[Skip to Content]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address 54.205.150.215. Please contact the publisher to request reinstatement.
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
[Skip to Content Landing]
Download PDF
Image description not available.
Figure 1.—Left, Axial proton-weighted magnetic resonance image for patient 1, with arrow demonstrating pronounced increased signal (edema) in splenium of corpus callosum. Right, Similar level axial T2-weighted magnetic resonance image of the same patient 6 weeks later, with no residual abnormal signal.
Image description not available.
Figure 2.—Patient 2, 3 hours after helicopter rescue from 5550 m altitude. Coronal image demonstrates pronounced increased signals in splenium in this proton-weighted magnetic resonance image. Cerebrospinal fluid is gray, splenium (arrow) is white (edematous).
Image description not available.
Figure 3.—Axial T2-weighted magnetic resonance image of patient 3 showing increased signal in splenium.
Image description not available.
Figure 4.—Left, Axial T2-weighted magnetic resonance image of patient 4 showing markedly increased signal in corpus callosum (arrows), including both the genu and the splenium, as well as increased signal of periventricular and subcortical white matter. Right, Axial T2-weighted magnetic resonance image of the same patient 5 weeks after original presentation, demonstrating no residual abnormality in splenium (arrow).
Image description not available.
Figure 5.—Axial T2-weighted magnetic resonance image of patient 8 showing markedly increased signal in splenium and centrum semiovale.
Image description not available.
Figure 6.—Left, Axial T2-weighted magnetic resonance image of patient 9 demonstrating high signal in splenium and mild increased signal in centrum semiovale. Right, Axial T2-weighted magnetic resonance image of the same patient demonstrating complete resolution of abnormal signals 11 months later.
1.
Singh I, Khanna PK, Srivastava MC, Lal M, Roy SB. Acute mountain sickness.  N Engl J Med.1969;280:175-218.
2.
Houston CS, Dickinson JG. Cerebral form of high altitude illness.  Lancet.1975;2:758-761.
3.
Hackett PH, Rennie ID, Levine HO. The incidence, importance, and prophylaxis of acute mountain sickness.  Lancet.1976;2:1149-1154.
4.
Klatzo I. Presidential address: neuropathologic aspect of brain edema.  J Neuropathol Exp Neurol.1967;26:1-14.
5.
Fishman RA. Brain edema.  N Engl J Med.1975;293:706-711.
6.
Hansen JE, Evans WO. A hypothesis regarding the pathophysiology of acute mountain sickness.  Arch Environ Health.1970;21:666-669.
7.
Lassen NA, Harper AM. High altitude cerebral oedema.  Lancet.1975;2:1154.
8.
Sutton JR, Lassen NA. Pathophysiology of acute mountain sickness and high altitude pulmonary edema: an hypothesis.  Bull Eur Physiopathol Respir.1979;15:1045-1052.
9.
Wohns RN. High altitude cerebral edema: a pathophysiological review.  Crit Care Med.1981;9:880-882.
10.
Dickinson JG, Heath J, Gosney J, Williams D. Altitude related deaths in seven trekkers in the Himalayas.  Thorax.1983;38:646-656.
11.
Krasney J. Cerebral hemodynamics and high altitude cerebral edema. In: Houston C, Coates G, eds. Hypoxia: Women at Altitude. Burlington, Vt: Queen City Press; 1997:254-268.
12.
Klatzo I. Pathophysiological aspects of brain edema.  Acta Neuropathol (Berl).1987;72:236-239.
13.
Bradley WB. Pathophysiologic correlates of signal alteration. In: Brant-Zawadzki M, Norman D, eds. Magnetic Resonance Imaging of the Central Nervous System. New York, NY: Raven Press; 1987:29-33.
14.
Kobayashi T, Koyoma S, Kubo K, Fukushima M, Kusama S. Clinical features of patients with high altitude pulmonary edema in Japan.  Chest.1987;92:814-821.
15.
Levine B, Yoshimura K, Kobayashi T, Fukushima M, Shibamoto T, Ueda G. Dexamethasone in the treatment of acute mountain sickness.  N Engl J Med.1989;321:1707-1713.
16.
Roach RC, Bartsch P, Oelz O, Hackett PH. The Lake Louise acute mountain sickness scoring system. In: Sutton JR, Coates G, Houston CS, eds. Hypoxia and Molecular Medicine. Burlington, Vt: Queen City Press; 1993:272-274.
17.
Matsuzawa Y, Kobayashi T, Fujimoto K.  et al.  Cerebral edema in acute mountain sickness. In: Ueda G, Reeves JT, Sekiguchi M, eds. High Altitude Medicine. Matsumoto, Japan: Shinshu University Press; 1992:300-304.
18.
