Catastrophic Neurologic Disorders in the Emergency Department , 2nd Edition

Chapter 9. Brain Edema

Brain edema at its core, particularly when profound and widespread, displaces brain tissue and results in compression of the diencephalic structures and in impairment of consciousness.

Edema may also occur in one hemisphere and be proportionally more severe than the primary lesion (e.g., brain metastasis). In contrast, brain edema may be rather inconsequential, as in anoxic-ischemic-related brain swelling, and an inevitable result of widespread brain damage. The complexity of diagnosis and management of acute cerebral edema warrant a separate discussion. Outcome in many instances is somber because of the rapid emergence of irreversible damage to the brain stem.

Classification and Presentation of Brain Edema

The structure and workings of the blood-brain barrier have been partly elucidated. A brief discussion is found in Box 9.1.

Brain edema has been conveniently classified, but the types are clinically indistinguishable. For comparative purposes, the different types are summarized in Table 9.1.1 Brain edema has been classified by Klatzo2 into (1) vasogenic edema, which is a consequence of damage to the blood-brain barrier leading to increased capillary permeability, and (2) cytotoxic edema, which is a consequence of a direct cellular insult leading to swelling without abnormalities in capillary permeability. An additional category has been proposed by Fishman,1 who includes interstitial or hydrocephalic edema caused by obstruction of the cerebrospinal fluid (CSF) pathways (Fig. 9.1).

Milhorat3 suggested an alternative classification, categorizing specific compartmental increases that cause enlargement of brain bulk. Enlargement of brain bulk is an increase in brain volume that may take place in one of three major compartments: the vascular compartment (by arterial dilatation or venous obstruction), the astrocytes (by ischemia or intoxication), or the interstitium, including the CSF compartment (by tumors, infections, trauma, or obstructive hydrocephalus).3

Any classification of brain edema suffers from the fact that many acute insults to the brain involve multiple compartments. In addition, the clinical manifestations are a consequence of brain tissue shift, and neither the initial presentation nor the evolution of clinical signs is much different among the different categories. The pathophysiology of brain edema is discussed in Box 9.2.

Diffuse cerebral edema commonly is rapid in onset and results in coma; focal cerebral edema can go largely unnoticed.

Clinical manifestations of brain edema in fulminant hepatic failure are dramatic.16 Often, patients rather suddenly lapse into coma and may exhibit significant extensor responses, sometimes with progression to brain death. When intracranial pressure is monitored, extreme increases with reduction in cerebral perfusion pressure are typical.

Cerebral edema in diabetic ketoacidosis usually is a devastating complication with high mortality, more often, unfortunately, in children and adolescents with juvenile diabetes. The clinical presentation is often without warning: rapidly developing stupor is soon followed by extensor posturing and fixed, dilated pupils. Many patients fulfill the clinical criteria for brain death in a matter of hours.17,19

Box 9.1. The Blood-Brain Barrier

The exchange of fluids and solute between blood and brain in both directions is governed by multiple mechanisms. Exchange is determined by an anatomical restriction (in the true sense, a barrier), transport systems to rapidly provide the main energy source (e.g., facilitated glucose transport system), and osmolality. Breakdown of the blood—brain barrier, therefore, may not be an anatomical defect but may involve any of these control systems.

The blood-brain barrier is located at the capillary level, and its morphology is unique. A characteristic feature is the crowding of the capillary with astrocyte processes. These astrocyte feet initially appeared to define the barrier, but electron microscopic studies clearly demonstrated an open intercellular space. The physiologic function of these astrocyte foot processes is not entirely known, and their influence on the barrier may be only to moderate permeability rather than to define it anatomically. The capillary consists of a single layer of endothelial cells with a well-organized basement membrane. The endothelial cells are continuous, connected with tight junctions, and without true gaps. Structures that appear to be gaps are in fact very thin layers suggesting an opening, but the membranes are intact, securing impermeability.

Focal cerebral hemisphere edema is manifested by one of the herniation syndromes (see Chapter 8) or, in less severe cases, only with more obvious drowsiness. Notably, edema surrounding neoplasms can be clinically silent. In general, degree of impaired consciousness and degree of focal edema may be very poorly related.

