Does Nasal Oxytocin Enter The Brain?

By Neuroskeptic | December 8, 2013 3:53 am

Oxytocin is hot. There are now hundreds of studies looking at the effect of this hormone on the human brain.


A dose of oxytocin, delivered in the form of a nasal spray, can make people nicer towards the ostracised, reduce marijuana cravings, and ‘enhance brain function’ in autistic children – and much more, if you believe it.

But not everyone does. Some doubt that nasal oxytocin even gets into the brain at all. Oxytocin is a peptide molecule, which means it can’t cross the blood-brain barrier and enter the central nervous system. Animal studies suggest that intranasal oxytocin might nonetheless get to the brain via some other route, perhaps along a nerve. But no-one has directly tested this in humans.

Now a new paper from Bonn, Germany claims to have put the fears to rest: Elevated cerebrospinal fluid and blood concentrations of oxytocin following its intranasal administration in humans.

They took 13 patients who were due to have a lumbar puncture (‘spinal tap’) for medical reasons. This procedure samples the cerebrospinal fluid (CSF) that surrounds the spinal cord and brain; if the oxytocin enters the brain, it would certainly be in the CSF.

The patients took either oxytocin or placebo nasal spray either 45, 60, or 75 minutes before the sample was taken, and oxytocin levels in the CSF were analyzed. Here’s what happened:


In the group of three people who had an oxytocin spray 75 minutes before the spinal tap, oxytocin levels were higher than in the group person who had a placebo 75 minutes earlier. “64%” higher, though an average of three points is almost meaningless.

It’s really hard to know what to make of such tiny numbers. Even pooling all the placebo patients, there were just 4 of them.

The authors say

[Using a one-way ANOVA], mean CSF concentrations of oxytocin significantly differed between the sampling time points and treatment groups (p=0.009). Specifically, post-hoc Bonferroni tests showed that CSF concentrations were elevated following intranasal oxytocin administration after 75 min compared with the placebo group (p=0.024) or with the group of oxytocin-treated subjects sampled at 45 min (p=0.011).

But with this amount of data, statistics don’t really work. There’s no ‘absolute’ minimum sample size for a one-way ANOVA, but the technique was certainly not intended for data with only 1 observation per group.

If we do take these data at face value, they show that nasal oxytocin doesn’t enter CSF until over 60 minutes after administration, which would be bad news given that many nasal oxytocin studies use a time delay of 30 or 45 minutes. But there’s insufficient data here to call those studies into question, just as it’s insufficient to verify 75 minute ones.

Using medical lumbar puncture to validate nasal oxytocin in humans is a great idea. This study is an important one and the authors are to be commended for getting these results (lumbar puncture research is a hassle.) But this is just not enough data.

ResearchBlogging.orgStriepens N, Kendrick KM, Hanking V, Landgraf R, Wüllner U, Maier W, & Hurlemann R (2013). Elevated cerebrospinal fluid and blood concentrations of oxytocin following its intranasal administration in humans. Scientific reports, 3 PMID: 24310737

  • petrossa

    The good stuff. Proper skepticism well put. Does a new hype medication actually enter the brain. Not really that much that we can tell.

    • James

      Nonsense paper with nonsense data. Where was the Oxytocin for the first 75 min? OT has a very short half-life (minutes). Clearly these authors are not pharmacokineticists. Worse, they publish such garbage in journals…

      • petrossa

        I was refering to Mister/Miss Neuro_skeptic being skeptical about this paper. Sometimes she/he is skeptical just to be skeptical.

      • NervousNeuron

        The paper isn’t making tall claims at all, in fact the authors note the small sample size problem. Although it can be used by secondary sources to make tall claims (nasal OXT does cross the BBB for sure in humans). They themselves say “the present findings do not necessarily represent evidence of a direct nose-to-brain penetration of OXT”

        A good point about the plasma half life though. According to a graph found in the full text version, plasma OXT levels start dropping off between 60 and 75 min (where CSF concentration starts to increase) suggesting that is the half life. But clearance of labeled OXT from plasma is around 5 min (highly variable depending on many factors – but whatever, I’ve never seen one that said it’s 75 min).