Zavasky DM, Hackett PH. Cerebral etiology of acute mountain sickness: MRI findings [abstract].  Wilderness Environ Med.1995;6:229-230.
19.
Yamaguchi S, Kobayashi T, Matsuzawa Y.  et al.  MRI findings in patients with high altitude pulmonary edema. In: Ueda G, Reeves JT, Sekiguchi M, eds. High Altitude Medicine. Matsumoto, Japan: Shinshu University Press; 1992:305-308.
20.
Garrido E, Castello A, Ventura JL.  et al.  Cortical atrophy and other brain magnetic resonance imaging (MRI) changes after extremely high altitude climbs without oxygen.  Int J Sports Med.1993;14:232-234.
21.
Garrido E, Segura R, Capdevila A.  et al.  New evidence from magnetic resonance imaging of brain changes after climbs at extreme altitude.  Eur J Appl Physiol.1995;70:447-481.
22.
Simonson TM, Yuh WTC. Stroke and cerebral ischemia. In: Edelman RR, Zlatkin MB, Hesselink JR, eds. Clinical Magnetic Resonance Imaging. 2nd ed. Philadelphia, Pa: WB Saunders Co; 1996:767-773.
23.
Moody DM, Bell MA, Challa VR. The corpus callosum, a unique white matter tract: anatomic features that may explain sparing in Binswanger disease and resistance to flow of fluid masses.  AJNR Am J Neuroradiol.1988;9:1051-1059.
24.
Ito T, Sakai T, Inagawa S, Utsu M, Bun T. MR angiography of cerebral vasospasm in preeclampsia.  AJNR Am J Neuroradiol.1995;16:1344-1346.
25.
Cordoliani Y-S, Sarrazin J-L, Felten D, Caumes E, Leveque C, Fisch A. MR of cerebral malaria.  AJNR Am J Neuroradiol.1998;9:871-874.
26.
Le Bihan D, Turner R. Diffusion and perfusion. In: Stark DD, Bradley WG, eds. Magnetic Resonance Imaging. 2nd ed. St Louis, Mo: Mosby–Year Book Inc; 1992:335-369.
27.
Moody DM, Bell MA, Challa VR. Features of the cerebral vascular pattern that predict vulnerability to perfusion or oxygenation deficiency: an anatomic study.  AJNR Am J Neuroradiol.1990;11:431-439.
28.
Ferrazzini G, Maggiorini M, Kriemler S, Bartsch P, Oelz O. Successful treatment of acute mountain sickness with dexamethasone.  Br Med J (Clin Res Ed).1987;294:1380-1382.
29.
Hackett PH, Roach RC, Wood RA.  et al.  Dexamethasone for prevention and treatment of acute mountain sickness.  Aviat Space Environ Med.1988;59:950-954.
30.
Hackett PH, Roach RC. High altitude medicine. In: Auerbach PS, ed. Wilderness Medicine, Management of Wilderness and Environmental Emergencies. St Louis, Mo: Mosby–Year Book Inc; 1995:1-37.
31.
Joo F, Klatzo I. Role of cerebral endothelium in brain oedema.  Neurol Res.1989;11:67-75.
32.
Greenwood J, Begley DJ, Segal MB. New Concepts of a Blood-Brain BarrierNew York, NY: Plenum Press; 1995.
33.
Hauser RA, Lacey DM, Knight MR. Hypertensive encephalopathy: magnetic resonance imaging demonstration of reversible cortical and white matter lesions.  Arch Neurol.1988;45:1078-1083.
34.
Raroque HG, Orrison WW, Rosenberg GA. Neurologic involvement in toxemia of pregnancy: reversible MRI lesions.  Neurology.1990;40:167-169.
35.
Yaffe K, Ferriero D, Barkovich J, Rowley H. Reversible MRI abnormalities following seizures.  Neurology.1995;45:104-108.
36.
Baumgartner RW, Spyridopoulos I, Bartsch P, Maggiorini M, Oelz O. Acute mountain sickness is not related to cerebral blood flow: a decompression chamber study.  J Appl Physiol.In press.
37.
Krasney JA. A neurogenic basis for acute altitude illness.  Med Sci Sports Exerc.1994;26:195-208.
38.
Schilling L, Wahl M. Brain edema: pathogenesis and therapy.  Kidney Int.1997;51(suppl 59):S69-S75.
39.
Plateel M, Dehouck M-P, Torpier G, Cecchelli R, Tessier E. Hypoxia increases the susceptibility to oxidant stress and the permeability of the blood-brain barrier endothelial cell monolayer.  J Neurochem.1995;65:2138-2145.
40.
Bartsch P, Shaw S, Francioli M, Gnadinger MP, Weidmann P. Atrial naturetic peptide in acute mountain sickness.  J Appl Physiol.1988;65:1929-1937.
41.
Richalet JP, Hormych A, Rathat C, Aumont J, Larmignant P, Remy P. Plasma prostaglandins, leukotrienes and thromboxane in acute high altitude hypoxia.  Respir Physiol.1991;85:205-215.
42.
Severinghaus JW. Hypothetical roles of angiogenesis, osmotic swelling, and ischemia in high altitude cerebral edema.  J Appl Physiol.1995;79:375-379.
43.
Ogawa S, Koga S, Kuwabara R.  et al.  Hypoxia-induced increased permeability of endothelial monolayers occurs through lowering of cellular cAMP levels.  Am J Physiol.1992;262:C546-C554.
Original Contribution
December 9, 1998

