24 units by nose, and 100 to 500 micrograms under the skin
What was studied
A drip into a vein, or ten units into a muscle after delivery
The gap
Convention runs above the studied dose
How long it lasts
About 1 to 6 minutes in the blood, from the approved label. The uterus responds almost at once to an intravenous dose and settles within an hour, or within 3 to 5 minutes into a muscle, persisting 2 to 3 hours.
What it actually is
Oxytocin is a hormone the body makes, nine amino acids long, and it is one of the few things in this reference that is a real approved drug. The approval is an injection used in childbirth: starting or reinforcing labour, helping the uterus empty, and controlling bleeding after delivery. Its label allows exactly two ways in, a drip into a vein or a shot into a muscle, and it is dosed in international units rather than milligrams. What is sold for bonding, trust, social ease and sleep is a nasal spray mixed to order by a compounding pharmacy. There was once an approved American nasal oxytocin; its record is still in the government database, marked discontinued.
What it is supposed to do
In the body it acts on the uterus and the breast, and in the brain it is involved in social and pair bonding behaviour, which is where the selling comes from. The bonding reputation starts with an infusion directly into a rodent's brain: in female prairie voles, oxytocin released into the brain ventricles produced a preference for the partner, while the same drug given under the skin did nothing. So the question was always how to get it in, and that question is genuinely unsettled. It also does not last, at a blood half life of about 1 to 6 minutes.
What people take it for
Bonding, trust, social anxiety, sexual effect and sleep. The trial record for those uses is mostly autism, and the two largest randomised trials I found both missed their primary endpoint, prespecified in the larger of the two. A 2021 systematic review counted 38 randomised trials between 1990 and 2010 across 1,529 participants, 79 percent male, and 8 percent of them people with developmental or mental health difficulties. What that record does not contain is the person buying a compounded spray for bonding or for sleep.
The dose question
What circulates
24 units by nose, and 100 to 500 micrograms under the skin
What was studied
A drip into a vein, or ten units into a muscle after delivery
The Pitocin label, which permits exactly those two routes
Above the studied dose
The gap here is route rather than indication, and the under the skin figures are the ones to look at twice. Converting at the international standard's 1.68 micrograms per unit, which is my arithmetic, 100 to 500 micrograms a day is roughly 60 to 300 units. A single dose vial of the approved American injection holds 10 units. I could not find a human trial of oxytocin under the skin at those amounts. The nasal convention is a different matter: 24 units is not a calculation but a physical object, six squirts of a commercial spray metered at 4 units each, and I could not find the study that chose it.
Reported because it is what people use. Nothing here recommends any amount.
How long it lasts
About 1 to 6 minutes in the blood, from the approved label. The uterus responds almost at once to an intravenous dose and settles within an hour, or within 3 to 5 minutes into a muscle, persisting 2 to 3 hours.
Route studied
Into a vein or into a muscle on the approved label. In the behavioural literature it is a nasal spray at 24 units, six squirts metered at 4 units each, and the two large autism trials targeted 48 units a day. The best direct measurement of whether the nose reaches the brain gave 24 units to 11 men and saline to 4, one spinal fluid sample each.
Route used
Sprayed into the nose, and secondarily injected under the skin. I could not find a human trial of oxytocin under the skin at the amounts that circulate for that route.
How to check you have the right molecule
The molecule is not the problem here, so check the product and the route instead. What you are buying is compounded rather than approved, which means a pharmacy mixed it to order and no regulator assessed that formulation. One compounded spray I read lists postpartum haemorrhage among its uses, which is a genuine approved indication being offered by a route the approved label does not carry. Nothing establishes that the Swiss research spray used in the studies and the discontinued American nasal application are the same formulation, and this entry does not treat them as such. Units are the other trap: oxytocin is dosed in international units, the standard being about 1.68 micrograms per unit, so any microgram figure has to be converted before it can be compared with anything in the literature.
The premise under the whole spray market is that sniffing it reaches the brain, and that is argued over inside the papers that measured it rather than settled. The founding human paper for the nose to brain route did not test oxytocin at all; it gave a melanocortin fragment, vasopressin and insulin. Oxytocin itself was measured eleven years later in neurology inpatients already having a spinal tap, one sample each, and nothing measurable had happened at 45 or 60 minutes, with only the three men sampled at 75 minutes above the placebo group. Since then macaque work found no advantage of the nose over a vein, a human imaging study explained the brain blood flow change entirely by the rise in the bloodstream, and blocking the nasal blood route removed most of an electrical brain effect. Pushing the other way, fluid sampled straight out of rodent brain regions rose after a nasal dose, and reviews split on the same literature. The honest position is that it is unresolved, and blood oxytocin, which is what nearly every study and every sales page measures, did not track the spinal fluid in the one human study that measured both.
What is known about harm
The serious harms on the approved label belong to labour rather than to the molecule: maternal deaths from blood pressure episodes, bleeding around the brain, rupture of the womb, and fetal deaths, with oxytocic drugs injected for induction. The harm that belongs to the molecule is water intoxication, which has killed, with convulsions and coma during a slow infusion over 24 hours, and the label attaches a rate to it that works out at 2.4 to 3 units an hour sustained into a vein. That is a different kind of exposure from six squirts of a spray. On long term risk the label's own answer is that the studies were not done: none on carcinogenicity or mutagenicity, and no information on fertility. For the nasal route in research settings the record is unremarkable, and a 2011 review says so in three numbered points, at 18 to 40 units for short term use in controlled research. The best powered posted safety numbers are from the largest autism trial, where over 24 weeks serious adverse events ran 2 of 146 on oxytocin against 1 of 144 on placebo.