Oxytocin: childbirth, bleeding, the "trust hormone" and what the studies showed
Ukraine, Dnepr, st. 25 Sicheslavskaya Brigade (Rybinskaya St.), 119 ‑ 120
Ukraine, Dnepr, st. 25 Sicheslavskaya Brigade (Rybinskaya St.), 119 ‑ 120

Oxytocin: childbirth, bleeding, the "trust hormone" and what the studies showed

Oxytocin: what it does in labor and for bleeding after birth, what studies showed for autism, trust, anxiety and breast milk, the numbers, how it is used and who should see a doctor first.

Oxytocin: childbirth, bleeding, the "trust hormone" and what the studies showed
Injectable oxytocin works in labor and for bleeding after birth, while the nasal spray gave different results in studies of autism, trust and anxiety. The numbers, the label regimen and who should see a doctor first.

The main point

Oxytocin is a hormone the body makes itself; as an injectable drug it is approved in the US (FDA label for Pitocin, revision March 2026) for two tasks: to start and strengthen uterine contractions in labor for medical indications and to stop bleeding after birth. This is where its benefit is best measured. In a Cochrane review of 4162 women (6 studies), oxytocin after birth reduced the share with blood loss of 500 mL or more: relative risk (RR) 0.51, that is, a risk about half that without a uterotonic (low certainty). In a study of 29,645 women, oxytocin served as the reference against which a new drug was compared. The second part of the story concerns the nasal spray: the "trust hormone" and links with autism, anxiety and milk. These uses are not in the label, and the results are different: in 290 children and adolescents with autism the social withdrawal scale changed almost the same in both groups, with no significant difference. Related topics: cortisol and the adrenal glands, the stress axis, sleep and energy and the mind and the thyroid gland.

Substance and formWhat it doesWhat the studies showedWho it is discussed for
Injectable oxytocin, laborcauses uterine contractionsFDA label: induction and stimulation of labor for medical indications; in a Cochrane review, oxytocin vs expectant management: not delivered within 24 hours 8.4% vs 53.8%pregnant women with medical indications (FDA indication)
Injectable oxytocin, after birthcontracts the uterus and reduces blood loss4162 women, 6 studies: blood loss of 500 mL or more, RR 0.51 vs no uterotonicwomen in labor (FDA indication)
Oxytocin nasal spray, autismgiven through the nose290 children and adolescents, 24 weeks: social withdrawal scale changed by −3.7 and −3.5 points in the two groupschildren and adults with autism, in studies (the FDA label has no nasal form)
Oxytocin nasal spray, milk in mothers of preterm babiesgiven before expressing milk51 mothers, 1 study: total milk volume did not differ between groupsmothers of preterm babies, in studies
Oxytocin nasal spray, trustgiven before a trust gamereplication with power above 95%: no effect found in the main condition of the gamehealthy volunteers, in studies

How oxytocin works

According to the FDA label, oxytocin causes uterine contractions by raising the calcium content inside cells. The uterine wall has special receptors, and their number grows sharply during pregnancy, peaking at the start of labor at term. That is why the response to the same dose is very individual and depends on the sensitivity of the uterus. Its plasma half-life is about 1–6 minutes. When given into a vein, the uterus responds almost at once and the effect fades within an hour; when given IM (by injection), the response comes after 3–5 minutes and lasts 2–3 hours. Oxytocin differs from vasopressin by only two amino acids out of eight, so in large doses it has a vasoconstrictor and antidiuretic effect, that is, it makes the body retain water.

Labor: induction and stimulation

The label names the indications: induction of labor for medical reasons (Rh conflict, maternal diabetes, preeclampsia at or near term, premature rupture of membranes), strengthening weak labor, and help with incomplete miscarriage. Elective induction without indications is not covered by the label: there is not enough data to assess the balance of benefit and risk. In a 2009 Cochrane review, oxytocin was compared with expectant management: 8.4% of women had not delivered vaginally within 24 hours vs 53.8%, RR 0.16 (95% confidence interval, CI, 0.10–0.25). Dissatisfaction with induction was also less frequent: 5.9% vs 13.7% (one study). Epidural analgesia was needed more often: RR 1.10 (1.04–1.17). The number of participants for these calculations is not given in the abstract.