High-Altitude Cerebral Edema Evaluated With Magnetic Resonance ImagingClinical Correlation and Pathophysiology

Author Affiliations

From the School of Health Professions, University of Alaska (Dr Hackett), and the Departments of Emergency Medicine (Dr Hackett) and Radiology (Drs Hill and McCormick), Alaska Regional Hospital, Anchorage; Department of Medicine, University of Washington School of Medicine, Seattle (Dr Hackett); and St Anthony Hospital (Drs Yarnell, Reynard, and Heit) and Department of Neurology, University of Colorado School of Medicine (Dr Yarnell), Denver. Dr Hackett is now with St Mary's Hospital and Medical Center, Grand Junction, Colo.

JAMA. 1998;280(22):1920-1925. doi:10.1001/jama.280.22.1920
Context.—

Context.— Because of its onset in generally remote environments, high-altitude cerebral edema (HACE) has received little scientific attention. Understanding the pathophysiology might have implications for prevention and treatment of both this disorder and the much more common acute mountain sickness.

Objectives.— To identify a clinical imaging correlate for HACE and determine whether the edema is primarily vasogenic or cytotoxic.

Design.— Case-comparison study.

Setting.— Community hospitals accessed by helicopter from mountains in Colorado and Alaska.

Patients.— A consecutive sample of 9 men with HACE, between 18 and 35 years old, 8 of whom also had pulmonary edema, were studied after evacuation from high-altitude locations; 5 were mountain climbers and 4 were skiers. The control group, matched for age, sex, and altitude exposure, consisted of 3 subjects with high-altitude pulmonary edema only and 3 who had been entirely well at altitude. Four patients with HACE were available for follow-up imaging after complete recovery.

Main Outcome Measures.— Magnetic resonance imaging (MRI) of the brain during acute, convalescent, and recovered phases of HACE, and once in controls, immediately after altitude exposure.

Results.— Seven of the 9 patients with HACE showed intense T2 signal in white matter areas, especially the splenium of the corpus callosum, and no gray matter abnormalities. Control subjects demonstrated no such abnormalities. All patients completely recovered; in the 4 available for follow-up MRI, the changes had resolved entirely.

Conclusions.— We conclude that HACE is characterized on MRI by reversible white matter edema, with a predilection for the splenium of the corpus callosum. This finding provides a clinical imaging correlate useful for diagnosis. It also suggests that the predominant mechanism is vasogenic (movement of fluid and protein out of the vascular compartment) and, thus, that the blood-brain barrier may be important in HACE.

HIGH-ALTITUDE CEREBRAL edema (HACE) is a potentially fatal neurologic syndrome that develops over hours or days in persons with acute mountain sickness (AMS) or high-altitude pulmonary edema (HAPE).13 This type of high-altitude illness typically presents with altered mental status, ataxia, and progressive neurologic deterioration and is considered to be the end stage of AMS. Although hypoxia is the triggering stimulus, the exact pathophysiology of HACE, as well as AMS, is still not known. In the framework of the 2 general types of cerebral edema,4,5 authors have postulated that HACE may be either cytotoxic (intracellular, due to cellular swelling),2,6 vasogenic (extracellular, due to blood-brain barrier leakage),1,7,8 or a combination of the 2.9 The distinction has important implications not only for prevention and treatment but also for determining the direction of future research on the basic mechanisms of altitude illness. Brain edema found on autopsy has been so severe and global that it has not been helpful in elucidating the initial pathophysiology.1,10 One brain biopsy done to exclude other diagnoses reported microscopic evidence of "edematous brain" without further distinction.1 Work with a sheep model of altitude illness that showed extravasation of Evans blue dye in the brain suggested vasogenic edema.11

In vivo brain imaging, by localizing the edema earlier in the course of the illness, might offer new insights into the pathophysiology. Vasogenic edema preferentially spreads along white matter tracts, whereas cytotoxic edema affects both gray and white matter, especially the former.4,5,12,13 Previous imaging studies were primarily with computed tomography (CT). Kobayashi et al14 performed axial CT scans on 9 mountain climbers with HAPE and neurologic signs. In 8 subjects, the CT scan revealed small ventricles and cisterns, disappearance of sulci, and a diffuse low-density appearance of the entire cerebrum, all of which resolved slowly on recovery. Levine et al15 also reported diffuse low density on CT of the brain in 1 subject with AMS. Although confirming the presence of cerebral edema, the CT scans did not differentiate regional differences or help define mechanisms.

