The Glymphatic System and Sleep
Sleep’s nightly “brain wash” rests on a mouse result that human experiments have not cleanly reproduced. In 2013, Lulu Xie’s team measured cortical interstitial space expanding from roughly 14% of tissue volume during wakefulness to 23% during sleep or ketamine-xylazine anesthesia, an increase above 60%. Amyloid-beta then cleared about twice as fast, but a 2024 mouse study reported the opposite result.
How the proposed pathway works
The brain has no conventional lymphatic vessels inside its tissue. The glymphatic model proposes that cerebrospinal fluid enters beside arteries, exchanges with interstitial fluid between cells, and carries solutes toward exit routes. Aquaporin-4 water channels concentrated in astrocyte endfeet appear to regulate part of this exchange.
Iliff et al. reported this route in mice in 2012. Deleting aquaporin-4 reduced interstitial solute clearance by about 70%, although later work showed that anesthesia, tracer size, injection injury, and measurement location can change the result.
What the experiments measure
| Study | Subjects | Signal | What it cannot prove |
|---|---|---|---|
| Xie et al., 2013 | Mice | Interstitial expansion; faster Aβ clearance | Same magnitude in humans |
| Fultz et al., 2019 | 13 humans | CSF oscillations near 0.05 Hz during NREM sleep | Net waste removal |
| Miao et al., 2024 | Male mice | Lower fluorescent-dye clearance during sleep | Transport of larger native proteins |
| Dagum et al., 2026 | 39 humans | Sleep-linked plasma Aβ and tau changes | Direct pathway visualisation |
The human brain moves cerebrospinal fluid differently during sleep. Fultz’s EEG-fMRI experiment found that neural slow waves preceded changes in blood volume, followed by CSF pulses. Movement, mixing, and clearance are three different measurements. Headlines often collapse them into one.
What’s contested
The central dispute is whether solutes travel through brain tissue mainly by pressure-driven bulk flow or ordinary diffusion, and whether sleep increases their net exit. Miao et al. found diffusion unchanged and clearance reduced during sleep and three forms of anesthesia. Plá et al. challenged that conclusion in 2025, arguing that the experiment measured redistribution between two brain locations rather than removal from the brain.
Human evidence remains indirect. MRI contrast agents enter from cerebrospinal fluid, while amyloid-beta and tau originate inside brain tissue and can also cross the blood-brain barrier or undergo local degradation. No single experiment yet watches a native human metabolite complete the entire route.
Why this has to do with other realms
This is a measurement problem before it is a plumbing problem. In concept information theory, a signal is useful only if the channel preserves what the observer wants to infer. CSF motion is a signal; waste removal is the hidden variable.
It is also concept queueing theory inside a skull. Amyloid concentration is inventory, not throughput: the same concentration can result from faster production, slower clearance, or both. One sleepless night can change every term in that equation.
An open question
What tracer could be produced naturally inside the human brain, followed without surgery, and measured again after it reaches blood? Until that instrument exists, how much of the nightly-cleaning story is physiology, and how much is inference?
Key Sources
- Iliff et al. (2012), “A Paravascular Pathway Facilitates CSF Flow Through the Brain Parenchyma,” Science Translational Medicine - the founding glymphatic experiment.
- Xie et al. (2013), “Sleep Drives Metabolite Clearance from the Adult Brain,” Science - the 60% interstitial-space result.
- Fultz et al. (2019), “Coupled Electrophysiological, Hemodynamic, and Cerebrospinal Fluid Oscillations in Human Sleep,” Science - direct measurement of sleep-linked human CSF waves.
- Miao et al. (2024), “Brain Clearance Is Reduced During Sleep and Anesthesia,” Nature Neuroscience - the opposing mouse result.
- Plá et al. (2025), “A Curious Concept of CNS Clearance,” Nature Neuroscience - the methodological challenge to Miao et al.
- Dagum et al. (2026), “The Glymphatic System Clears Amyloid Beta and Tau from Brain to Plasma in Humans,” Nature Communications - randomized crossover evidence linking sleep to plasma amyloid-beta and tau changes.
Abhishek's take
The plumbing metaphor is useful until it starts pretending that motion proves removal. What holds my attention here is the measurement gap: a 0.05 Hz CSF pulse is visible, but the fate of one native metabolite remains hidden.
Tags: #glymphatic-system #sleep #cerebrospinal-fluid #brain-clearance #neurodegeneration