Cerebellar softening and swelling may quickly compress the upper brain stem, leading to sudden pontomedullary dysfunction and respiratory arrest. However, the critical threshold of tolerable swelling is not known, and we have observed dramatic swelling with spontaneous resolution.20

Different types of edema can be visualized by computed tomography (CT) scanning or magnetic resonance imaging (MRI). CT is not very sensitive to global cerebral edema in early stages, but the severity of edema can be graded by a simple system that characterizes different areas of involvement (Table 9.2). Most difficult in the evaluation of edema is the absence of sulci. This often becomes an issue in young persons who have a catastrophic illness that may produce edema but that may be difficult to appreciate because of an age-related lack of sulci (Fig. 9.2).

Vasogenic edema produces increased signal intensity on T2-weighted images, particularly in the white matter. Cytotoxic edema in stroke increases the T2 signal as well but often is found in the boundary zone of the infarct between the central area of infarction and the surrounding normal brain tissue. Diffusion-weighted MRI specifically measures free water diffusion. The apparent diffusion coefficient (ADC value) can be calculated. Elevated ADC values indicate vasogenic edema (Fig. 9.3). Restricted diffusion with reduced ADC values indicate vasogenic edema due to neuronal injury, be it trauma or ischemia. Normal ADC values may indicate intravoxel averaging of ADC values when both mechanisms are present.21,22

Table 9.1. Types of Brain Edema





Pathophysiologic mechanism

Proteinaceous plasma filtrate in extracellular space

Cellular swelling from influx of water and sodium

Cerebrospinal fluid migration from increased ventricular pressure


Preferentially white matter (often sparing gray matter)

Preferentially gray matter (often adjacent white matter)

Preferentially periventricular white matter


·  Primary or metastatic brain tumor

·  Inflammation

·  Head injury

·  Cerebrovascular disorders

·  Global anoxic-ischemic insult

·  Fulminant hepatic failure

·  Water intoxication, dysequilibrium syndrome

Obstructive hydrocephalus

Capillary permeability




Source: Modified from Fishman.1 By permission of WB Saunders Company.

Figure 9.1 Changes in brain tissue with cytotoxic edema, vasogenic edema, and interstitial edema.

Specific Clinical Circumstances of Brain Edema

Brain Edema in Postanoxic-Ischemic Encephalopathy

Brain edema usually affects both hemispheres and invariably is present in comatose patients. Brain edema in anoxic-ischemic injury is a global astroglial swelling involving the entire hemisphere. Tissue necrosis results in the development of brain edema several days later, and its appearance is not associated with a significant increase in intracranial pressure. Therefore, brain edema in postanoxic-ischemic encephalopathy, whether from cardiac arrest or asphyxia, is a measure of the severity of the insult and implies a poor prognosis.23

Brain edema in survivors of cardiopulmonary resuscitation is noted only when CT scanning is performed several days after the event. Brain edema on CT scans is more common in patients with absent motor responses or extensor posturing responses, abnormal cranial nerve reflexes, and generalized myoclonus. These patients are often in cardiogenic shock and need progressively higher doses of vasopressors.

Brain edema in postanoxic-ischemic encephalopathy rarely results in further clinical deterioration in these already ravaged patients. In some patients, however, further progression to brain death may occur and is accompanied by further worsening of brain edema revealed by CT scans. CT scanning often shows loss of definition of cortical sulci, lack of gray and white matter differentiation, and, in the most severe cases, obliteration of the basal cisterns (Fig. 9.4).24 In some patients, the initial CT scans may show hypodensity of the basal ganglia (caudate and putamen), which are fields of terminal vascular supply, or of the thalamus. MRI may be useful in delineating the extent of abnormalities, but experience is limited.22

Brain Edema from Acute Metabolic Derangements and Organ Failure

The water content of the brain may increase significantly as a result of changes in plasma osmolality. Brain edema may also occur as a result of an increased osmotic effect of a toxic intermediate, as in acute liver failure.25,26 Fulminant hepatic failure has been associated with the development of brain edema, and its emergence has been linked to early mortality.27

Experiments in hepatectomized rats found that an increase in cortical glutamine may act as an osmotically active molecule (osmolyte), increasing brain water and resulting in astrocyte swelling. The increase in astrocyte content of glutamine in hepatic failure correlates with an increase in arterial ammonia. An increase in glutamine may therefore reflect an attempt by astrocytes to detoxify themselves from ammonia by producing glutamine, creating an osmolyte. The development of brain edema may be further amplified by changes in cerebral blood flow. Increased arterial ammonia may induce vasodilatation, and a relative increase in cerebral blood flow despite the decreased metabolic demand from encephalopathy (so-called luxury perfusion) may increase the development of vasogenic edema.28,29 Other potential mechanisms are inhibition of Na+,K+-adenosine triphosphatase, which may result in astrocyte swelling. This explanation is supported by one study in which serum from patients with fulminant hepatic failure inhibited this pump.30