Using magnetic resonance imaging (MRI), we sought to differentiate white from gray matter edema to identify the operant pathophysiologic mechanism. In addition, we hoped to find a reliable clinical imaging correlate that could be helpful in the diagnosis and assessment of HACE. We studied 9 consecutive patients with a clinical diagnosis of HACE, 8 of whom also had HAPE. Magnetic resonance imaging revealed characteristic changes (increased T2 signal in the white matter, especially in the splenium of the corpus callosum, and no gray matter edema) in 7 of the 9 subjects. Control subjects had no such changes. Four subjects with repeat MRI after complete recovery showed total resolution of the abnormalities. Magnetic resonance imaging may provide a useful in vivo clinical imaging correlate for HACE, and the findings strongly suggest a predominantly vasogenic mechanism.

METHODS

The study subjects were all patients admitted to 2 community hospitals during a 4-year period with the diagnosis of HACE. The authors (P.H.H. or P.R.Y.) cared for the study subjects. The control group consisted of 6 climbers without symptoms of HACE who had been climbing higher than 5000 m on Mt McKinley and who had MRI performed within 24 hours of returning to sea level. Three control subjects were entirely well at high altitude and 3 had developed HAPE. The control group matched the patient group for mean age, sex, and length of altitude exposure. Magnetic resonance images were obtained with either a GE 1.5-T Signa (Anchorage, Alaska) or a Siemens 1.0-T Magneton (Denver, Colo). We were able to perform repeat MRI during recovery in 5 study patients and after complete recovery in 4 study patients.

RESULTS

Table 1 presents the clinical and imaging data on our 9 patients. All were vigorous, healthy men aged 18 to 35 years (mean, 27.9 years). None had a history of significant medical problems and none were being treated for any acute illness. Previous high-altitude experience varied. Four had history of AMS but none had history of HACE or HAPE. Four developed HACE while skiing in Colorado and 5 while attempting to climb Mt McKinley (6194 m), also known by its Alaskan name, Denali. Eight received initial treatment in the field with oxygen, 5 received dexamethasone, 2 were given acetazolamide and nifedipine, and 1 was treated in a portable hyperbaric bag, which simulates descent. All 9 patients required hospitalization either in Denver (1610 m) or in Anchorage (sea level).

All subjects met the criteria for diagnosis of AMS as well as for HACE (all had mental status changes and/or ataxia in association with AMS).16 Eight had clinical diagnosis of HAPE confirmed by radiographs. Respiratory alkalosis of varying degree was present in all patients and severe hypoxemia was present in the 8 with HAPE. Five patients had retinal hemorrhages and all had ataxic gait.

Hospital course varied considerably among patients. Average stay was 5.6 days, with a range of 1 to 15 days. Three patients required intubation. All recovered without evidence of permanent sequelae. Time to normal neurologic examination averaged 2.4 weeks, with a range of 1 day to 6 weeks. The imaging findings are presented in Table 1 and Figure 1, Figure 2, Figure 3, Figure 4, Figure 5, through Figure 6. The 3 most seriously ill patients (1, 4, and 9) had the most striking changes on MRI, but 2 moderately ill patients (6 and 7) had normal MRI findings. Time from onset of clinical cerebral edema to MRI varied from 16 to 132 hours (mean, 58 hours). Four patients with initial abnormal MRI results still had the abnormalities on repeat imaging performed between 3 and 11 days later, although they were clinically improved. These 4 subjects demonstrate that resolution of MRI findings can lag behind clinical improvement. Patients 1 and 4 had normal MRI findings at 6 weeks, and patients 2 and 9 had normal MRI findings on follow-up at 3 and 11 months, respectively. All control subjects had normal MRI results.

COMMENT

Magnetic resonance imaging in 7 of 9 patients with HACE demonstrated strikingly increased T2 signal in the corpus callosum, particularly in the splenium, with additional involvement in the centrum semiovale. The gray matter was normal. These abnormalities had resolved on subsequent MRI studies obtained in 4 patients, and all of the patients had complete clinical recovery. High-altitude controls, both healthy individuals and those with HAPE, did not show MRI abnormalities. This MRI pattern of reversible edema limited to white matter provides an imaging correlate for the syndrome of HACE and strongly suggests a predominantly vasogenic mechanism as the basis of the edema.

