A literature review of DSIP dosage: why the human studies all used 25 nmol/kg intravenously, how that converts to micrograms, reconstitution math, and DSIP's minutes-long plasma half-life.
The DSIP dosage literature is unusually narrow. Nearly every human study of delta sleep-inducing peptide converged on one quantity — 25 nmol/kg body weight, intravenously — and nearly all of it was published between 1981 and 1992 in cohorts of six to sixteen subjects. There is no dose-ranging programme, no modern replication, and no published human pharmacokinetics for any non-intravenous route. What follows is a reading of that literature, not a protocol.
DSIP is a nonapeptide, Trp-Ala-Gly-Gly-Asp-Ala-Ser-Gly-Glu (WAGGDASGE), with a molecular weight of roughly 848.9 g/mol. The dosing literature is written in nmol/kg, not mg — a convention inherited from the neuropeptide field that has to be converted before any vial math makes sense.
The conversion: 1 nmol of DSIP is about 0.849 µg, so 25 nmol/kg is roughly 21.2 µg/kg — about 1.5 mg for a 70 kg subject. Run the other way, 100 µg is approximately 118 nmol. Getting this wrong by a factor of 1,000 is the common arithmetic failure here; see the dosing math guide for the general mg/mcg/IU framework.
Acute single administration. Schneider-Helmert and colleagues (Int J Clin Pharmacol Ther Toxicol, 1981) gave 25 nmol/kg as a slow morning intravenous infusion to six healthy volunteers in a double-blind crossover. The companion study in six chronic insomniacs (Experientia, 1981) used the same acute dose and reported a timing detail worth noting: sleep-promoting effects appeared only in the second hour after injection, with a slight arousing effect in the first.
Repeated nightly administration. Monti and colleagues (Int J Clin Pharmacol Res, 1987) gave 25 nmol/kg intravenously across four nights, double-blind crossover, with polysomnography. Bes and colleagues (Neuropsychobiology, 1992) gave the same dose in the afternoon before each of three consecutive nights to 16 chronic insomniacs, parallel groups.
A separate indication, repeated dosing. Dick and colleagues used the same 25 nmol/kg intravenously in alcohol and opiate withdrawal (Neuropsychobiology, 1983; Eur Neurol, 1984), across 67 and then 107 inpatients — the only published context where injection frequency was titrated to clinical response rather than fixed, and more injections were needed for opiate than alcohol withdrawal.
Preclinical. Rodent work uses mg/kg rather than nmol/kg — for example 0.1 mg/kg intraperitoneally, roughly 118 nmol/kg, in ethanol-withdrawal sleep studies. Species and route differences make those figures poor translation material.
DSIP is cleared extremely fast. In an enzyme-immunoassay clearance study (Neuroendocrinology, 1984), intravenous DSIP had a mean plasma half-life of 4.0 ± 0.7 minutes in anaesthetised dogs, 2.9 minutes in a monkey, and 2.0 ± 0.54 minutes in rats, with clearance around 30.7 mL/kg/min in dogs. In vitro work attributes the rapid loss to aminopeptidase-like activity in plasma and serum.
Two consequences follow. There is no meaningful accumulation across doses, so steady-state reasoning borrowed from long-half-life compounds does not apply. And the reported pharmacodynamics are decoupled from the pharmacokinetics: effects in the sleep studies persisted for hours after plasma clearance took minutes. That gap is an open question, not a settled mechanism. A canine CSF study (Brain Res Bull, 1986) found central penetration tracked plasma concentration, plasma half-life, and lipophilicity — one reason investigators used slow infusions rather than bolus injection.
Reconstitution is the same volumetric exercise as for any lyophilised peptide, covered in reconstitution 101. For a 5 mg vial:
Because the published dosing is weight-normalised, any reconstruction has to carry the per-kg term through, not just the per-vial concentration. Store powder and reconstituted solution per the Certificate of Analysis — short plasma stability says nothing about vial stability.
The dosing literature is thin in every dimension that matters, and the most rigorous designs reached the least favourable conclusions: Monti concluded sleep improvement under DSIP was "of little clinical significance," and Bes concluded short-term treatment of chronic insomnia was unlikely to be of major therapeutic benefit. No maximum tolerated dose has been established, no dose-response curve published, and DSIP has never been approved as a medicine anywhere. Tolerability reports from the era were unremarkable — headaches in a minority of withdrawal patients — but cohorts were small and follow-up short.
For mechanism and the broader picture, see the DSIP research overview and sleep and longevity peptides. Research-grade DSIP ships with a Certificate of Analysis on each lot.
Why is DSIP dosing expressed in nmol/kg instead of mg? Because the human studies came out of the neuropeptide field, where molar dosing is standard. At about 848.9 g/mol, 25 nmol/kg converts to roughly 21.2 µg/kg — about 1.5 mg for a 70 kg subject. Vial calculations have to pass through that conversion first.
What dose did the human DSIP studies use? Essentially all used 25 nmol/kg intravenously — the 1981 healthy-volunteer and insomniac studies, the 1987 four-night polysomnographic study, the 1992 double-blind study, and the early-1980s withdrawal work. No published human dose-ranging study exists.
How long does DSIP stay in circulation? Not long. Measured plasma half-life was 4.0 ± 0.7 minutes in dogs, 2.9 minutes in a monkey, and about 2 minutes in rats, with rapid aminopeptidase-mediated degradation. Reported effects outlasting that window is an unresolved feature of the literature.
Was DSIP dosed nightly or as a single administration? Both appear. The 1981 studies used a single acute intravenous administration; the 1987 and 1992 studies dosed before each of four and three consecutive nights; the withdrawal studies used repeat injections titrated to clinical response. None were long-term protocols.
Does the dosing literature support DSIP as a sleep agent? Weakly at best, and the better-controlled studies were the least supportive. Improvements in sleep efficiency and latency were reported, but both double-blind studies judged the effects to be of little clinical significance. This is an open research question, not an established result.