Box 9.2. Brain Edema: Physiology and Pathology

In vasogenic edema, the breakdown of the harrier results in fluid accumulation into the white matter. Myelin sheets are swollen and filled with vacuoles, which may further result in myelin breakdown, and cysts appear in the white matter. The astrocytes are swollen at a later stage. The breakdown of the blood-brain barrier is most illustrative in vasogenic edema. Whatever disorder triggers the insult, the result is transudation of plasma into the extracellular white matter space. With this flooding of the white matter, however, cerebral blood flow remains unaffected, and cellular mechanisms remain intact.4

In cytotoxic or cellular edema, a preferential astrocyte swelling (gemistocyte) is observed, often maximal in the astrocyte foot processes. Because cytotoxic edema represents intracellular swelling, gray matter is more involved than white matter.

The mechanisms of cytotoxic edema are more complex than opening of the blood-brain barrier. Experimental studies have indicated that compounds blocking the release of excitatory amino acids reduce the water content of the brain, an indirect suggestion that glutamate has a potential role. Free radicals, prostaglandins, arachidonic acid, and possibly leukotrienes may potentiate cerebral swelling,5,6,7 Other evidence indicates that initial cellular acidity could activate ion antiport channels, such as Na+/H- and Cl-/HCO3-, to extrude H+ but at the expense of an increase in osmolarity.

Clearing of brain edema occurs predominantly through the CSF, Clearance of extravasated proteins by the glial cells is also closely linked to resolution of edema fluid; this suggests a major role for colloid osmotic pressure generated by the proteins.8,9 Edema spreads through bulk flow and a downhill pressure gradient between the white matter and the CSF compartment, a mechanism that may be further facilitated when CSF pressure is reduced.10 A centrally located atrial natriuretic factor has been found to moderate the brain water content, and it might decrease edema formation,11,12,13,14 Another mechanism may be due to water channel proteins (aquaporins), widely expressed in die brain. Up-regulation of aqua-porin-4 has been found in contusional brain lesions.15

Table 9.2. Calculation of Brain Edema Severity Score on the Basis of CT Findings in Patients with Fulminant Hepatic Failure



Visibility of cortical sulci

   3 CT scan slices of upper cerebral area (L/R)


Visibility of white matter

   Internal capsule (L/R)


   Centrum semiovale (L/R)


   Vertex (L/R)


Visibility of basal cisterns

   Sylvian fissure (horizontal-vertical, L/R)


   Frontal interhemispheric fissure


   Quadrigeminal cistern


   Paired suprasellar cisterns (L/R)


   Ambient cistern (L/R)


   Maximal total*


CT, computed tomographic; L/R, left and right cerebral hemispheres.
*In CT scan with normal findings.
Source: Wijdicks et al.
16 By permission of Mayo foundation for Medical Education and Research.

CT scans may demonstrate disappearance of the sylvian fissures, and later, complete compression of the basal cisterns and loss of white-gray matter differentiation may be seen (Fig. 9.5). These abnormalities may reverse entirely with control of intracranial pressure and after liver transplantation.24

Brain edema is rarely present in patients with fluctuating drowsiness; it is usually not visualized on CT until patients become stuporous. A linear relationship between the severity of cerebral edema and the degree of hepatic encephalopathy has been found16 and implies that brain edema is the final common pathway by which coma occurs.

Figure 9.2 Resolving brain edema in encephalitis. Virtual lack of sulci and sylvian fissures, poor white-gray matter differentiation, and effacement of basal cistern and third ventricle with improvement (arrows).

Cerebral edema may develop in acute, often dramatic, sodium and glucose derangements. Hyperosmolality can be brought on by severe dehydration or from infusion of hypertonic solutions. This may lead to shrinkage of the endothelial cells, causing gaps and possible rupture of the interendothelial connection that result in increased permeability. Its effect is brief and reversed in a matter of hours.31

Acutely induced hypo-osmolality or hyponatremia of a sufficient degree (usually 110 mmol/L or less) may induce significant cerebral edema. Cerebral edema in this hypo-osmolar state has been noted in young healthy women after general anesthesia and in patients with polydipsia, predominantly in schizophrenia. Excessive administration of free water results in acute onset of massive brain edema,32 rapid displacement of the diencephalon, and respiratory arrest.