A potential problem of this study is the diagnosis of HACE. Although our patients' illnesses were consistent with HACE, other diagnoses must be considered. History, clinical course, and diagnostic evaluation, including brain MRI, excluded cerebrovascular events and brain trauma. Three patients presented with elevated temperature, but this is common in HAPE, and brain infection was excluded in these 3 by negative cerebrospinal fluid cultures. The possibility of anoxic encephalopathy secondary to the pulmonary edema was excluded by lack of the typical gray-matter lesions on MRI and by clinical course. Vasculitis may cause white-matter MRI changes and was also considered in this group, perhaps secondary to sympathomimetic drug use. However, all subjects denied such drug use and toxicology findings were negative in all 5 patients tested. In summary, given the setting of acute ascent to high altitude, the reversibility of the syndrome, and the exclusion of other illnesses, we are reasonably certain of the clinical diagnosis of HACE.

Another concern is whether white matter edema might be an incidental finding of high altitude. The fact that 3 asymptomatic climbers studied within 24 hours of returning from high altitude (6194 m) had no MRI abnormalities speaks against this, as does other work showing no increased T2 signal intensity in 8 subjects with mild or absent AMS after simulated altitude exposure.17,18 The brain edema does not seem related directly to HAPE or severe hypoxemia because 3 extremely hypoxemic individuals with HAPE but without HACE also had normal MRI results. Although a greater number of control MRIs is desirable, we feel confident that our findings truly correlate with the diagnosed clinical illness.

A third concern is that 2 patients with HACE did not demonstrate the MRI abnormalities. The reason for this is not clear. Overall, all 5 high-altitude climbers, but only 2 of 4 skiers, had positive MRI results for HACE. Possible factors are that the climbers had all been at considerably higher altitude, their length of altitude exposure was longer, and their evacuation to low altitude was slower. In addition, the climbers were engaged in activities more likely to elevate cerebral capillary pressure, such as lifting and carrying heavy packs and pulling themselves up ropes. Duration and severity of symptoms and the treatment prior to MRI were similar in those with and without MRI changes. Because the MRI findings were not present in all cases, they are considered characteristic but not prerequisite for the diagnosis of HACE. Since resolution of MRI abnormalities lagged behind clinical improvement, their presence may help establish the diagnosis in a person recently recovered.

Our findings are consistent with other recent studies regarding imaging of AMS and HACE in the literature. Levine et al15 found CT evidence of mildly decreased white matter density in the sickest of 6 subjects with AMS after a 48-hour simulated altitude exposure. Similarly, Matsuzawa et al17 were able to detect slight increased T2 signal of white matter in the sickest 4 of 7 subjects with AMS in a 24-hour simulated altitude experiment. Yamaguchi and colleagues19 performed MRI on 4 patients with HAPE and neurologic symptoms and found only slightly increased T2 signal, in contrast with the dramatic signal alterations in the current study. This is most likely because they studied the 2 subjects with cerebral edema only during recovery and the other 2 had only mild neurologic dysfunction. Follow-up studies were not performed in any of these studies; the changes were presumed to be transient. Slight increased T2 signal of white matter in AMS and intense signal in HACE support the notion that the 2 illnesses reflect a continuum of the same pathophysiology.3 However, only the advanced pathology of HACE provided the clear and unusual pattern of corpus callosum edema, the predilection for the splenium, and a useful clinical imaging correlate.

Our present findings are distinct from the white-matter hyperintense changes noted on MRI in subjects after climbs to extreme altitude without oxygen. Garrido et al20 detected signal abnormalities in 7 of 26 subjects 1 month to 3 years after the most recent high-altitude exposure. There was no prior or subsequent imaging and no diagnoses could be established retrospectively. Although many subjects reported some neurologic symptoms, none apparently had HACE. The increased signals in this study and the subsequent investigation21 were found in various deep white-matter tracts but not in the corpus callosum or splenium, and they were considered nonreversible. The mechanism of these changes is unknown. Although other MRI abnormalities may appear after extreme altitude exposure, our findings seem to be specific for acute HACE.

The finding that HACE is primarily manifested in the white matter in vivo has pathophysiologic implications. Reversible increased T2 signal abnormality of white matter without gray matter involvement suggests a vasogenic type of cerebral edema as the predominant mechanism at this stage of the illness and at least a relative absence of cytotoxic edema.13 Gray matter consists of tightly packed, tangled cellular structures, whereas white matter has an orderly network of extracellular channels, is less dense, and offers less resistance to invasion by edema fluid.5,12 Vasogenic edema thus spreads preferentially through the white matter. In contrast, gray matter is more sensitive to imbalance of cellular energy demand and supply and, therefore, more susceptible to the cell swelling of cytotoxic edema. Also, cytotoxic edema most commonly displays morphologic changes on T1-weighted images,22 which we did not observe in gray or white matter. Thus, the presence of reversible white-matter changes and absence of gray-matter changes both support the conclusion of a vasogenic edema.