Against other methods the result is different. Against vaginal prostaglandins, oxytocin lost on timing: not delivered within 24 hours 70% vs 21% (2 studies). Against intracervical prostaglandins, cesarean section was performed in 19.1% vs 13.7%, RR 1.37 (1.08–1.74). Low-dose oral misoprostol vs intravenous oxytocin: 737 women, 6 randomized controlled trials (RCTs), cesarean section RR 0.67 (0.50–0.90) in favor of misoprostol, moderate certainty; for the share of births within 24 hours the difference appears to be small or absent (466 women, 3 RCTs, low certainty). Stopping the drip in the active phase of labor reduced the combination of uterine tachysystole (contractions that are too frequent) and fetal heart rate changes: RR 0.15 (0.05–0.46), 486 women, 3 RCTs, moderate certainty. For the cesarean rate among those who reached the active phase, the difference is probably small or absent: RR 0.92 (0.65–1.29), 787 women, 4 RCTs.

Comparison of high and low dose for weak labor: a meta-analysis of 21 studies (18 RCTs), 14,834 women. For cesarean section, newborn death, postpartum hemorrhage and instrumental delivery there were no statistically significant differences. Uterine tachysystole was significantly more frequent on the high dose (CI 1.30–1.94). The authors concluded that a low-dose regimen may be safe and effective.

Bleeding after birth

This is the most weighty use of oxytocin. A 2019 Cochrane review (24 studies) compared preventive administration in the third stage of labor. Against no uterotonic or placebo, blood loss of 500 mL or more: RR 0.51 (0.37–0.72), 4162 women, 6 studies; 1000 mL or more: RR 0.59 (0.42–0.83), 4123 women, 5 studies; the certainty of both estimates is low. Additional uterotonics were needed less often: RR 0.54 (0.36–0.80), 3135 women, 4 studies, moderate certainty. For blood transfusions the difference is not significant: RR 0.88 (0.44–1.78), 3081 women, 3 studies.

Against ergot alkaloids, it is unclear whether there is a difference in blood loss of 500 mL or more: RR 0.84 (0.56–1.25), 3082 women, 10 studies, very low certainty. But oxytocin has other advantages: vomiting is less frequent, RR 0.09 (0.05–0.14), 1991 women, 7 studies, moderate certainty. A third stage of labor longer than 30 minutes was more frequent: RR 4.69 (1.63–13.45), 450 women, 2 studies, moderate certainty.

The largest study with oxytocin as the reference: CHAMPION, 29,645 women, 23 centers in 10 countries, double-blind. After vaginal birth, carbetocin 100 µg or oxytocin 10 IU was given by injection (IM). Blood loss of 500 mL or more or use of additional uterotonics: 14.5% on carbetocin and 14.4% on oxytocin, RR 1.01 (0.95–1.06). Blood loss of 1000 mL or more: 1.51% and 1.45%. The study was funded by the maker of carbetocin (Merck Sharp & Dohme). It shows that even with a uterotonic given, about 14% of women lose 500 mL or more or need an additional drug.

Breast milk

In a double-blind RCT, mothers who gave birth before 35 weeks received an oxytocin spray or placebo before each expression of milk until day five: 51 mothers (27 and 24). Total milk volume did not differ between groups. The shape of the production curve differed: at first there was more milk on oxytocin, then the groups converged. The weight and fat content of the milk on day five and the mothers' opinion of expressing also did not differ significantly. Most mothers thought they were receiving the active spray, and the authors explained this by the placebo effect and the extra support with breastfeeding (Fewtrell, 2006). An early 1981 study in first-time mothers who gave birth prematurely found milk volume from day 2 to day 5 to be 3.5 times greater on oxytocin (Ruis, PMID 6788318); the number of participants is not given in the abstract, and a later double-blind study did not reproduce this difference.

