The Side You Sleep On | nammu.academy
Sleep · Glymphatics · Female Health
nammu.academy
Sleep Position · Brain Clearance

The Side You Sleep On.
One Rodent Study, Two Contradicting Answers, and the Finding Nobody Repeats.

You have read that sleeping on your side flushes toxins from your brain. That claim traces to a single study in anaesthetised rodents. The system it depends on was contradicted in 2024 by a paper finding clearance goes down during sleep, not up. Meanwhile there is one sleep-position finding in humans with an odds ratio of 2.63, and it gets almost no attention at all.

nammu.academy
3 interactive sections
10 peer-reviewed sources
60%Increase in interstitial space measured during sleep in mice
1 studyBehind the whole side-sleeping claim, in anaesthetised rodents
2024Year a paper found clearance reduced during sleep
2.63Odds ratio for late stillbirth with supine going-to-sleep position

This is one of my favourite examples in all of health content, because the true version is stranger and more useful than the version being sold.

The claim goes like this. Your brain has a waste clearance system. It runs mostly during deep sleep. Body position changes how well it works, and lying on your side works best. So sleep on your side to detoxify your brain.

Each sentence in that chain is doing something different, and they degrade in quality as you go. The first is a genuine and exciting discovery. The second is contested, seriously, by a paper from 2024. The third comes from one study in anaesthetised rodents. The fourth is a pillow advertisement.

And running underneath all of it is a sleep-position finding in humans, with a real effect size, that almost nobody puts in a wellness post because it is about pregnancy and stillbirth rather than about detoxing.

So let me take the chain apart link by link, because the interesting part is not whether the advice is right. It is what happens to a finding on its way from a mouse to a mattress.

Four Positions, and What Was Actually Measured

Note how many of the human cells are empty. Tap through.

The positional comparison comes from Lee and colleagues, in anaesthetised rodents. No human study has compared glymphatic clearance across sleep positions. [3]

The first link is strong and it deserves the excitement it got.

2012: a clearance pathway nobody had described. Jeffrey Iliff, Maiken Nedergaard and colleagues described a paravascular route by which cerebrospinal fluid moves through brain tissue and carries interstitial solutes away, including amyloid beta. Because it appeared to depend on glial water channels and to perform a lymphatic-like function in an organ with no lymphatic vessels, they called it glymphatic. [2]

2013: and it runs during sleep. Lulu Xie and colleagues then measured it across sleep and waking in mice and found that the interstitial space expanded by around 60 percent during sleep, with a roughly twofold increase in the rate at which amyloid beta was cleared. This is the paper that produced the phrase about the brain washing itself at night. [1]

2019: the human version, sort of. Nina Fultz and colleagues used simultaneous EEG and fMRI in sleeping humans and found large pulsing waves of cerebrospinal fluid flow in the fourth ventricle, tightly coupled to slow-wave electrical activity and to blood oxygenation changes. They did not measure clearance of anything. They measured fluid moving, in humans, locked to the rhythm of deep sleep. [4]

And the reason anybody cares. Ehsan Shokri-Kojori and colleagues used PET imaging in healthy adults before and after one night of total sleep deprivation and found increased amyloid beta burden in regions implicated in Alzheimer's disease, including the hippocampus and thalamus. [6] One night. Twenty people. Preliminary, in the authors' own word.

That is a real, coherent and genuinely important research programme. Nothing in it is about which side you lie on.

The Contradiction

Two sets of experiments, opposite conclusions, both in Nature-family journals. Switch between them.

This is an unresolved live disagreement in the field, not a settled overturning. Anyone telling you the brain definitely washes itself at night is ahead of the evidence. [1][5][7][8]

Now the specific claim, which is where the chain gets thin.

One study. Hedok Lee, Nedergaard, Helene Benveniste and colleagues compared glymphatic transport in rodents positioned on their side, on their back and face down, using contrast MRI and fluorescence. Transport was most efficient in the lateral position. [3]

What the paper is. A careful mechanistic experiment in anaesthetised animals, published in the Journal of Neuroscience, testing a plausible hypothesis about posture and fluid dynamics. It is a good study.

What it is not. It is not human. It is not in natural sleep. It does not measure any health outcome. And the animals in question are quadrupeds whose head, neck and venous drainage geometry differ from a human lying in a bed. Every one of those gaps has to be crossed before you arrive at advice, and the literature has crossed none of them.

Zero
Human studies comparing brain clearance across sleeping positions. The entire side-sleeping recommendation rests on a rodent posture experiment, extended by analogy.