Figure 9.3 Series of magnetic resonance sequences (fluid attenuation inversion recovery [FLAIR], diffusion-weighted imaging [DWI], perfusion-weighted imaging [PWI]) showing increased apparent diffusion coefficient values in hypertensive encephalopathy indicative of edema (arrows).

Nonetheless, brain edema is uncommon after acute severe hyponatremia and may be explained by a corrective mechanism due to rapid loss of organic osmolytes. Loss of osmolytes permits transport of potassium outside the cell and leads to reduction in the content of intracellular solute, minimizing the risk of cell swelling induced by this rapid osmotic change. A linear correlation between the degree of hyponatremia and the loss of important osmolytes, such as taurine, glutamate, and aspartame, by the brain has been documented.33 Reduction in the number of osmolytes was already detectable 3 hours after the onset of hyponatremia, decreasing to a minimal concentration in 24 hours.

Figure 9.4 Anoxic-ischemic encephalopathy with generalized brain edema. Arrowheads indicate loss of gray–white matter definition. Arrows point to absence of sulci.

Figure 9.5 Cerebral edema in fulminant hepatic failure. Left: Note pseudo-subarachnoid hemorrhage, which occurs when the brain tissue becomes very low in attenuation and the dura and blood vessels appear comparatively hyperdense (arrowheads point to the basal cisterns and arrows to the tentorium region). Right: Loss of sylvian fissures and loss of cortical sulci.

The mechanism of brain edema in diabetic ketoacidosis may be related to rapid fluid management to correct the ketoacidosis and dehydration, but this is not clear. At least theoretically, rapid administration of fluid may lead to intracellular shift of water because of osmotically active molecules that are intrinsically present to protect the brain from excessive shrinkage. Rapid administration of fluid may override the washout of osmolytes when plasma osmolality rapidly corrects itself. Alternatively, insulin administration may activate the Na+/H+ pump, enhancing sodium and water influx.34 Cerebral edema may be more prevalent than appreciated, and serial CT scan studies have shown the development of clinically unrecognizable cerebral edema.35,36 In one study, increased serum urea nitrogen concentrations, more severe hypocapnia, and treatment with bicarbonate emerged as important factors, even more important than rate of fluid, sodium, and insulin infusion. The vasoconstrictive effect of hypocapnia and extreme dehydration could cause anoxic-ischemic injury, leading to cerebral edema.37 In clinical practice, brain edema may also be related to anoxic-ischemic damage from acidosis-related cardiopulmonary arrest.

Treatment with osmotic diuretic agents is often unsuccessful, evidence that massive cerebral edema has resulted in irreversible herniation.

Brain Edema and Acute Bacterial Meningitis

Brain edema in acute bacterial meningitis may complicate the clinical course and almost certainly increases mortality.

The true prevalence of brain edema in bacterial meningitis is not known, and reported series have been biased toward autopsy material. Brain edema occurs early in the course of bacterial meningitis but is not invariably the cause of early death. In a series of 29 patients with bacterial meningitis, 15 had pathologic evidence of cerebral edema by the appearance of tonsillar herniation at autopsy.38

There may be several mechanisms of brain edema in inflammatory disorders. It has been proposed that granulocyte (“granulocytic brain edema”) products may induce edema.39 Cytokines originating from leukocytes may cause endothelial alterations that can lead to vasogenic edema.39,40 In addition, it has been documented that high doses of these chemotactically active agents administered intrathecally increased brain water content, but it is not clear whether this experiment, which used very high doses, reflected the changes in vivo.41,42 The cytokine–endothelium–leukocyte interaction is currently an active field of research, and recent studies have shown that both dexamethasone and monoclonal antibodies against leukocyte adhesion receptors attenuate meningeal inflammation and brain edema in rats inoculated with Haemophilus influenzae.43

CT scanning of brain edema in acute bacterial meningitis typically shows generalized edema, with edematous white matter and cortical effacement. The ventricles may become extremely small. In contrast-enhanced CT scans, enhancement of the basal cisterns due to hypervascularity or gyral configurations in cortical zones may represent extensive meningitis. MRI may document pus (see Chapter 16).

Brain Edema Associated with Hemispheric Mass

Brain edema often only surrounds a hemispheric mass irrespective of whether it is an intra-parenchymal hematoma, a large territorial infarct, or a tumor.

