shorter latency and better efficiency versus placebo, but effects on slow-wave (delta) sleep, the peptide's namesake target, largely failed to replicate. A consumer wearable's latency/efficiency/HRV/resting-HR metrics are the closest accessible proxy, though only an approximation of lab EEG. Flip side: controlled human infusion studies found no consistent cortisol or ACTH change, so stress-hormone (urine/blood) assays are NOT a valid readout here, sleep and autonomic metrics are the only defensible target.

Collect a baseline of nightly wearable data first. A single before/after cannot separate signal from placebo and regression to the mean, and the endpoints are highly suggestible, so read any change as directional only.How to do it right

Bottom line

DSIP has no validated blood marker, and its only human-studied action is on sleep, shown only in small, mixed, unreplicated 1970s-80s IV-infusion PSG studies whose effect on slow-wave (delta) sleep, the peptide's own namesake, largely failed to replicate. A consumer wearable (Oura/WHOOP) can track the right category of things, onset latency, sleep efficiency, deep-sleep minutes, nocturnal HRV, resting HR, but only as an imperfect proxy for lab EEG, and deep-sleep staging specifically is the least reliable wearable metric. The felt endpoints (falling asleep faster, feeling rested) are highly suggestible, so the honest verdict is partial: never trust a single before/after. Urine or blood cortisol/ACTH is NOT a valid readout, since controlled studies showed no consistent HPA-axis effect.

Logging baseline vs retest is the whole game. The Truth Tracker is our free log for bloodwork, DEXA, genetics, and wearable data, bring your own numbers.

How to track it

History

First documented human use: First documented human use was in the early 1980s: small double-blind crossover and parallel-group studies (Schneider-Helmert and colleagues, ~1981-1984) gave synthetic DSIP by slow IV infusion at 25 nmol/kg to healthy volunteers and chronic insomniacs. These were genuine human trials, but small (6-16 subjects each) and produced weak, partly placebo-confounded results. No modern, adequately powered, registered RCT of DSIP for sleep has been completed; no DSIP product is FDA-approved.

Update: the FDA advisory panel voted this one DOWN on July 24, 2026

Of the seven peptides reviewed at the July 23-24, 2026 PCAC meeting, this was the only one the committee declined to recommend for the 503A Bulks List, and the committee recommended the other six over its own agency's objections, so a no here is a meaningful signal. The vote was 6-7 with one abstention, as reported by the press; FDA has not posted a tally. The stated reason was unusually weak evidence: a single patient case report for narcolepsy and two small uncontrolled studies for opioid withdrawal, with the additional problem that the human studies used an intravenous route while the proposal was for subcutaneous injection. That last point generalizes, so it is worth sitting with: evidence collected by one route does not automatically carry to another. As of November 2, 2026, the draft enforcement-discretion policy HHS published in the Federal Register on October 30 does not name this peptide in the press and law-firm summaries we read (we could not open its appendix).

Vote tallies as reported by the press (STAT, McDermott Will and Schulte); FDA has not posted tallies, minutes or a transcript, and its meeting materials are on its meeting page. Rechecked November 2, 2026 against FDA and Federal Register primary pages: FDA has published no proposed rule, docket FDA-2025-N-6895 closed July 22, 2026, and HHS published a draft enforcement-discretion policy in the Federal Register on October 30, 2026 (docket HHS-OASH-2026-0496). Nothing here is final, so check the current policy and rule before acting.. FDA, PCAC meeting, July 23-24 2026

1974-1977Schoenenberger and Monnier (University of Basel) isolate DSIP from cerebral venous blood of rabbits put into electrically-induced sleep; sequence determined as the nonapeptide WAGGDASGE.
1981-1984First human trials: small double-blind IV studies (25 nmol/kg) in healthy volunteers and chronic insomniacs report modest, inconsistent sleep improvements.
1992Double-blind study in chronic insomniacs (PMID 1299794) finds only weak effects on sleep efficiency/latency, partly confounded by placebo-group change.
2001European Journal of Anaesthesiology editorial (Pollard & Pomfrett) calls DSIP 'incredibly safe,' noting no lethal animal dose and only transient headache/nausea/vertigo in humans.
2006Review titled 'DSIP: a still unresolved riddle' concludes the peptide's gene, receptor and sleep mechanism remain unidentified after three decades.
2020sDSIP becomes a popular gray-market 'sleep peptide' sold as lyophilized powder for self-administered subcutaneous injection (and nasal sprays), despite no modern efficacy trials.
April 2026FDA's interim Category 2 list is updated: Emideltide (DSIP) now shows as previously in Category 2 and withdrawn by its nominators, ahead of the July advisory review.
July 24, 2026The FDA Pharmacy Compounding Advisory Committee (PCAC) declines to recommend DSIP/Emideltide for the 503A Bulks List, the only rejection among the seven peptides it reviewed, citing thin and route-mismatched human evidence.

All sources

PubMed: DSIP human sleep / insomnia double-blind trials(Primary small human trials from the 1980s-90s (IV dosing, mixed/weak results); three of them were randomized, double-blind and placebo-controlled, which is what sets the evidence grade.)
PubMed 16539679: Kovalzon and Strekalova, DSIP: a still unresolved riddle (J Neurochem 2006)(Skeptical review; no receptor/gene found, sleep link uncharacterized after decades.)
PubMed: DSIP review (Graf & Kastin) and pharmacology(Background on structure, history, short half-life, and breadth of claimed (mostly animal) effects.)
ClinicalTrials.gov: DSIP / delta sleep-inducing peptide(Use to confirm the absence of modern registered efficacy trials for sleep.)
FDA Pharmacy Compounding Advisory Committee (PCAC) materials(Regulatory status: the PCAC voted Emideltide/DSIP down for the 503A Bulks List on July 24, 2026; immunogenicity raised as class-level concern.)
FDA briefing document, Emideltide-related bulk drug substances, July 2026(Primary FDA review: 209 people given IV DSIP, a 107-patient withdrawal series with three serious effects, and the evidence gaps behind the 6-7 vote.)
PubMed 11437870: Pollard and Pomfrett, Delta sleep-inducing peptide (Eur J Anaesthesiol 2001, editorial)(The editorial behind the 'incredibly safe' quote. The PubMed record has no abstract, so the quote was not checked against the text.)
Schneider-Helmert D, Schoenenberger GA. 1981 (PMID 7028502); design from the FDA briefing document, Emideltide, July 2026(dose table)
FDA briefing document, Emideltide-related bulk drug substances, PCAC July 23-24, 2026, summary of Monti et al. 1987(dose table)
Bes F, Hofman W, Schuur J, Van Boxtel C. 1992 (PMID 1299794)(dose table)
FDA briefing document, Emideltide-related bulk drug substances, PCAC July 23-24, 2026, summaries of Schneider-Helmert 1986 and 1987(dose table)
FDA briefing document, Emideltide-related bulk drug substances, PCAC July 23-24, 2026, clinical studies assessing safety(dose table)

Not medical advice. Grades reflect the evidence as of the date above and will change as data arrives. Reports from users are anecdote, not proof. Talk to a licensed clinician before you act on anything here.