Sleep Science
The glymphatic system: what the brain does with its waste while you sleep
A drainage network described only in 2012 appears to run far more actively during sleep than wakefulness. It is one of the most exciting findings in the field and one of the most over-claimed.
The brain has no lymphatic vessels of the conventional kind, which raised a long-standing question about how it disposes of metabolic waste. In 2012, Maiken Nedergaard's group at Rochester described a system that appeared to answer it.
Cerebrospinal fluid enters the brain along the outside of arteries, in a space between the vessel wall and the surrounding astrocyte endfeet. It exchanges with interstitial fluid, picking up soluble waste, and exits along veins. Because astrocytes and their aquaporin-4 water channels are essential to the flow, they named it the glymphatic system — glial plus lymphatic.
The sleep finding
The following year, Lulu Xie and colleagues published the result that made this famous. Using two-photon imaging in live mice, they found that glymphatic flow increased dramatically during sleep and under anaesthesia compared with wakefulness — by a factor commonly quoted around ten to twenty.
The proposed mechanism was mechanical: the volume of the interstitial space increased by roughly 60 per cent during sleep, offering far less resistance to fluid movement. And they showed that clearance of injected labelled amyloid-beta — the protein that aggregates in Alzheimer's disease — was substantially faster in sleeping animals.
The narrative that followed was irresistible: sleep is when the brain washes itself, and failing to sleep allows the proteins of dementia to accumulate.
The link between sleep and amyloid in humans has independent support. A study by Ehsan Shokri-Kojori and colleagues found that a single night of sleep deprivation increased amyloid-beta burden measurably on PET imaging in healthy adults. Longitudinal cohort studies associate poor sleep with later amyloid accumulation and with dementia risk. The relationship is also bidirectional — Alzheimer's pathology disrupts sleep, which complicates every causal claim.
The criticisms, which are substantial
This is a field in active dispute, and the popular account has run well ahead of the settled science.
The mechanism is contested. Several modelling studies have argued that diffusion alone cannot account for the transport rates claimed, and that bulk flow through the brain parenchyma faces resistance that makes the proposed convective mechanism physically difficult. Alternative accounts emphasise arterial pulsation-driven flow confined to perivascular spaces rather than through the tissue.
Anaesthesia is a confound. Much of the work uses anaesthetised animals, and different anaesthetics produce different results — which suggests some of the effect attributed to sleep may be an effect of the drug.
A 2024 study challenged the core claim directly. Nick Franks and Bill Wisden's group at Imperial College used a different method — fluorescent dye injected into the brain and measured over time — and reported that clearance was reduced during sleep and anaesthesia relative to wakefulness. That is the opposite of the original finding, and the disagreement has not been resolved.
Human evidence is indirect. Most direct measurements are in rodents. Human work relies on imaging proxies and on the amyloid findings, which are consistent with the theory but do not establish the mechanism.
The related discovery that is not in dispute
In 2015, two independent groups — Jonathan Kipnis's and Antoine Louveau's at Virginia, and Aleksanteri Aspelund's in Helsinki — described genuine lymphatic vessels in the dural meninges surrounding the brain. These drain to deep cervical lymph nodes and carry immune cells.
This was a substantial anatomical finding: a system that had been looked for and declared absent for a century turned out to exist. It is separate from the glymphatic proposal and much better established.
What can be said responsibly
Sleep and the clearance of brain waste are connected. The association between poor sleep and amyloid accumulation is real, replicated, and clinically interesting. The direction of causation is probably circular.
Whether the glymphatic system as originally described is the mechanism — and whether flow really increases tenfold in sleep — is genuinely uncertain and currently contested by competent researchers using different methods.
What should not be said is that sleeping washes plaques out of your brain and that a bad night causes Alzheimer's. That is a compression of a contested rodent finding into a health claim, and it has been repeated so widely that it is now treated as established fact.
The practical upshot
Unchanged, and unexciting. Sleep well because the evidence for its importance across memory, metabolism, immunity, cardiovascular health and mood is strong and independent of this particular mechanism.
If the glymphatic account survives, it will be one more good reason. If it does not, none of the others go away.