There is one human study that touches it from a different direction, and it is worth knowing precisely because it is so easy to misread. Daniel Levendowski and colleagues used multi-night in-home recordings to compare sleep position in 45 people with diagnosed neurodegenerative disease against 120 age and sex matched controls with normal cognition. Spending more than two hours a night with the head supine was significantly more frequent in the neurodegenerative group, odds ratio 3.7, and the association survived adjustment for age, sex, snoring and sleep apnoea diagnosis. [9]

Read the design before you react to the number. It is case-control, measured after diagnosis. Neurodegenerative disease changes how people sleep, move and reposition themselves at night. The arrow could easily run the other way, and the authors say their findings warrant further investigation rather than claiming a direction. [9]

A rodent posture study and a case-control association in 45 patients. That is the entire evidential basis for a recommendation currently printed on pillow packaging.

The Evidence Ladder

Six studies, sorted by species and design rather than by conclusion. Filter them.

Notice that the only human study with a hard clinical outcome is the one about pregnancy, and it is not about clearance at all. [1][3][4][8][9][10]

The part that should make you angry

There is a real sleep-position finding in humans. It is about pregnancy, and it barely travels.

Hold two things next to each other.

One: a rodent posture study, extended through several unjustified steps into an instruction to sleep on your side for brain detoxification, which has been repeated across wellness media for a decade and printed on products.

Two: an individual participant data meta-analysis pooling the available case-control studies of maternal going-to-sleep position and late stillbirth, which found an adjusted odds ratio of 2.63, with a confidence interval of 1.72 to 4.04 and a p value below 0.0001. Going to sleep on the left or the right appeared equally safe. [10]

The second one is a real human effect, in a real clinical outcome, in a population that is medically monitored more intensively than almost any other. It produces a piece of advice so simple it fits in a sentence: in late pregnancy, go to sleep on your side rather than on your back, either side.

Now ask how many pregnant women you know were told that clearly, and how many people you know have heard that side-sleeping detoxifies the brain.

I want to be careful about the numbers here, because the individual risk of late stillbirth is low and the point is not to frighten anyone who woke up on her back. The point is about which findings get amplified. An unproven mechanism with a consumer product attached travels effortlessly. A well-evidenced, free, actionable finding about pregnant bodies moves through professional guidelines slowly and through public awareness barely at all.

That is not an accident of science communication. It is the same pattern this whole blog keeps running into, which is that female-specific findings need to be extraordinary to get the attention that a mouse gets for being interesting.

Where the evidence gets thin

Four things to hold lightly

The glymphatic system is real as a described anatomy and contested as a mechanism. Nathan Smith, Alan Verkman and colleagues tested it directly and found solute transport in brain tissue that was diffusive and independent of the water channel the model relies on, rather than the convective flow the model predicts. [7] That was 2017 and the argument has not ended.

The 2024 result is the sharpest challenge. Miao and colleagues measured movement and clearance of fluorescent tracers in mouse brains and reported that movement was independent of sleep, wake or anaesthesia, and that clearance was markedly reduced during sleep and anaesthesia rather than increased. [8] Different tracers and methods can give different answers, and that is precisely why nobody should be selling certainty here.

Almost all of this is in mice. The human work shows fluid moving in time with slow waves [4] and amyloid rising after a night without sleep [6], both of which are consistent with the story and neither of which tests it. No human study has compared positions.

And if you have back pain, hip pain or reflux, the position that lets you sleep is the right position. Sleep duration and continuity have vastly better evidence behind them than sleep geometry, and trading sleep quality for a posture derived from anaesthetised rodents would be a bad exchange. My own back would make that choice for me. [1][8]

01
If you are pregnant, this one is realSide, either side, from late pregnancy

Going to sleep on your side rather than your back in late pregnancy has an actual human evidence base with an adjusted odds ratio of 2.63 for late stillbirth. Left and right looked equally safe, so use whichever you can sustain. If you wake on your back, roll over and go back to sleep; the studies measured going-to-sleep position, not every minute of the night. [10]

02
Otherwise, sleep on whichever side you likeIncluding your back

There is no human evidence that any sleeping position clears more from your brain. If lying on your side is comfortable, do that. If it wrecks your shoulder, do not. Nothing measurable is being lost either way, because nothing measurable has been measured.

03
Protect the deep sleep insteadThat is where the evidence actually is

The human findings that exist are about sleep itself, not posture. Fluid pulses coupled to slow-wave activity [4], and amyloid burden rising after a single deprived night [6]. Duration, continuity and the conditions that allow deep sleep are the variables with evidence attached. Position is not.

