DSIP

Many signalling peptides can be traced back to a gene, a parent protein and a receptor that receives them. For DSIP, as of the 2006 review, none of the three had been found. The 2006 review called the hypothesis of DSIP as a sleep factor extremely poorly documented and still weak, largely because of that gap.

DSIP is also a loner. Its structure does not resemble any known peptide family. Antibodies that recognize it light up specific hormone-making nuclei of the hypothalamus in many vertebrates, but those regions are not especially involved in sleep. The review’s authors proposed that a related, still-undiscovered peptide may explain much of what has been attributed to DSIP.

A second problem is survival. In 1987, the Basel group incubated DSIP in human and rat blood and found that it was broken down quickly. Two modified versions, including a phosphorylated analogue, degraded more slowly and persisted longer. The authors raised the possibility that this explains the sometimes stronger and more consistent effects of DSIP analogues.

The third thread involves opioids. Earlier animal work found that DSIP, like morphine and alcohol, induced slow-wave sleep when injected into a specific brainstem circuit, and naloxone, an opioid blocker, reversed it. A 2018 rat study found that naloxone also blunted the memory benefit of phosphorylated DSIP. These hints have never added up to a defined mechanism.¹⁵⁷⁹

The high point came in 1981, when The Lancet published a short report from the Basel group titled synthetic delta-sleep-inducing peptide improves sleep in insomniacs. It was brief, early and from the team that discovered the peptide, and it set the tone for much of what followed.

Independent tests followed. In 1987, a separate team gave DSIP or placebo intravenously to chronic insomniacs over four nights in a double-blind crossover design and recorded their sleep in the lab. Awakenings, time to fall asleep and time awake during the night fell, but not significantly compared with baseline or placebo nights. Total sleep rose a little, almost entirely as light stage 2 sleep, but the stage 2 difference between DSIP and placebo nights was already there at baseline. The authors concluded the improvement was of little clinical significance.

In 1992, researchers at the University of Amsterdam ran a double-blind trial in 16 people with chronic insomnia, who slept five nights in the lab. Sleep efficiency was slightly higher and sleep onset slightly faster with DSIP than with placebo. On closer analysis, the effects were weak and partly reflected a chance change in the placebo group. Subjective sleep quality did not change.

Their conclusion was blunt. Short-term DSIP was not likely to be of major benefit in chronic insomnia. None of the studies cited here is a later human insomnia trial that overturns it.²³⁴

The 1987 trial did more than measure total sleep. It broke the night into stages. Stages 3 and 4, the slow-wave sleep that gave delta sleep-inducing peptide its name, did not change under DSIP. REM sleep did not change either. The small gain in sleep time came from stage 2, the lighter sleep that fills much of the night, and even that difference from placebo was already present at baseline, so it is hard to credit to DSIP.

The 2006 review adds a twist from animal research. In early studies by the authors’ own group in rabbits and rats, certain artificial structural analogues of DSIP significantly promoted slow-wave sleep. DSIP itself did not. A naturally occurring dermorphin-related peptide, sharing five of DSIP’s nine positions, also promoted deep sleep in rabbits.

Put those findings beside the 1987 blood study, which found that analogues last longer in blood than DSIP itself, and the 2018 rat study, in which a phosphorylated analogue increased both non-REM and REM sleep at simulated high altitude. A pattern appears. The effects that match the name tend to belong to modified relatives of DSIP rather than to the original molecule.

This is a hypothesis, built from a handful of small studies, and none of the studies cited here tested an analogue in people with insomnia. But it reframes the question. When people say DSIP, the evidence suggests the interesting biology may sit with sturdier cousins of the peptide, or with the undiscovered DSIP-like peptide the 2006 review proposed, and not with the nine amino acids sold under the name.³¹⁵⁹

The opioid link inspired a second line of human work. If DSIP acted like morphine and alcohol in that brainstem circuit, perhaps it could ease withdrawal from them. In 1984, a clinical team gave DSIP intravenously to 107 inpatients, 47 in alcohol withdrawal and 60 in opiate withdrawal.

Among those who could be evaluated, symptoms disappeared or improved markedly and rapidly in 97 percent of opiate patients and 87 percent of alcohol patients, according to physicians and nurses. Anxiety eased more slowly. Tolerance was good, apart from headaches in a few patients. There was, however, no control group, and withdrawal symptoms naturally fade with time and care.

The same issue of European Neurology carried a pilot study in 7 people with chronic, severe pain, including migraine, vasomotor headache, chronic tinnitus and psychogenic pain attacks. Pain levels fell significantly in 6 of the 7 compared with their own earlier records, and depressive symptoms eased too.