Autism: what was measured

The largest study: SOARS-B, 24 weeks, 290 children and adolescents aged 3–17 with autism spectrum disorder (ASD), intranasal oxytocin at a total dose of 48 IU a day or placebo. The analysis included 139 and 138 people. The social withdrawal subscale (scale 0–39, higher is worse) changed by −3.7 in the oxytocin group and by −3.5 in the placebo group: difference −0.2 (95% CI −1.5 to 1.0), P = 0.61. Two other social function scales and the IQ assessment also differed insignificantly, and the frequency and severity of adverse events were similar (Sikich, 2021). The study was funded by a government institute, NICHD.

In children aged 3–12: 87 people (45 and 42), 12 weeks, 16 IU in the morning and evening. For the primary outcome, the social responsiveness scale (SRS-2) as rated by parents, no significant treatment effect over time was shown (P = 0.686). The interaction with age was significant (P = 0.028): in children aged 3–5 the parents' rating improved, but by the clinician's rating and on secondary scales there was no improvement (Guastella, 2023). In adult men with ASD, a crossover dose-finding study: 109 were randomized, 103 completed. A new spray form, TTA-121, was tested. Primary outcome at 6 IU a day vs placebo: P = 0.118 in the full data set, P = 0.012 in the per-protocol set, which kept those who took the drug well (Yamasue, 2022).

Objective assessment of facial expression in adult men at a dose of 48 IU a day (multicenter study, 106 people): by week 2 the neutral facial expression measure had decreased, d = −0.78, by week 4 d = −0.46, by week 6 d = −0.35 (P = 0.10, difference not significant) (Owada, 2019). In pooled data the picture is closer to zero. A meta-analysis of 12 RCTs: of 11 studies on social cognition, 7 reported improvement, but the overall result for this and for repetitive behavior was not significant (Ooi, 2017). A meta-analysis of 10 RCTs in adults on four outcomes: social function, standardized mean difference (SMD) −0.018 (−0.279 to 0.242); anxiety −0.168 (−0.387 to 0.050); repetitive behavior −0.078 (−0.382 to 0.225); autism severity −0.084 (−0.343 to 0.175); all differences were not significant (Kiani, 2023). A network meta-analysis found a single benefit in adults: repetitive behavior, SMD 0.41 (0.16–0.66), 147 people, 6 RCTs, moderate certainty; for social communication in adults oxytocin showed no significant improvement, and in children and adolescents none for any outcome (Siafis, 2022).

Trust and recognizing emotions

The label "trust hormone" grew out of a 2005 experiment: after a nasal spray, participants in an economic game more often entrusted money to a partner (Kosfeld, PMID 15931222; the number of participants is not given in the abstract). A double-blind replication with a preregistered protocol and power above 95% (Declerck, 2020), in the condition of minimal social contact, that is, the one where the original work found an effect, found no influence of oxytocin on trusting behavior. Among the authors of the replication is a co-author of the original work. A review of three lines of data (the spray, plasma oxytocin, variants of the receptor gene) concluded that there is no convincing converging evidence linking trust with oxytocin (Nave, 2015). In five double-blind experiments, oxytocin strengthened preference for one's own group: ethnocentrism (De Dreu, 2011). So the "love hormone" describes only part of the picture.

Recognizing emotions: a review of 33 studies and 10 meta-analyses showed that in healthy volunteers a single dose improves recognition of basic emotions, especially fear, and increases the expression of positive emotions. In people with diagnoses (ASD, schizophrenia, eating disorders, depression and others) there is no significant effect on the interpretation and expression of emotions (Leppanen, 2017).

Anxiety, schizophrenia and sexual function

Social anxiety disorder: 25 people received 24 IU of oxytocin or placebo together with exposure therapy. Self-ratings of their appearance and the quality of their own speech improved over the sessions, but symptom severity, dysfunctional thoughts and life impairment after treatment were similar in both groups (Guastella, 2009). Schizophrenia: a meta-analysis of 9 RCTs, 308 people, mean dose 47.6 IU a day, about 7 weeks; negative symptoms SMD −0.26 (−0.61 to +0.09), P = 0.14 (Sabe, 2021). After excluding one study with high heterogeneity, SMD −0.08 (−0.32 to +0.16).