The remarkable predilection for the splenium and corpus callosum in our MRI images is puzzling. This is in contrast with the classic teaching of resistance to flow of fluid in the corpus callosum because of its thickly packed fibers.23 However, because edema is seen in the corpus callosum in certain situations, this is merely a relative resistance. With the increased availability and use of MRI, more reports of edema of the corpus callosum and splenium are appearing.24,25 Additionally, recent measurements indicate that the degree of water movement (or accumulation) depends much on orientation of the myelin fibers.26 Flow along the axons is much less impeded than across them and the very high signals seen in our MRIs may indicate accumulation of water parallel to the axons. Another possibility is that the edema is caused by increased vascular permeability in the corpus callosum itself. Because of the short arterioles and relative lack of pressure drop along the vessels, the unique vascular anatomy of the corpus callosum that provides relative protection from hypoperfusion and ischemia23,27 may render it more susceptible to edema in the setting of hypoxic cerebral vasodilatation at high altitude.7 However, these suggestions are very speculative. The fact that the signals are on T2-weighted images, reversible, and only in white matter indicate this is a vasogenic edema. An explanation for this relatively unusual location of water in the brain may yield important insights into the mechanisms of AMS and HACE. Development of an animal model evincing a similar edema pattern should be a high priority for further investigation.

Further evidence for the concept of vasogenic edema is based on important clinical observations. Acute mountain sickness, the early form of HACE, is effectively prevented and treated by steroids15,28,29 and, anecdotally, steroids are also beneficial when given early in the course of HACE.1,30 It is well known that vasogenic edema responds to steroids whereas cytotoxic brain edema does not.5 The slow resolution of edema is also consistent with a vasogenic mechanism because removal of extravasated proteins is primarily by the relatively slow process of astrocyte pinocytosis.12,31 Furthermore, a predominantly vasogenic edema tends to leave the brain tissue well preserved after resolution, as is demonstrated in HACE by the usual complete clinical recovery and normalization of MRI. Cytotoxic edema, in contrast, is generally not so benign.12

The actual mechanism of this vasogenic edema is still unclear. By definition, vasogenic edema results from opening of the blood-brain barrier. A complete discussion of the blood-brain barrier and vasogenic edema is beyond the scope of this article. Recent excellent reviews are available32 and the summary by Krasney11 with respect to AMS, HACE, and the blood-brain barrier is of particular interest. A few facts are worth mentioning for the purposes of our discussion. Some researchers have proposed that hypoxic vasodilatation results in a failure of autoregulation,1,7,8 a hypothesis that has not yet been directly tested. Their view is that increased cerebral capillary hydrostatic pressure, not necessarily with systemic hypertension, may result in blood-brain barrier opening and outflow of fluid along white-matter tracts. Indeed, a number of clinical conditions exhibiting vasogenic edema on MRI are thought to share this mechanism. These include hypertensive encephalopathy, toxemia of pregnancy, seizures, acute intermittent porphyria, cyclosporine toxic effects, and migraine.3335 Hypoxic cerebral vasodilatation on ascent to high altitude is well documented but whether this in itself explains vasogenic edema is doubtful. All persons at high altitude have cerebral vasodilatation and those with AMS or HACE are not appreciably different in this regard than those who are well.36 Whether regional changes in autoregulation, cerebral blood flow, and cerebral capillary pressure are sufficient to produce vasogenic edema remains unanswered.

The animal model of AMS and HACE developed by Krasney37 also supports the vasogenic hypothesis but suggests additional mechanisms. In animals that became ill, Krasney found increased wet-dry brain tissue ratio and extravasation of Evans blue dye, confirming leak of the blood-brain barrier, in conjunction with increased cerebral blood flow and elevated brain capillary pressures, without systemic hypertension. However, further experiments disclosed that brain edema from large increases in capillary hydrostatic pressure induced by carbon dioxide breathing or nitroglycerin was much less than that resulting from hypoxic edema, suggesting that another factor than capillary hydrostatic pressure must be in play.11 This other important factor is the conductance or permeability of the blood-brain barrier itself and, primarily, the role of the cerebral endothelium.31 A quite complex regulatory mechanism, the blood-brain barrier is influenced by many factors, including adrenergic and cholinergic systems, neurotransmitters and neuromodulators, cyclic nucleotides, nitric oxide, histamine, and cytokines and other factors released by white cell–endothelial interaction.11,31,38 All of these are candidates for being altered by hypoxia39 and some, like atrial naturetic peptide, norepinephrine, and eicosanoids, are known to be altered in AMS.40,41 In addition, the beneficial effect of steroids in altitude illness might likely be due to an effect on blood-brain barrier permeability. Dexamethasone is known to suppress lipid peroxidation as well as block vascular endothelial growth factor and therefore angiogenesis, both of which actions make the barrier less permeable.11,31,42 Dexamethasone also prevents the increased permeability of cultured endothelial cell monolayers that are subjected to hypoxia.43 Of course, increased blood-brain barrier permeability and capillary hydrostatic pressure may be working in concert; flux of fluid is greatly influenced by hydrostatic pressure in the presence of an opening of the blood-brain barrier.12