Results like these are why DSIP drew interest beyond sleep, and why that interest faded. Uncontrolled studies in a few dozen or a hundred people can only suggest. None of the studies cited here is a controlled trial that confirms them.⁷⁸

Interest in DSIP did not end in the 1990s, but it moved back to the lab. In 2018, researchers in Delhi exposed rats to simulated high altitude, where oxygen is scarce and sleep falls apart. A phosphorylated form of DSIP improved both non-REM and REM sleep and restored memory in the water maze, along with a memory-linked signalling protein in the hippocampus called CREB.

In 2021, a Russian team gave rats DSIP through the nose before and after an induced stroke. The damaged area of brain was smaller in treated animals, though not significantly. Motor coordination on a rotating rod recovered significantly faster.

Both studies are interesting, and both are in rats under extreme conditions. Neither tells us what DSIP does for an ordinary adult who sleeps badly, which is the question most readers actually have.

They also share a limit. Each tested one model in one lab, and neither result has been repeated by another group in the studies reviewed here. Encouraging rat results often fade in people, and DSIP’s own history, from the confident 1981 report to the weak 1992 trial, is a small example of exactly that.⁹¹⁰

The human record is a set of small studies from roughly 1981 to 1992. Two controlled insomnia trials, with 16 people in one, found weak effects or effects of little clinical significance. Two uncontrolled studies, in 107 withdrawal patients and 7 pain patients, reported striking improvement without a comparison group.

That pattern is common in medicine. Open studies look strong, and controlled studies shrink the effect. For DSIP, the controlled evidence points to small, inconsistent changes in sleep and no change in the deep sleep the peptide was named for. None of the studies cited here revisits it with current methods.

It also helps to look at who ran which study. The 1981 Lancet report and the 1987 blood-stability work came from the Basel group that discovered DSIP. The two double-blind insomnia trials came from separate teams, and they found the weakest effects. On its own, that pattern proves nothing about bias. It is still a good reason to want replication before trusting the early results.³⁴⁷⁸

Most DSIP studies were too short and small to define a safety profile. The largest, the 1984 withdrawal study, reported good tolerance with headaches in a few patients. The sleep trials, run over a few nights, focused on sleep measures rather than side effects.

FDA’s April 2026 compounding safety page lists emideltide, the generic name for DSIP. It states that FDA found no safety information for the proposed route and notes risks from immunogenicity, an immune reaction to the peptide, and from peptide-related impurities.

The 1987 blood study showed DSIP breaks down quickly once it is in blood. Separately, nothing in the research cited here tested what vials sold online actually contain, so identity, purity and strength are unknown. People taking sedatives or opioids, or in recovery from alcohol or opioid use, should not experiment with a peptide studied partly for its links to the opioid system.⁷⁴¹²⁵

If you are reading about DSIP, you probably sleep badly. A clinician’s assessment comes first, and some contributors can show up on a blood test. TSH screens the thyroid, because an overactive gland produces light, restless sleep. Cortisol is worth checking only when an adrenal problem is suspected, since one morning reading cannot describe your daily rhythm or explain insomnia.

Ferritin measures iron stores. Low ferritin is linked with restless legs, an urge to move the legs at night that breaks up sleep. Red-cell magnesium gives a partial read of a mineral involved in nerve and muscle function. None of these is a DSIP marker, and an abnormal result does not by itself prove the cause of poor sleep, but thyroid disease and low iron are common, treatable contributors.

This is how Vitals Vault approaches sleep and peptide care. A blood panel at Quest Diagnostics comes first, and a licensed clinician reviews your results and history before discussing options. Since DSIP is not available through compounding, finding the treatable cause is the most direct route to better sleep.¹²¹³

DSIP, listed by FDA as emideltide, is not FDA-approved for any use. In July 2026, FDA’s Pharmacy Compounding Advisory Committee voted against adding it to the 503A bulks list, and it is not one of the six peptides covered by the October 29, 2026 enforcement discretion.

That leaves no standard US prescription route. Vials sold online as research use only are not made for people.¹²¹³

DSIP is a peptide with a memorable name, a 1977 discovery story and a human record that never matched either. Its controlled sleep trials found weak effects, the 1987 trial found no change in deep sleep, and the most sleep-like effects in animals came from modified analogues.

If sleep is the problem, the better first question is what is breaking it up. Light, timing, stress, iron and thyroid are all measurable. And if a peptide comes up, ask which molecule the evidence was actually about.³⁴¹