High-Altitude Cerebral Edema

Abstract & Commentary

Synopsis: A recent publication that appeared in December in JAMA addresses the pathophysiology of high-altitude cerebral edema and its etiology, as demonstrated on magnetic resonance imaging. Before prevention and treatment of this disorder can be approached effectively, studies such as this are necessary to gain a better understanding of the basis for the disorder and to define its earliest manifestations.

Source: Hackett PH, et al. JAMA 1998;280:1920-1925.

At the far end of the spectrum we call acute mountain sickness, often following the onset of high-altitude pulmonary edema (HAPE), lies the realm of reversible hypoxia induced high-altitude cerebral edema (HACE). Brain edema justifiably has been considered the pathophysiologic basis for this disorder, but, based upon limited animal data obtained in sheep, a single brain biopsy performed at autopsy and previous computerized tomographic (CT) imaging studies. The latter clearly demonstrated cerebral edema in vivo, but could not elucidate either the mechanisms or regional differences within brain tissue for this slowly resolving process. Magnetic resonance imaging (MRI) has demonstrated white matter changes following climbs to extreme altitude in subjects who apparently did not demonstrate HACE.1,2 However, the deep lesions of white matter that were shown in such climbers may not be specifically related to HACE and were not considered reversible.

For the purposes of both this study and others that preceded it, a clear distinction has been made by Hackett and associates between two general types of cerebral edema. Cytotoxic cerebral edema is intracellular in origin and dependent upon cellular swelling in response to insult or injury. Vasogenic cerebral edema is extracellular and ascribed to leaks in the blood-brain barrier. White matter is less dense than gray matter and is more prone to vasogenic edema, since its orderly array of extracellular channels offers less resistance than gray matter to invasion and spread of edema fluid. Any combination of these mechanisms could also account for the acute changes observed in patients with HACE. The objective of this study was to determine whether a clinical imaging correlate for HACE could be identified using MRI and, in the investigative process, to determine to what extent either vasogenic or cytotoxic cerebral edema contributed to the process. This distinction is extremely important if both prevention and therapy of HACE are to have a rational basis, given the large number of potential interventions that could be applied to the problem.

This study was designed as a case-comparison investigation performed in hospitals, which were accessed by helicopter and located in either Alaska or Colorado. Brain MRIs were performed during acute, convalescent, and recovery phases of HACE in consecutive study subjects and once in matched controls, immediately following high-altitude exposures. Study patients consisted of nine men, either climbers or skiers between ages 18 and 35, eight of whom also had HAPE, and had been evacuated from high-altitude locations. None had a history of previous HACE. Prior to hospitalization, study patients received initial treatment in the field with oxygen. Five had also received dexamethasone. Two received acetazolamide and nifedipine, and one was treated in a portable hyperbaric bag. All had mental status changes and/or ataxia in association with acute mountain sickness (AMS). Anoxic encephalopathy and toxic drug effects were excluded. Five study patients received repeat MRI scans during recovery, and four were also available for follow-up MRI after complete recovery. Six age-matched men who served as controls were equally divided between those who had been entirely well at altitude and three who had experienced HAPE without HACE. Their MRI scans were performed within 24 hours of returning to sea level, and they also matched the patient study group in their length of altitude exposures, which for all control subjects had occurred on Denali (Mt. McKinley).

Seven of nine study patients demonstrated intense white matter abnormalities suggesting vasogenic cerebral edema. Two patients with moderate illness had normal MRI findings. These appeared as intense T2 signals exclusively in white matter, particularly in the posterior rounded end of the corpus callosum, known as the splenium, and in the centrum semiovale, but no changes were observed in gray matter. (See Figure.) All study patients recovered completely and their MRI changes resolved entirely on follow-up evaluations, although MRI resolution lagged behind clinical improvement. Control subjects showed no abnormalities on MRI. Hackett et al demonstrated what may be a useful diagnostic MRI correlate (i.e., reversible white matter edema and an early predilection for the posterior region of the splenium of the corpus callosum in patients with acute clinical HACE). The predominant mechanism was likely to be vasogenic edema, which primarily spreads along white matter tracts and involves a breakdown in the blood-brain barrier with movement of fluid and protein out of the vascular compartment. Hackett et al point out that this form of edema, although it cannot be entirely explained as yet, tends to respond to corticosteroid treatment and to resolve slowly, with preservation of brain tissue.

Comment by Frank J. Bia, MD, MPH

Hackett et al address several concerns in their own discussion of this work. Is white matter edema an incidental finding at high altitude? Three of their asymptomatic control climbers who were evaluated within 24 hours of returning from high altitude had no MRI abnormalities. This study also does confirm previous work indicating that brain edema does not seem to be caused by HAPE; however, a greater number of controls in both groups (climbers and skiers) would be of importance to truly exclude such causal associations. It is also not clear why two of four skiers with HACE in the study group did not demonstrate MRI changes. Hackett et al speculate this may be related to the increased effort, at higher altitudes, for more prolonged periods, among the climbers with HACE. One should also be aware that this study cannot be generalized to all those who experience HACE at high altitude since both study subjects and controls were all men. Whether there are gender differences in the cerebral manifestations of HACE on MRI scanning will have to await further study, but it should not be assumed that future results that might be obtained in women at high altitude will be the same as those obtained here.

In reviewing this informative work, it brought to mind a series of three unusual cases reported by Schlim and Meijer earlier this decade.3 They described acute manifestations of previously unsuspected brain tumors, which suddenly became symptomatic during acute exposure to high altitudes. Their symptoms might have easily been attributed solely to AMS, had they not persisted and been carefully evaluated. These included the onset of progressive right-sided hemiparesis in a 69-year-old woman after she had first experienced headaches upon arrival in Lhasa, Tibet. Subsequently, CT scanning revealed a left frontoparietal meningioma, later removed in Sweden. They also reported a 20-year-old woman with an unsuspected right frontal lobe meningioma who experienced left hemiparesis, fever, and seizures on arrival in Cusco, Peru, and a 16-year-old boy on a ski vacation in the Dolomites, who developed symptoms of headache, vomiting, and ataxia, later found due to a unsuspected malignant astrocytoma obstructing the third ventricle. Perhaps these lesions became clinically evident as acute exposures to high altitude induced acute cerebral edema in white matter that would have resolved, as described in the current study by Hackett et al, had they not occurred in patients who already had underlying lesions of the CNS? The current study contributes further to our understanding of both pathogenesis and natural history of HACE. When HACE fails to resolve or presents with atypical features, then we have to recognize such episodes as potentially representing more than HACE alone. It is studies such as these which will permit clinicians to do so with a greater degree of certainty in the future.


1. Garrido E, et al. Int J Sports Med 1993;14:232-234.

2. Garrido E, et al. Eur J Appl Physiol 1995;70:447-481.

3. Schlim DR, Meijer HJ. Ann Emerg Med 1991;20:315.