Research Article

Cerebrospinal fluid influx drives acute ischemic tissue swelling

See allHide authors and affiliations

Science  13 Mar 2020:
Vol. 367, Issue 6483, eaax7171
DOI: 10.1126/science.aax7171

You are currently viewing the abstract.

View Full Text

Log in to view the full text

Log in through your institution

Log in through your institution

Spreading edema after stroke

The brain is enveloped in a cushion of cerebrospinal fluid (CSF), which normally provides protection and helps to remove metabolic waste. CSF transport has also recently been shown to play unexpected roles in neurodegeneration and sleep. Mestre et al. used multimodal in vivo imaging in rodents and found that, after a stroke, an abnormally large volume of CSF rushes into the brain, causing swelling (see the Perspective by Moss and Williams). This influx of CSF is caused by constrictions of arteries triggered by a well-known propagating chemical reaction-diffusion wave called spreading depolarization. CSF transport can thus play a role in brain swelling after stroke.

Science, this issue p. eaax7171; see also p. 1195

Structured Abstract


Cerebrospinal fluid (CSF) covers and protects the brain from mechanical injury. CSF also flows along an interconnected network of perivascular spaces surrounding blood vessels and communicates with interstitial fluid permeating brain tissue, aiding in the removal of metabolic waste produced by cells. This glial cell–mediated lymphatic (glymphatic) function of CSF represents a continuous source of fluid and ions for the brain. When a cerebral artery is occluded, nearby brain tissue is abruptly deprived of blood flow, oxygen, and glucose. This process, known as acute ischemic stroke, is a leading cause of morbidity and mortality worldwide. After stroke, fluid accumulates in ischemic tissue, and the brain becomes edematous and begins to swell, a dangerous complication of the disease. In the first hours after occlusion, the degree of swelling correlates with the net gain of cations, primarily sodium, and this gain draws in fluid from surrounding sources.


Because the brain is already encased by CSF, we asked if glymphatic flow could play a role in early edema formation. To test this, we evaluated CSF dynamics using in vivo magnetic resonance (MR) and multimodal optical imaging after occluding the middle cerebral artery in mice. Edema was assessed using diffusion-weighted MR, and edema fluid sources were labeled using radionuclides. Changes in the flow of CSF in perivascular spaces were explored using a network model of the mouse middle cerebral artery. Histology was used to evaluate edema formation in regions adjacent to CSF inflow routes in mouse and human autopsy tissue.


We found that within minutes of ischemic stroke, CSF flowed rapidly into brain tissue along perivascular spaces. Its entry coincided with the onset of swelling and increased brain water content. Radionuclides and multimodal imaging confirmed that CSF was the earliest contributor of both fluid and ions. Calcium imaging in transgenic mice expressing GCaMP7 in cortical neurons and astrocytes revealed that this process was initiated by spreading depolarizations that were triggered when tissue was deprived of blood flow. Diffusion-weighted MR imaging showed that this was the earliest phase of edema formation. This aberrant CSF inflow was found to be caused by spreading ischemia, the pathological constriction of cerebral blood vessels that follows spreading depolarizations. We present a network model that predicts that the space left unoccupied after vessels constrict would be filled by an inrush of CSF that nearly doubles flow speed. That prediction was confirmed experimentally using particle tracking velocimetry of CSF flow in live mice. Inflow depended on the aquaporin-4 water channel that is highly expressed by glial cells (astrocytes), which is a key contributor to glymphatic function. Postmortem examination of rodent and human brains showed increased fluid accumulation in tissue surrounding perivascular spaces and the cerebral ventricles compared with regions deep in the brain that were far from large CSF reservoirs.


Here, we demonstrate that CSF can provide a source of ischemic edema. Glymphatic inflow of CSF appears to be the primary initial event driving tissue swelling. This finding challenges our current understanding of edema formation after stroke and may provide a basis for treatment of acute ischemic stroke. Spreading depolarizations continue several days after stroke and are also present in many other neurological conditions, ranging from traumatic brain injury to migraine; therefore, it will be important to determine if spreading edema is also a feature of these diseases and whether CSF influx contributes to worsening at more delayed time points. It is also intriguing to speculate that abnormal CSF inflow could be a source of edema fluid in other types of chronic cerebrovascular disease, such as small-vessel disease characterized by enlarged perivascular spaces and transient accumulations of fluid in periventricular white matter.

CSF influx is responsible for early tissue swelling after stroke.

(Left) Spreading ischemia accelerates CSF inflow to the region deprived of blood flow. (Top right) Spreading ischemia constricts the cortical vessels, increasing the perivascular space and resulting in CSF inflow. (Bottom right) Histology of postmortem tissue shows fluid accumulation in the ischemic human brain (diffuse white empty space, right) that is not present in a control brain (left).



Stroke affects millions each year. Poststroke brain edema predicts the severity of eventual stroke damage, yet our concept of how edema develops is incomplete and treatment options remain limited. In early stages, fluid accumulation occurs owing to a net gain of ions, widely thought to enter from the vascular compartment. Here, we used magnetic resonance imaging, radiolabeled tracers, and multiphoton imaging in rodents to show instead that cerebrospinal fluid surrounding the brain enters the tissue within minutes of an ischemic insult along perivascular flow channels. This process was initiated by ischemic spreading depolarizations along with subsequent vasoconstriction, which in turn enlarged the perivascular spaces and doubled glymphatic inflow speeds. Thus, our understanding of poststroke edema needs to be revised, and these findings could provide a conceptual basis for development of alternative treatment strategies.

View Full Text

Stay Connected to Science