Our findings do not exclude an element of intracellular edema as well but strongly suggest that vasogenic edema is the major operant factor in the pathophysiology of HACE, at least in the phase in which it becomes clinically evident. As Klatzo4,12 has described, progressive extracellular vasogenic edema, by increasing intercapillary distances, will affect the energy requirements of the cells, eventually rendering them ischemic and leading them to swell, thus contributing to a further increase in intracranial pressure. This cytotoxic mechanism, traditionally invoked to explain HACE,2 likely becomes operant only at this late stage. Successful treatment must begin while the condition is still reversible, before onset of neuronal damage. The next focus of research into AMS and HACE should be on the role of the blood-brain barrier.

References
1.
Singh I, Khanna PK, Srivastava MC, Lal M, Roy SB. Acute mountain sickness.  N Engl J Med.1969;280:175-218.
2.
Houston CS, Dickinson JG. Cerebral form of high altitude illness.  Lancet.1975;2:758-761.
3.
Hackett PH, Rennie ID, Levine HO. The incidence, importance, and prophylaxis of acute mountain sickness.  Lancet.1976;2:1149-1154.
4.
Klatzo I. Presidential address: neuropathologic aspect of brain edema.  J Neuropathol Exp Neurol.1967;26:1-14.
5.
Fishman RA. Brain edema.  N Engl J Med.1975;293:706-711.
6.
Hansen JE, Evans WO. A hypothesis regarding the pathophysiology of acute mountain sickness.  Arch Environ Health.1970;21:666-669.
7.
Lassen NA, Harper AM. High altitude cerebral oedema.  Lancet.1975;2:1154.
8.
Sutton JR, Lassen NA. Pathophysiology of acute mountain sickness and high altitude pulmonary edema: an hypothesis.  Bull Eur Physiopathol Respir.1979;15:1045-1052.
9.
Wohns RN. High altitude cerebral edema: a pathophysiological review.  Crit Care Med.1981;9:880-882.
10.
Dickinson JG, Heath J, Gosney J, Williams D. Altitude related deaths in seven trekkers in the Himalayas.  Thorax.1983;38:646-656.
11.
Krasney J. Cerebral hemodynamics and high altitude cerebral edema. In: Houston C, Coates G, eds. Hypoxia: Women at Altitude. Burlington, Vt: Queen City Press; 1997:254-268.
12.
Klatzo I. Pathophysiological aspects of brain edema.  Acta Neuropathol (Berl).1987;72:236-239.
13.
Bradley WB. Pathophysiologic correlates of signal alteration. In: Brant-Zawadzki M, Norman D, eds. Magnetic Resonance Imaging of the Central Nervous System. New York, NY: Raven Press; 1987:29-33.
14.
Kobayashi T, Koyoma S, Kubo K, Fukushima M, Kusama S. Clinical features of patients with high altitude pulmonary edema in Japan.  Chest.1987;92:814-821.
15.
Levine B, Yoshimura K, Kobayashi T, Fukushima M, Shibamoto T, Ueda G. Dexamethasone in the treatment of acute mountain sickness.  N Engl J Med.1989;321:1707-1713.
16.
Roach RC, Bartsch P, Oelz O, Hackett PH. The Lake Louise acute mountain sickness scoring system. In: Sutton JR, Coates G, Houston CS, eds. Hypoxia and Molecular Medicine. Burlington, Vt: Queen City Press; 1993:272-274.
17.
Matsuzawa Y, Kobayashi T, Fujimoto K.  et al.  Cerebral edema in acute mountain sickness. In: Ueda G, Reeves JT, Sekiguchi M, eds. High Altitude Medicine. Matsumoto, Japan: Shinshu University Press; 1992:300-304.
18.
Zavasky DM, Hackett PH. Cerebral etiology of acute mountain sickness: MRI findings [abstract].  Wilderness Environ Med.1995;6:229-230.
19.
Yamaguchi S, Kobayashi T, Matsuzawa Y.  et al.  MRI findings in patients with high altitude pulmonary edema. In: Ueda G, Reeves JT, Sekiguchi M, eds. High Altitude Medicine. Matsumoto, Japan: Shinshu University Press; 1992:305-308.
20.
Garrido E, Castello A, Ventura JL.  et al.  Cortical atrophy and other brain magnetic resonance imaging (MRI) changes after extremely high altitude climbs without oxygen.  Int J Sports Med.1993;14:232-234.