Intracranial hematomas, when located in the basal ganglia, often have a perihematoma rim of edema that invariably represents vasogenic edema. Its significance is unknown, and an increase in the volume of edema is common within the first 24 hours after the ictus.44 Secondary clinical deterioration from progressive perihematoma swelling commonly occurs in lobar hematomas. The products of degraded erythrocytes are responsible for edema. Hemoglobin breakdown products take time to form, which explains delayed edema in most cases. Hemin, bilirubin, and FeCl2 produce brain edema and leave the way open to explore the use of iron-chelating agents in patients with intracerebral hematomas.45,46,47

Brain edema from a cerebral arteriovenous malformation (AVM) most commonly is associated with recent hemorrhage. Brain edema may also be correlated with increased pressure on the draining veins and marked dilatation or varices of the draining veins. Acute venous occlusion may be another mechanism of venous outflow.48 However, brain edema can be explained by seizures in patients with documented AVM and is most often seen on MRI. The focal hyperintensity on T2-weighted images in the white matter may disappear after control of seizures.

Large hemispheric infarcts may swell, usually after 3–5 days.49 In many patients, brain swelling is heralded by increasing headache and a fluctuating level of consciousness. Outcome in patients with brain stem involvement from herniation is invariably poor (Fig. 9.6). Tumor-associated edema most likely involves multiple mechanisms (Box 9.3, Fig. 9.7). A preliminary study suggested that reduced brain tissue oxygenation improves with decompression in peritumoral brain edema.56

Brain Edema Associated with Head Injury

Diffuse brain swelling is more common in children than in adults. Brain swelling is present commonly in comatose patients with closed head injury and rarely in patients who remain alert. Subsequent deterioration from diffuse brain swelling in a previously alert patient is uncommon (<5%). It has been reported after repeated brain injury—the second impact syndrome—but its mechanism and risk factors are not known.57 Brain swelling occurs more often in patients who suffered a systemic insult, such as hypotension or pre-hospital hypoxemia. Brain swelling often is evident on CT scans and may involve absence of the third ventricle and basal cisterns in most cases. Intracranial pressure is almost always significantly increased.58 Brain swelling in adults with closed head injury is often associated with multiple parenchymal contusions, shearing lesions, and traumatic subarachnoid hemorrhage.58,59,60,61,62

Management of Brain Edema

Acute brain edema is associated with high morbidity and mortality, and results of aggressive intervention have not been encouraging. In the course of several hours, brain shift may be extensive, resulting in permanent damage even after the water content of the brain has been ameliorated.

Figure 9.6 Cerebral swelling associated with middle cerebral artery territory infarct. Top row: Initial scan with hypodensity and loss of sulci in the middle cerebral artery territory (arrows). Bottom row: Massive edema and shift with contralateral hydrocephalus and enlargement of temporal horn (arrows).

Osmotic diuretics, such as mannitol or hypertonic saline, may be useful initially, while the true value of intracranial pressure is determined by placement of a monitor. Continuous infusion with 3% hypertonic saline should be considered if a bolus of mannitol (1 g/kg) is unsuccessful in head trauma, intracranial hematoma, or cerebral infarct. The dose is increased until serum sodium concentrations are 145–155 mmol/L.63 A pilot study suggested an effect from intravenous bolus administration of 23.4% saline in refractory increased intracranial pressure. Cerebral perfusion pressure was augmented as well, and intravascular volume was not depleted.64

Treatment in diabetic ketoacidosis would be normal saline 10 mL/kg in the first hour, followed by 3.5–5 mL/kg hourly infusion. Insulin could be added at 0.1 unit/kg hourly. However, with cerebral edema, decreased partial pressure of arterial CO2 (PaCO2) should be corrected to normal values, intravenous fluid and insulin infusion should be markedly reduced, and mannitol 1–2 g/kg should be added.65 Corticosteroids are useful only in metastasis or glioma with mass effect from perilesional edema. Corticosteroids (10 mg of dexamethasone intravenously) should be considered in brain edema from fulminant bacterial meningitis (see Chapter 16). The major beneficial effect is improvement of CSF dynamics, predominantly the CSF outflow tract over the convexity. Corticosteroids have no documented value in brain edema from endocrine or hepatic disturbances.

