The brain is a complex place: makes more than a hundred different peptides, messengers by which different populations of neurones communicate with other populations.
Perhaps the “hottest” of these is oxytocin. Interest in oxytocin has been amplified by hundreds of studies that have given oxytocin by intranasal spray. Such sprays, it is claimed, enhance libido, promote weight loss, cure migraine, and generally make us better people by promoting trust and love, and might help to treat a multitude of behavioral disorders including schizophrenia and autism.
The jury is out. As to whether oxytocin is important in pain, fear, stress, appetite, sexual responses, and reproductive and social behaviours, the consensus is probably yes. But whether intranasal oxytocin has any of the beneficial effects ascribed to it is questionable.
I am a skeptic, and the fact that hundreds of papers have reported beneficial effects of intranasal oxytocin does not impress me. These papers repeat each other’s flaws, and reference each other to build an appearance of overwhelming confirmation, while generally neglecting inconvenient evidence. But John Ioannidis’ paper “Why most published research results are false” (1) did impress me, and it provides a catalogue of errors common in the scientific literature, errors abundantly present in the literature on intranasal oxytocin. (2)
I suspect that, in years to come, scientists may look back in bemusement, at how scientists, having identified a peptide with so many interesting actions on so many diverse pathways in the brain, should imagine that it was a good idea to stuff it up the nose.
The rationale is that oxytocin reaches the brain easily this way. But it does not: almost none of it gets into the brain. Large amounts do reach the lungs, from where it enters the blood, raising blood concentrations to abnormally high levels. Oxytocin is a hormone; it is secreted into the blood from the pituitary gland which holds most of the oxytocin in the body. It can act on the uterus and mammary gland of course, where its actions are well-known, but also on the colon, pancreas, kidneys, heart, bone, various blood vessels, and on the prostate and penis. Certainly intranasal oxytocin has effects on the brain. But how many of these are consequences of its actions on the colon, pancreas, kidneys, heart, bone etc.
Some don’t regard this as an important question; if intranasal oxytocin has beneficial effects, does it matter how those effects come about?
So are the claimed benefits real? In 2015, Hofman et al. published a meta-analysis of placebo-controlled studies on psychiatric patients in which they concluded that oxytocin does have potential benefit. (3)
But, in 2018, they found an error. They had assumed that individuals in the placebo group would never show greater reductions in symptoms than individuals receiving oxytocin, and this led them to miscode many of their effect sizes – they coded all effects as positive when many were negative. Correcting this changed their conclusions: “Whereas our data previously indicated that oxytocin had some potential benefit for reducing depression, anxiety, psychotic symptoms, and general psychopathology, we now conclude that oxytocin has no benefit for any of these symptom domains.” (3)
Confirmation bias is not a stratagem used deliberately by authors to deceive their readers. More often, it’s an unconscious bias – we tend to see what we expect to see. I am a skeptic about intranasal oxytocin; I find errors in studies that report benefits because I expect to see errors, and I look for them. Hofman et al. did not see their error because they did not foresee the possibility that, through the vagaries of chance in sampling, oxytocin might sometimes appear to have significant negative effects. They retracted their paper, and we should applaud them for doing the right thing. Many others leave known errors to fester in the literature, being too embarrassed to admit them.
Daniel Quintana (4), a proponent of intranasal oxytocin who is exceptionally thoughtful and rigorous, has observed that ”researchers should apply the same constructive scepticism for non-significant studies as for significant studies,” and he is right. We have to try to weigh the totality of evidence – but this is hard because we never see the totality of evidence. We see what has been published, and, studies that report positive effects are much more likely to be written up and published than studies that find no significant effect.
I may be wrong about intranasal oxytocin. It is hard to weigh the evidence of benefits with any confidence when that evidence is extensively flawed and incompletely catalogued.
I am a skeptic because the case that intranasal oxytocin actually reaches the brain in adequate amounts seems to me to be hopelessly flimsy.
Recent studies have indicated that very little intranasal oxytocin reaches the brain; and what little does reach it, does so via the blood (5). Most studies on rats that have shown behavioral effects of oxytocin have given it directly into the brain. The lowest doses with a clear effect are those that influence the milk-ejection reflex in lactating rats – the one physiological function for which oxytocin is clearly essential. Injections of I ng into the brain of a rat are effective in this.
When oxytocin is given intravenously to rats, some reaches the brain. But only about 0.002% of the applied dose does so (6). So, to deliver 1 ng into the brain of a rat by intranasal application would require giving a dose of 50 µg.
This is the dose of oxytocin commonly administered to human by intranasal spray, and it’s a big dose in the sense that this is more than the total present in the human pituitary gland.
But whereas a rat has about 12 ml of blood, a human has about 5 litres, and while the rat brain weighs about 2g, a human weighs about 1300 g. So to get a direct effect of intranasal oxytocin on the human brain we would expect to have to give him or her a dose of oxytocin that is at least 100 times higher than we would need for a rat – about 5 mg.
But most of the oxytocin that is given intranasally ends up in the gut or in the blood, and even a dose of 50 ug in a human raises blood levels excessively. Giving enough oxytocin intranasally to ensure adequate entry into the brain seems impossible.
Author: Gareth Leng | Date: February 2020
Notes
1. Ioannidis JP (2005) Why most published research findings are false. PLoS Medicine Aug;2(8):e124
2. Walum H et al. Statistical and Methodological Considerations for the Interpretation of Intranasal Oxytocin Studies. Biological Psychiatry 79:251-7
3. Hofmann SG et al. RETRACTED: Effect of intranasal oxytocin administration on psychiatric symptoms: A meta-analysis of placebo-controlled studies. Psychiatry Research 2015 228:708-14. Retraction in: Psychiatry Research 2018 May; 263:299.
4. Quintana DS (2018) Revisiting non-significant effects of intranasal oxytocin using equivalence testing. Psychoneuroendocrinology 87:127-30
5. Lee MR et al. (2016) Oxytocin by intranasal and intravenous routes reaches the cerebrospinal fluid in rhesus macaques: determination using a novel oxytocinassay. Molecular Psychiatry 2018 23:115-22
6. Mens WB et al. (1983) Penetration of neurohypophyseal hormones from plasma into cerebrospinal fluid (CSF): half-times of disappearance of these neuropeptides from CSF. Brain Research 262:143-9.
7. (A similar conclusion was reached in 1974 by Zaidi and Heller using very different techniques: ZaidiSM, Heller H (1974) Can neurohypophysial hormones cross the blood-cerebrospinal fluid barrier? Journal of Endocrinology 60:195-6.)