Sexual function in women: a crossover double-blind RCT, 30 pre- and postmenopausal women, 32 IU intranasally 50 minutes before intimacy, 8 weeks of oxytocin and 8 weeks of placebo. Female Sexual Function Index (FSFI): +26% on oxytocin and +31% on placebo; sexual distress −36% and −45%. There was no significant treatment effect: both groups improved (Muin, 2015).

The blood test and the nasal spray

The test "oxytocin in the blood as a measure of attachment" is not supported by the sources. The review by Leng and Ludwig (2016) notes that many oxytocin measurements were made with a method judged unsound, and that claims that peripheral measurements reflect release in the brain are at best questionable. Nave (2015) reaches the same conclusion: plasma oxytocin data are spoiled by the way it is measured. The same source says about the spray that only a very small part of what is given through the nose reaches the cerebrospinal fluid, while concentrations in peripheral blood exceed physiological ones, with a likely effect on the gastrointestinal tract, heart and reproductive organs.

A methodological analysis by Walum and colleagues (2016): the average statistical power of intranasal oxytocin studies in healthy people is 16%, in clinical groups 12%; the median sample is 49 and 26 people. The authors concluded that most published findings have a high probability of not reflecting a true effect. This may explain the discrepancy: individual studies often report improvement, while the pooled result is close to zero.

How it is used

Only the injectable form is approved. For induction and stimulation of labor, oxytocin is given only into a vein, by drip, through an infusion pump, only in hospital. The standard solution: 10 IU in 1000 mL of 0.9% sodium chloride or lactated Ringer's solution, that is, 10 mIU in 1 mL. The starting rate is 0.5–1 mIU a minute (3–6 mL an hour), increased every 30–60 minutes by 1–2 mIU a minute until the desired contraction pattern; at 5–6 cm of dilation the dose can be reduced in the same steps. A rate of up to 6 mIU a minute gives blood levels like those in spontaneous labor, and more than 9–10 mIU a minute is rarely needed. To control bleeding after birth, 10–40 IU is added to a running drip (no more than 40 IU per 1000 mL) or 10 IU is given by injection (IM) after the placenta is delivered. Patients are under continuous observation by trained staff with a physician available and electronic fetal monitoring. The label has no intranasal form.

Who should see a doctor first

Injectable oxytocin is given by medical staff in hospital, and the decision is made by the obstetrician. According to the label, before delivery it must not be given with a clinically narrow pelvis, unfavorable fetal positions, fetal distress, when delivery is not imminent, an already overactive uterus, invasive cervical cancer, active genital herpes, complete placenta previa, vasa previa, cord presentation or prolapse, and hypersensitivity to the drug. Apart from unusual circumstances, it is not given with a predisposition to uterine rupture, including after cesarean section and other operations on the uterus or cervix, with polyhydramnios, fetal immaturity, an overdistended uterus, a high number of previous births, or a history of uterine sepsis or traumatic birth. The label describes that with an excessive dose or heightened sensitivity, uterine hypertonus, tetanic contractions and uterine rupture are possible; also described are water intoxication (antidiuretic effect), hypertensive episodes, rhythm disturbances, subarachnoid hemorrhage, afibrinogenemia and, in newborns, bradycardia, a low Apgar score, jaundice and seizures. The label lists maternal and fetal deaths associated with parenteral drugs used for induction and stimulation of labor. The doctor should be told about fluid intake during a long infusion and about anesthesia: severe hypertension has been described when it was given 3–4 hours after a vasoconstrictor with caudal anesthesia.