Box 9.3. Edema in Brain Tumors

Four processes have been suggested that may be related to edema associated with primary brain tumors or metastasis:50 (1) tumor angiogenesis of vessels with defective blood–brain harrier, characterized by large interendothelial gaps;51 (2) Increased microvascular permeability from production of mediators, such as prostaglandin E2 and thromboxane B2, in this process;52 (3) an immunologic mechanism such as interleukin-2, which when injected has resulted in brain edema in experimental studies;53 and (4) less likely, an inflammatory mechanism through substances, such as platelet-activating factor, released from polynuclear leukocytes surrounding the tumor-associated edema. Other factors that may potentiate tumor-associated edema are seizures, chemotherapeutic agents, and therapeutic radiation.54 Plasma osmolality may play an important role, and one experimental study found a direct relation between plasma osmolality and formation of brain edema.55

Figure 9.7 Peritumoral edema (arrows) in mass lesion (glioma).

Focal hemispheric edema is more difficult to manage, and craniectomy with duraplasty greatly increases the possibility for swelling outside the skull, relieving pressure. Removal of additional swollen brain tissue (anterior temporal lobectomy or frontal lobectomy) is optional but considered only if the primary lesion is in this location (e.g., temporal lobe swelling from herpes simplex encephalitis,66 metastasis with malignant edema).

Surgical decompression in patients with head injury but retained brain stem reflexes has been advocated but only in young patients (<40 years), with reimplantation of bone flaps as early as 6 weeks after surgery. In one experience, results were “surprisingly good.”67 The craniotomy is extensive (frontotemporoparietal), leaving a bone rim on top of the superior sagittal sinus.


1. Fishman EA: Cerebrospinal Fluid in Diseases of the Nervous System, 2nd ed. Philadelphia: WB Saunders, 1992.

2. Klatzo I: Presidential address. Neuropathological aspects of brain edema. J Neuropathol Exp Neurol 26:1, 1967.

3. Milhorat TH. Cerebrospinal Fluid and the Brain Edemas. New York: Neuroscience Society of New York, 1987.

4. Sutton LN, Barranco D, Greenberg J, et al.: Cerebral blood flow and glucose metabolism in experimental brain edema. J Neurosurg 71:868, 1989.

5. Ikeda Y, Long DM: The molecular basis of brain injury and brain edema: the role of oxygen free radicals. Neurosurgery 27:1, 1990.

6. Ito U, Baethmann A, Hossmann KA, et al.: Brain edema IX. Acta Neurochir 60:1, 1993.

7. Kimelberg HK: Current concepts of brain edema. Review of laboratory investigations. J Neurosurg 83:1051, 1995.

8. Marmarou A, Hochwald G, Nakamura T, et al.: Brain edema resolution by CSF pathways and brain vasculature in cats. Am J Physiol 267:H514, 1994.

9. Marmarou A, Nakamura T, Tanaka K: The kinetics of fluid movement through brain tissue. Semin Neurol 4:439, 1984.

10. Reulen HJ, Graham R, Spatz M, et al.: Role of pressure gradients and bulk flow in dynamics of vasogenic brain edema. J Neurosurg 46:24, 1977.

11. Doczi T, Joó F, Szerdahelyi P, et al.: Regulation of brain water and electrolyte contents: the possible involvement of central atrial natriuretic factor. Neurosurgery 21:454, 1987.

12. Nakao N, Itakura T, Yokote H, et al.: Effect of atrial natriuretic peptide on ischemic brain edema: changes in brain water and electrolytes. Neurosurgery 27:39, 1990.

13. Rosenberg GA, Estrada EY: Atrial natriuretic peptide blocks hemorrhagic brain edema after 4-hour delay in rats. Stroke 26:874, 1995.

14. Vajda Z, Pedersen M, Doczi T, et al.: Effects of centrally administered arginine vasopressin and atrial natriuretic peptide on the development of brain edema in hyponatremic rats. Neurosurgery 49:697, 2001.

15. Sun MC, Honey CR, Berk C, et al.: Regulation of aquaporin-4 in a traumatic brain injury model in rats. J Neurosurg 98:565, 2003.

16. Wijdicks EFM, Plevak DJ, Rakela J, et al.: Clinical and radiologic features of cerebral edema in fulminant hepatic failure. Mayo Clin Proc 70:119, 1995.

17. Clements RS Jr, Prockop LD, Winegrad AI: Acute cerebral edema during treatment of hyperglycaemia. An experimental model. Lancet 2:384, 1968.

18. Fitzgerald MG, O'Sullivan DJ, Malins JM: Fatal diabetic ketosis. BMJ 2:247, 1961.

19. Young E, Bradley RF: Cerebral edema with irreversible coma in severe diabetic ketoacidosis. N Engl J Med 276:665, 1967.