04
Learn to spot the species jumpIt is the single most useful habit in reading health claims

When a recommendation arrives with a mechanism attached, ask what species it was measured in, whether the animals were awake, and whether any outcome was measured or only a proxy. That question resolves a surprising proportion of wellness advice in under a minute.

I came in expecting to confirm the side-sleeping advice with a few caveats. Instead I found a chain where each link is weaker than the one before it, and a 2024 paper arguing that the foundation points the opposite way.

None of which makes the underlying science less interesting. A fluid clearance system in the brain, pulsing in time with the slow waves of deep sleep, visible in a human scanner, is a remarkable thing to have discovered in the last fifteen years. It may well turn out to matter enormously for neurodegeneration. It is also unfinished, actively disputed, and nowhere near the stage where it should be generating instructions about pillows.

What stays with me is the asymmetry again. A mechanism in mice, with a product attached, crossed from a journal into general knowledge in about two years. A finding in pregnant women, with a clinical outcome and an odds ratio of 2.63, has been sitting in the literature since 2019 and most women have never heard it.

So: sleep however you can sleep. Unless you are pregnant, in which case sleep on your side, either side, and know that this is the one piece of sleep-position advice that was actually earned.

Findings about female bodies moving slowly while findings about mice move fast is the structural problem The Art of Female Health was written about. The chapter on how evidence travels covers the stillbirth and sleep-position story in more detail. nammu.academy/book

Sleep well. Whichever side. Love, Nina ❤

References

  1. Xie, L., Kang, H., Xu, Q., et al. (2013). Sleep drives metabolite clearance from the adult brain. Science, 342(6156), 373–377. https://doi.org/10.1126/science.1241224
  2. Iliff, J. J., Wang, M., Liao, Y., et al. (2012). A paravascular pathway facilitates CSF flow through the brain parenchyma and the clearance of interstitial solutes, including amyloid beta. Science Translational Medicine, 4(147), 147ra111. https://doi.org/10.1126/scitranslmed.3003748
  3. Lee, H., Xie, L., Yu, M., et al. (2015). The effect of body posture on brain glymphatic transport. Journal of Neuroscience, 35(31), 11034–11044. https://doi.org/10.1523/JNEUROSCI.1625-15.2015
  4. Fultz, N. E., Bonmassar, G., Setsompop, K., et al. (2019). Coupled electrophysiological, hemodynamic, and cerebrospinal fluid oscillations in human sleep. Science, 366(6465), 628–631. https://doi.org/10.1126/science.aax5440
  5. Hablitz, L. M., Vinitsky, H. S., Sun, Q., et al. (2019). Increased glymphatic influx is correlated with high EEG delta power and low heart rate in mice under anesthesia. Science Advances, 5(2), eaav5447. https://doi.org/10.1126/sciadv.aav5447
  6. Shokri-Kojori, E., Wang, G.-J., Wiers, C. E., et al. (2018). Beta-amyloid accumulation in the human brain after one night of sleep deprivation. PNAS, 115(17), 4483–4488. https://doi.org/10.1073/pnas.1721694115
  7. Smith, A. J., Yao, X., Dix, J. A., Jin, B.-J., & Verkman, A. S. (2017). Test of the glymphatic hypothesis demonstrates diffusive and aquaporin-4-independent solute transport in rodent brain parenchyma. eLife, 6, e27679. https://doi.org/10.7554/eLife.27679
  8. Miao, A., Luo, T., Hsieh, B., et al. (2024). Brain clearance is reduced during sleep and anesthesia. Nature Neuroscience, 27(6), 1046–1050. https://doi.org/10.1038/s41593-024-01638-y
  9. Levendowski, D. J., Gamaldo, C., St. Louis, E. K., et al. (2019). Head position during sleep: Potential implications for patients with neurodegenerative disease. Journal of Alzheimer’s Disease, 67(2), 631–638. https://doi.org/10.3233/JAD-180697
  10. Cronin, R. S., Li, M., Thompson, J. M. D., et al. (2019). An individual participant data meta-analysis of maternal going-to-sleep position, interactions with fetal vulnerability, and the risk of late stillbirth. EClinicalMedicine, 10, 49–57. https://doi.org/10.1016/j.eclinm.2019.03.014
nammu.academy  ·  The Side You Sleep On  ·  All sources peer-reviewed & DOI-verified
Next
Next

Learning a new language