        The question is, where does all this extra OXT in the plasma come from? Why do levels in plasma drop off but rise in the CFS? A bit hard to interprate because of the small n and variability in the data.

    • James

      Any paper with making tall claims with n=3 is not worth the paper it was written on…

  • Dan Quintana

    Very timely study considering the prior assumption that oxytocin enters the brain is based on a vasopressin lumbar puncture study done 10 years ago. Important to consider that intranasally administered molecules can enter the brain via trigeminal nerve fibres without needing to pass through CSF (fibers from trigeminal ganglion cells lie close to the surface of nasal cavity in the respiratory epithelium). What will really help answer this question is radiolabelling oxytocin and tracking via PET after intranasal administration.

    • Peter

      From sources,oxytocin relieves anxiety by affecting or influencing the amygdala which is located in the temporal lobe…through intranasal administration,the only way the administered oxytocin would get into the brain would be through the olfactory tract which opposingly does not path anywhere close through the temporal lobe where our amygdala is situated.How then is it that intranasal administration affects anxiety?

  • Pingback: Does Nasal Oxytocin Enter The Brain? - Neuroske...()

  • Wiley

    But is it the same oxytocin in CSF that was administered via nasal spray, or is it possible that the elevated CSF oxytocin is just that which was released via an indirect mechanism following the sprayed oxytocin’s contact with the nasal passage? Would need labeled oxytocin to tell.

    • Dan Quintana

      In animals the distribution within brain CSF after intranasal administration is quite uniform (i.e., not concentrated at target sites – brainstem, hypothalamus, olfactory bulb) suggesting that exogenous oxytocin stimulates the release of endogenous oxytocin (possibly amplified by positive feedback) so it’s probably a bit of both. The other (speculative) alternative is afferent input to the hypothalamus via the vagus stimulates release due to the impact of oxytocin on the heart (oxytocin is a cardiovascular peptide too!). Oxytocin can enter the peripheral nervous system quite easily via blood vessels in the nasal cavity and has been shown to increase vagal input to the heart.

      • Neuroskeptic

        Thanks, I didn’t know that. But I suppose from the point of view of nasal oxytocin’s psychoactive effects, all that matters is whether the spray increases CNS oxytocin – the mechanism is secondary.

  • Matthew Klipstein

    One should note that we do know that intranasal administration of molecules too large to traverse the BBB (eg. Insulin, various growth factors, etc.) has been proven effective to reach the brain. See work of W. Frey, J. Fallon, and others.

  • Alan Magid

    The BBB is leaky in the ventral hypothalamus. That’s how neuropeptides such as oxytocin and ADH (vasopressin) get out. It is a possible route of entry from the circulation.

  • GeoffZoref

    I bought nasal Oxytocin not long ago. I I felt absolutely nothing. On the other hand (and please do not judge me for this) when I took it intravenously, I felt it come on extremely strong, but it was so unpleasant that I would never do it again.

    Taking it IV, I felt a rush at first, which quickly started my heart racing, extreme anxiety and a bad headache. The effects wore off within about 3 or 4 nominates, but it was horrible.

    • jhewitt123

      And that’s why I say n=1 observations, not the study, are the tools of discovery

  • Pingback: 2013-12-13 Spike activity « Mind Hacks()

  • Pingback: Morsels For The Mind – 13/12/2013 › Six Incredible Things Before Breakfast()

  • anon

    uhh, what about this paper? seems pretty conclusive:


    The possibility to improve socio-emotional behaviors in humans by intranasal administration of synthetic oxytocin (OXT) attracts increasing attention, but its uptake into the brain has never been demonstrated so far. Here we used simultaneous microdialysis in both the dorsal hippocampus and amygdala of rats and mice in combination with concomitant blood sampling from the jugular vein to study the dynamics of the neuropeptide in brain extracellular fluid and plasma after its nasal administration. OXT was found to be increased in microdialysates from both the hippocampus and amygdala with peak levels occurring 30–60min after nasal administration. Despite a similar temporal profile of OXT concentrations in plasma, peripheral OXT is unlikely to contribute to dialysate OXT as calculated from in vitro recovery data, indicating a central route of transport. Moreover, intraperitoneal administration of synthetic OXT in identical amounts caused rapid peak levels in brain dialysates and plasma during the first 30min after treatment and a subsequent return toward baseline. While the precise route(s) of central transport remain to be elucidated, our data provide the first evidence that nasally applied OXT indeed reaches behaviorally relevant brain areas, and this uptake is paralleled by changes in plasma OXT.