20. Wijdicks EFM, Maus TP, Piepgras DG: Cerebellar swelling and massive brain stem distortion: spontaneous resolution documented by MRI. J Neurol Neurosurg Psychiatry 65:400, 1998.

21. Albers GW: Diffusion-weighted MRI for evaluation of acute stroke. Neurology 51:S47, 1998.

22. Wijdicks EFM, Campeau NG, Miller GM: MR imaging in comatose survivors of cardiac resuscitation. AJNR Am J Neuroradiol 22:1561, 2001.

23. Lupton BA, Hill A, Roland EH, et al.: Brain swelling in the asphyxiated term newborn: pathogenesis and outcome. Pediatrics 82:139, 1988.

24. Wijdicks EFM, Parisi JE, Sharbrough FW: Prognostic value of myoclonus status in comatose survivors of cardiac arrest. Ann Neurol 35:239, 1994.

25. Hilgier W, Olson JE: Brain ion and amino acid contents during edema development in hepatic encephalopathy. J Neurochem 62:197, 1994.

26. Olafsson S, Gottstein J, Blei AT: Brain edema and intracranial hypertension in rats after total hepatectomy. Gastroenterology 108:1097, 1995.

27. Blei AT, Olafsson S, Therrien G, et al.: Ammonia-induced brain edema and intracranial hypertension in rats after portacaval anastomosis. Hepatology 19:1437, 1994.

28. Butterworth RF: Hepatic encephalopathy and brain edema in acute hepatic failure: does glutamate play a role? Hepatology 25:1032, 1997.

29. Durham S, Yonas H, Aggarwal S, et al.: Regional cerebral blood flow and CO2 reactivity in fulminant hepatic failure. J Cereb Blood Flow Metab 15:329, 1995.

30. Seda HW, Hughes RD, Gove CD, et al.: Inhibition of rat brain Na+,K+-ATPase activity by serum from patients with fulminant hepatic failure. Hepatology 4:74, 1984.

31. Cserr HF, DePasquale M, Patlak CS: Volume regulatory influx of electrolytes from plasma to brain during acute hyperosmolality. Am J Physiol 253:F530, 1987.

32. Arieff AI: Hyponatremia, convulsions, respiratory arrest, and permanent brain damage after elective surgery in healthy women. N Engl J Med 314:1529, 1986.

33. Sterns RH, Baer J, Ebersold S, et al.: Organic osmolytes in acute hyponatremia. Am J Physiol 264:F833, 1993.

34. Van der Meulen JA, Klip A, Grinstein S: Possible mechanism for cerebral oedema in diabetic ketoacidosis. Lancet 2:306, 1987.

35. Hoffman WH, Steinhart CM, el Gammal T, et al.: Cranial CT in children and adolescents with diabetic ketoacidosis. AJNR Am J Neuroradiol 9:733, 1988.

36. Krane EJ, Rockoff MA, Wallman JK, et al.: Subclinical brain swelling in children during treatment of diabetic ketoacidosis. N Engl J Med 312:1147, 1985.

37. Glaser N, Barnett P, McCaslin I, et al.: Risk factors for cerebral edema in children with diabetic ketoacidosis. The Pediatric Emergency Medicine Collaborative Research Committee of the American Academy of Pediatrics. N Engl J Med 344:264, 2001.

38. Dodge PR, Swartz MN: Bacterial meningitis—a review of selected aspects. II. Special neurologic problems, post-meningitic complications and clinicopathological correlations. N Engl J Med 272:954, 1965.

39. Fishman RA, Sligar K, Hake RB: Effects of leukocytes on brain metabolism in granulocytic brain edema. Ann Neurol 2:89, 1977.

40. Montesano R, Orci L, Vassalli P: Human endothelial cell cultures: phenotypic modulation by leukocyte interleukins. J Cell Physiol 122:424, 1985.

41. Niemöller UM, Täuber MG: Brain edema and increased intracranial pressure in the pathophysiology of bacterial meningitis. Eur J Clin Microbiol Infect Dis 8:109, 1989.

42. Täuber MG, Borschberg U, Sande MA: Influence of granulocytes on brain edema, intracranial pressure, and cerebrospinal fluid concentrations of lactate and protein in experimental meningitis. J Infect Dis 157:456, 1988.

43. Sáez-Llorens X, Jafari HS, Severien C, et al.: Enhanced attenuation of meningeal inflammation and brain edema by concomitant administration of anti-CD18 monoclonal antibodies and dexamethasone in experimental Haemophilus meningitis. J Clin Invest 88:2003, 1991.