21.
Garrido E, Segura R, Capdevila A.  et al.  New evidence from magnetic resonance imaging of brain changes after climbs at extreme altitude.  Eur J Appl Physiol.1995;70:447-481.
22.
Simonson TM, Yuh WTC. Stroke and cerebral ischemia. In: Edelman RR, Zlatkin MB, Hesselink JR, eds. Clinical Magnetic Resonance Imaging. 2nd ed. Philadelphia, Pa: WB Saunders Co; 1996:767-773.
23.
Moody DM, Bell MA, Challa VR. The corpus callosum, a unique white matter tract: anatomic features that may explain sparing in Binswanger disease and resistance to flow of fluid masses.  AJNR Am J Neuroradiol.1988;9:1051-1059.
24.
Ito T, Sakai T, Inagawa S, Utsu M, Bun T. MR angiography of cerebral vasospasm in preeclampsia.  AJNR Am J Neuroradiol.1995;16:1344-1346.
25.
Cordoliani Y-S, Sarrazin J-L, Felten D, Caumes E, Leveque C, Fisch A. MR of cerebral malaria.  AJNR Am J Neuroradiol.1998;9:871-874.
26.
Le Bihan D, Turner R. Diffusion and perfusion. In: Stark DD, Bradley WG, eds. Magnetic Resonance Imaging. 2nd ed. St Louis, Mo: Mosby–Year Book Inc; 1992:335-369.
27.
Moody DM, Bell MA, Challa VR. Features of the cerebral vascular pattern that predict vulnerability to perfusion or oxygenation deficiency: an anatomic study.  AJNR Am J Neuroradiol.1990;11:431-439.
28.
Ferrazzini G, Maggiorini M, Kriemler S, Bartsch P, Oelz O. Successful treatment of acute mountain sickness with dexamethasone.  Br Med J (Clin Res Ed).1987;294:1380-1382.
29.
Hackett PH, Roach RC, Wood RA.  et al.  Dexamethasone for prevention and treatment of acute mountain sickness.  Aviat Space Environ Med.1988;59:950-954.
30.
Hackett PH, Roach RC. High altitude medicine. In: Auerbach PS, ed. Wilderness Medicine, Management of Wilderness and Environmental Emergencies. St Louis, Mo: Mosby–Year Book Inc; 1995:1-37.
31.
Joo F, Klatzo I. Role of cerebral endothelium in brain oedema.  Neurol Res.1989;11:67-75.
32.
Greenwood J, Begley DJ, Segal MB. New Concepts of a Blood-Brain BarrierNew York, NY: Plenum Press; 1995.
33.
Hauser RA, Lacey DM, Knight MR. Hypertensive encephalopathy: magnetic resonance imaging demonstration of reversible cortical and white matter lesions.  Arch Neurol.1988;45:1078-1083.
34.
Raroque HG, Orrison WW, Rosenberg GA. Neurologic involvement in toxemia of pregnancy: reversible MRI lesions.  Neurology.1990;40:167-169.
35.
Yaffe K, Ferriero D, Barkovich J, Rowley H. Reversible MRI abnormalities following seizures.  Neurology.1995;45:104-108.
36.
Baumgartner RW, Spyridopoulos I, Bartsch P, Maggiorini M, Oelz O. Acute mountain sickness is not related to cerebral blood flow: a decompression chamber study.  J Appl Physiol.In press.
37.
Krasney JA. A neurogenic basis for acute altitude illness.  Med Sci Sports Exerc.1994;26:195-208.
38.
Schilling L, Wahl M. Brain edema: pathogenesis and therapy.  Kidney Int.1997;51(suppl 59):S69-S75.
39.
Plateel M, Dehouck M-P, Torpier G, Cecchelli R, Tessier E. Hypoxia increases the susceptibility to oxidant stress and the permeability of the blood-brain barrier endothelial cell monolayer.  J Neurochem.1995;65:2138-2145.
40.
Bartsch P, Shaw S, Francioli M, Gnadinger MP, Weidmann P. Atrial naturetic peptide in acute mountain sickness.  J Appl Physiol.1988;65:1929-1937.
41.
Richalet JP, Hormych A, Rathat C, Aumont J, Larmignant P, Remy P. Plasma prostaglandins, leukotrienes and thromboxane in acute high altitude hypoxia.  Respir Physiol.1991;85:205-215.
42.
Severinghaus JW. Hypothetical roles of angiogenesis, osmotic swelling, and ischemia in high altitude cerebral edema.  J Appl Physiol.1995;79:375-379.
43.
Ogawa S, Koga S, Kuwabara R.  et al.  Hypoxia-induced increased permeability of endothelial monolayers occurs through lowering of cellular cAMP levels.  Am J Physiol.1992;262:C546-C554.
×