    • Neuroskeptic

      That’s conclusive evidence that it enters the rodent brain from the rodent nose, but it might not apply to humans.

      Our brain is much further from our nose, and our nose is much smaller relative to our brain & body size, for example.

  • Pingback: The Quality of System One | Theories of Mind()

  • Defenestrator

    Proinflammatory cytokines are much larger proteins than the 9 amino acid oxytocin, and yet peripheral proinflammatory cytokine release leads to increases on the brain side of the BBB. They shouldn’t be able to diffuse across like a steroid hormone would.

    Of of the mechanisms by which this can happen is that there appears to be active pumps within circumventricular organs. Another mechanism is that the afferent vagus nerve detects proinflammatory cytokines in the periphery, and signals for them to be produced locally on the other side of the BBB.

    If oxytocin levels aren’t increased for more than an hour, this would argue for the nerve method because it allows time for de novo transcription and translation.

  • NervousNeuron

    I have used oxytocin analogues in rodent research, and the
    question I always have is whether or not oxytocin administration has central effects in humans. Researchers who I’ve questioned say that ‘oxytocin is having behavioural effects in so many other human studies who use the nasal administration, so clearly effects are central and not peripheral’. More convincing is that other routes of administrations do not have a behavioural (or changes in brain activation) effect, although I’m not too sure about this
    one in humans, never looked it up. But there are other viable explanations, such as peripheral effects to cardiovascular system, guts, etc which is detected by the brain, producing a central response. Like adrenaline.

    Anyway, I still think it’s important to understand the
    mechanism of action, and this study is a good start. It is difficult to interoperate the stats (the authors of the paper mentioned it themselves, very difficult to ethically do this study), so it would be better to see individual data points graphed.

    Anyway, I think the important take away point from this
    study is that plasma OT levels do not accurately reflect CSF levels. Looking at the data, it actually appears to be inversely proportional, but can’t say because of the large variability in the plasma administered data. However there is another study by Kagerbauer et al ( which looked at how well CSF and plasma levels correlate, and the answer is not very well. This study also presented graphs with individual data points when comparing mean levels, so you can get an idea of the variance in a bigger sample size at baseline. The OT data is pretty tight so perhaps the low n in the placebo isn’t too bad (Kagerbauer et al has a slightly lower mean for CSF OT levels).

    The most interesting thing is your point that many studies
    use pretreatment times less than 75mins. These do have behavioural/cognitive effects before then, which suggests that what happens in the CFS may not even represent what’s going on at the (brain) receptor. A much more conclusive study would be to tag oxytocin and measure how long it takes to show up in a PET scanner. I assume there is no such labeled ligand yet, so it hasn’t been done.

  • Gideon

    This was just published on the topic. I think it’s more convincing.

    What do you guys think?

  • Pingback: サバンナ速報 » 自閉症、アスペの治療薬が開発? 東大が治験中()

  • Pingback: Sceptical Neuropragmatism | The Gelatinous Blog()

  • Pingback: Neuroskeptic on Media Neurons: Serotonin and Oxytocin |

  • Pingback: The self-justification molecule: how have we accumulated a vast behavioral oxytocin literature for over a decade | The NeuroEconomist()

  • Pingback: How Can We Measure Human Oxytocin Levels? ⋆ Epeak . Independent news and blogs()

  • Pingback: How Can We Measure Human Oxytocin Levels? – Credence Nation()

  • Pingback: How Can We Measure Human Oxytocin Levels? – Neuroskeptic | Newze - Web Magazine()

  • Pingback: How Can We Measure Human Oxytocin Levels? - Neuroskeptic()

  • Pingback: » Spray that cures obesity !()



No brain. No gain.

About Neuroskeptic

Neuroskeptic is a British neuroscientist who takes a skeptical look at his own field, and beyond. His blog offers a look at the latest developments in neuroscience, psychiatry and psychology through a critical lens.


See More

@Neuro_Skeptic on Twitter


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Collapse bottom bar