44. Gebel JM Jr, Jauch EC, Brott TG, et al.: Natural history of perihematomal edema in patients with hyperacute spontaneous intracerebral hemorrhage.Stroke 33: 2631, 2002.

45. Huang FP, Xi G, Keep RF, et al.: Brain edema after experimental intracerebral hemorrhage: role of hemoglobin degradation products. J Neurosurg 96:287, 2002.

46. Xi G, Hua Y, Bhasin R, et al.: Mechanisms of edema formation after intracerebral hemorrhage: effects of extravasated red blood cells on blood flow and blood–brain barrier integrity. Stroke 32:2932, 2001.

47. Xi G, Reiser G, Keep RF: The role of thrombin and thrombin receptors in ischemic, hemorrhagic and traumatic brain injury: deleterious or protective? J Neurochem 84:3, 2003.

48. Miyasaka Y, Kurata A, Tanaka R, et al.: Mass effect caused by clinically unruptured cerebral arteriovenous malformations. Neurosurgery 41:1060, 1997.

49. Wijdicks EFM, Diringer MN: Middle cerebral artery territory infarction and early brain swelling: progression and effect of age on outcome. Mayo Clin Proc 73:829, 1998.

50. Del Maestro RF, Megyesi JF, Farrell CL: Mechanisms of tumor-associated edema: a review. Can J Neurol Sci 17:177, 1990.

51. Coomber BL, Stewart PA, Hayakawa K, et al.: Quantitative morphology of human glioblastoma multiforme microvessels: structural basis of blood-brain barrier defect. J Neurooncol 5:299, 1987.

52. Cooper C, Jones HG, Weller RO: Production of prostaglandins and thromboxane by isolated cells from intracranial tumours. J Neurol Neurosurg Psychiatry 47:579, 1984.

53. Saris SC, Patronas NJ, Rosenberg SA, et al.: The effect of intravenous interleukin-2 on brain water content. J Neurosurg 71:169, 1989.

54. Posner JB: Neurologic Complications of Cancer. Philadelphia: FA Davis, 1995.

55. Hansen TD, Warner DS, Traynelis VC, et al.: Plasma osmolality and brain water content in a rat glioma model. Neurosurgery 34:505, 1994.

56. Pennings FA, Bouma GJ, Kedaria M, et al.: Intraoperative monitoring of brain tissue oxygen and carbon dioxide pressures reveals low oxygenation in peritumoral brain edema. J Neurosurg Anesthesiol 15:1, 2003.

57. McCrory PR, Berkovic SF: Second impact syndrome. Neurology 50:677, 1998.

58. Eisenberg HM, Gary HE Jr, Aldrich EF, et al.: Initial CT findings in 753 patients with severe head injury. A report from the NIH Traumatic Coma Data Bank. J Neurosurg 73:688, 1990.

59. Lang DA, Teasdale GM, Macpherson P, et al.: Diffuse brain swelling after head injury: more often malignant in adults than children? J Neurosurg 80: 675, 1994.

60. Pearl GS: Traumatic neuropathology. Clin Lab Med 18:39, 1994.

61. Feldman Z, Zachari S, Reichenthal E, et al.: Brain edema and neurological status with rapid infusion of lactated Ringer's or 5% dextrose solution following head trauma. J Neurosurg 83:1060, 1995.

62. Shapira Y, Artru AA, Cotev S, et al.: Brain edema and neurologic status following head trauma in the rat. No effect from large volumes of isotonic or hypertonic intravenous fluids, with or without glucose. Anesthesiology 77:79, 1992.

63. Qureshi AI, Suarez JI, Bhardwaj A, et al.: Use of hypertonic (3%) saline/acetate infusion in the treatment of cerebral edema: effect on intracranial pressure and lateral displacement of the brain. Crit Care Med 26:440, 1998.

64. Suarez JI, Qureshi AI, Bhardwaj A, et al.: Treatment of refractory intracranial hypertension with 23.4% saline. Crit Care Med 26:1118, 1998.

65. Bohn D, Daneman D: Diabetic ketoacidosis and cerebral edema. Curr Opin Pediatr 14:287, 2002.

66. Ebel H, Kuchta J, Balogh A: Operative treatment of tentorial herniation in herpes encephalitis. Childs Nerv Syst 15:84, 1999.

67. Guerra W, Gaab MR, Dietz H: Surgical decompression for traumatic brain swelling: indications and results. J Neurosurg 90:187, 1999.