What the studies showed

What was studiedWho and how manyResultSource
Prevention of bleeding after birth: oxytocin vs no uterotonic4162 women, 6 studiesblood loss of 500 mL or more: RR 0.51 (95% CI 0.37–0.72), low certaintySalati Cochrane Database Syst Rev 2019. PMID 31032882
Oxytocin vs ergotamine: length of the third stage of labor450 women, 2 studieslonger than 30 minutes: RR 4.69 (1.63–13.45) on oxytocinSalati Cochrane Database Syst Rev 2019. PMID 31032882
Oxytocin as the control in prevention of bleeding29,645 women, 1 studyblood loss of 500 mL or more or an additional uterotonic: 14.4% on oxytocin, 14.5% on carbetocinWidmer N Engl J Med 2018. PMID 29949473
Induction of labor: oxytocin vs expectant managementreview of randomized trials, number of participants for this calculation not stated in the sourcenot delivered within 24 hours: 8.4% vs 53.8%, RR 0.16 (0.10–0.25)Alfirevic Cochrane Database Syst Rev 2009. PMID 19821304
Oxytocin vs oral misoprostol737 women, 6 studiescesarean section: RR 0.67 (0.50–0.90) in favor of misoprostolKerr Cochrane Database Syst Rev 2021. PMID 34155622
Stopping the drip in the active phase486 women, 3 studiesuterine tachysystole with a change in fetal heart rate: RR 0.15 (0.05–0.46)Boie Cochrane Database Syst Rev 2018. PMID 30125998
High vs low dose for weak labor14,834 women, 21 studiescesarean section, newborn death: no significant differences; tachysystole more frequent on the high dose (CI 1.30–1.94)Moraes J Pers Med 2024. PMID 39063978
Nasal spray, milk in mothers of preterm babies51 mothers (27 and 24), 1 studytotal milk volume did not differFewtrell Arch Dis Child Fetal Neonatal Ed 2006. PMID 16223754
Nasal spray, autism in children aged 3–17139 and 138 people, 1 study, 24 weeks−3.7 vs −3.5; difference −0.2 (−1.5 to 1.0); P = 0.61Sikich N Engl J Med 2021. PMID 34644471
Nasal spray, autism in children aged 3–1287 children (45 and 42), 1 study, 12 weeksprimary outcome: no significant treatment effect over time shown (P = 0.686)Guastella Mol Psychiatry 2023. PMID 36302965
Nasal spray, dose finding in adults with autism109 men, 103 completed, 1 studynew form TTA-121, 6 IU a day vs placebo: P = 0.118Yamasue Brain 2022. PMID 35067719
Autism in adults, four outcomes10 studiessocial function: SMD −0.018 (−0.279 to 0.242); anxiety: SMD −0.168 (−0.387 to 0.050)Kiani CNS Neurol Disord Drug Targets 2023. PMID 35585805
Network meta-analysis, repetitive behavior in adults147 people, 6 studiesSMD 0.41 (0.16–0.66), moderate certaintySiafis Mol Autism 2022. PMID 35246237
Trust game, replication of the original experimentpower above 95%, number of participants not stated in the sourceno effect on trust found in the main condition of the gameDeclerck Nat Hum Behav 2020. PMID 32514040
Recognizing emotions33 studiesno significant effect found in people with diagnosesLeppanen Neurosci Biobehav Rev 2017. PMID 28467893
Schizophrenia, negative symptoms308 people, 9 studiesSMD −0.26 (−0.61 to +0.09); P = 0.14Sabe Int J Neuropsychopharmacol 2021. PMID 33890987
Social anxiety disorder25 people, 1 studysymptom severity after treatment similar in the two groupsGuastella Psychoneuroendocrinology 2009. PMID 19246160
Sexual function in women30 women, 1 crossover studyFSFI: +26% on oxytocin, +31% on placeboMuin Fertil Steril 2015. PMID 26151620

Summary

Injectable oxytocin works where it is approved: in labor and in preventing bleeding after birth. In the Cochrane review the risk of blood loss of 500 mL or more was about half that without a uterotonic, and in a study of almost 30,000 women it served as the reference for a new drug. For the nasal spray the results are different: in 290 children and adolescents the social withdrawal scale changed almost the same in both groups, the replication of the trust experiment at power above 95% found no effect, and the sources call the blood oxytocin test as a measure of attachment doubtful. Decisions about labor and hormones are made together with a doctor. You can discuss your questions at a consultation.

Source: Oxytocin: childbirth, bleeding, the "trust hormone" and what the studies showed — universum.earth. Dr. Pereligyn.

Come back
Request a call back