Time Course of Early Brain Edema Following Reversible Forebrain Ischemia in Rats

Transcription

1 565 Time Course of Early Brain Edema Following Reversible Forebrain Ischemia in Rats Pekka Mellergard, MD, Finn Bengtsson, MD, PhD, Maj-Lis Smith, PhD, Vendela Riesenfeld, MD, and Bo K. Siesjo, MD, PhD Downloaded from by on January, 209 Cerebral ischemia is known to be accompanied by brain edema. This increase in brain tissue water content probably influences the final outcome of an ischemic insult negatively. Despite extensive investigations on different aspects of brain edema, information on edema development during the early recirculation period following ischemia is sparse. We assessed changes in brain water content, as reflected by changes in tissue density, during the early recirculation period following severe forebrain ischemia. Fasted rats were subjected to 5,5, or 30 minutes of ischemia and 5 to 0 minutes of recirculation. The specific gravity of specimens from the caudoputamen, frontoparietal cortex, hippocampus, and mesencephalon were measured with a Percoll linear density gradient. Five minutes of ischemia followed by recirculation did not produce any significant regional brain edema. However, following 5 minutes of ischemia, transient edema developed in the caudoputamen, frontoparietal cortex, and hippocampus. This edema was maximal after 30 minutes of reperfusion and was normalized after 0 minutes of reperfusion. Similar edema was seen following 30 minutes of ischemia. In the mesencephalon (where blood flow is approximately 50% of control during the ischemic insult) no brain edema was noted following 5, 5, or 30 minutes of ischemia. We discuss to what extent this transient regional brain edema may influence the selective neuronal vulnerability and cell damage observed in rats subjected to reversible forebrain ischemia and how these findings may correlate with neurochemical alterations observed during the early recirculation period. (Stroke 99;20: ) Cerebral ischemia is known to be accompanied by an increase in brain tissue water content, that is, brain edema. - This edema probably influences the outcome following an ischemic insult. ' 5 Edema may thus compromise cerebral microcirculation and aggravate an ongoing ischemic process, or it may prevent adequate recovery of tissue blood flow during reperfusion. 5 ' 6 In addition, previous as well as more recent studies emphasize the possibility of localized cellular swelling and osmolytic damage of dendrites secondary to the release of glutamate or related excitatory amino acids. 7 - Postischemic brain edema has been suggested to be of two main types. ' 2 Early edema (cytotoxic or From the Laboratory for Experimental Brain Research (M- L.S., B.K.S.) and the Departments of Neurosurgery (P.M.), Clinical Pharmacology (F.B.), and Diagnostic Radiology (V.R.), University of Lund, Lund, Sweden. Supported by Swedish Medical Research Council Grant No. X-263, United States Public Health Service Grant No. 5 RO NS073, the Tore Nilsson Foundation for Medical Research, Stockholm, Sweden, and the Crafoordska Foundation, Lund, Sweden. Address for correspondence: Dr. Pekka Mellergard, Department of Neurosurgery, University Hospital, S-22 5 Lund, Sweden. Received January 2, 99; accepted June 23, 99. cellular edema) is characterized by an intracellular accumulation of water, with or without a concomitant reduction of the extracellular space. 9 Later (on the order of hours) this early edema is followed by a second edema (vasogenic edema) associated with changes in the cerebral blood vessels, permitting leakage of large molecules such as proteins. 2-3 ' 0 ' The mechanisms of the two types of edema are still not fully understood Although much information is available on the pathophysiology of brain edema, only a few studies have explored the time course of edema during the early recirculation period following ischemia. Knowledge of this time course is required to allow better characterization of the mechanisms involved in the formation of edema. Therefore, we measured the time course of cerebral edema in four brain regions during the first 0 minutes of recirculation following severe forebrain ischemia in rats. Materials and Methods Transient forebrain ischemia was induced in male Wistar rats weighing g (M0llegaard's Breeding Center, Copenhagen, Denmark). The rats were fasted overnight but had free access to tap water.

2 Downloaded from by on January, Stroke Vol 20, No, November 99 The operative procedure has been described in detail. 6 In short, isoflurane (Abbott Laboratories Ltd., Kent, U.K.) in a mixture of nitrous oxide/ oxygen (70%/30%) was used to induce (3%) and maintain (-.5%) anesthesia. The rats were endotracheally intubated and connected to an animal ventilator. The common carotid arteries were exposed via a ventral midline incision, and a silicone catheter was inserted into the right jugular vein to allow withdrawal of blood for induction of hypotension during ischemia. The tail artery was cannulated for continuous blood pressure monitoring and blood sampling. The electroencephalogram (EEG) was monitored by placing needle electrodes bilaterally in the temporal muscles via a small skin incision. The rats were then heparinized (50IU) and allowed a 30-minute steady-state period. Skull and rectal temperatures were measured with a thermocouple (A-05, Ellab, Copenhagen, Denmark) and a standard mercury thermometer, respectively, and were maintained at close to 37 C by a distant heating lamp (55 W). Blood pressure was maintained at mm Hg. Pao 2 and Paco 2 were maintained at >90 mm Hg and 35-0 mm Hg, respectively, by adjusting ventilation. At that Paco 2, plasma ph was kept between 7.35 and 7.5. Plasma glucose levels were measured with a Beckman Glucose Analyzer 2 (Beckman Instruments Inc., Los Angeles, California). Isoflurane administration was discontinued, and minute later near-complete forebrain ischemia was induced by bilateral clamping of the common carotid arteries, combined with exsanguination to achieve a blood pressure of approximately 50 mm Hg. Under these conditions EEG activity rapidly became isoelectric. 6 Ischemia was maintained for 5, 5, or 30 minutes, and the clamps were then removed and the withdrawn blood was reinfused. Fifty-six rats were subjected to 5 minutes of ischemia and assigned to one of seven groups (n= in each group). Rats in one group were killed immediately upon termination of ischemia (i.e., the clamps were not removed and no blood was reinfused), and rats in the other six groups were decapitated after 5, 5, 30, 60, 90, or 0 minutes of recirculation. Sixteen additional rats were subjected to 5 or 30 minutes of ischemia and 30 or 60 minutes of recirculation (n= in each group). The same procedures were followed in nine control rats, except that ischemia was not induced. Following decapitation of the rats, the brains were rapidly transferred to a closed chamber with a relative humidity of >90% at ambient temperature. Samples of approximately 25 mg from the frontoparietal cortex, the caudoputamen, the dorsal hippocampus, and the mesencephalon were dissected bilaterally on an ice-cold plate immediately before being submerged in a density gradient of polyvinyl pyrrolidone-coated silica particles (Percoll, Pharmacia AB, Uppsala, Sweden) as described by Tengvar et al. 7 - Glass spheres of known densities (.0300,.0350,.000,.050, and.0500 g/cm 3, Scientific Glass, Bloomfield, New York) were introduced into the column. The position of these spheres in the gradient reflects the linearity of the column, and when equidistantly spaced, these spheres provide a simple and reliable calibration for the continuous gradient. The tissue samples were introduced into the gradient, left hemisphere samples first, and their position was measured 3 minutes later, thus allowing determination of their specific gravity (SG) in relation to the positions of the glass spheres. Thereafter, the same procedure was undertaken for the right hemisphere samples. The data were subjected to one-way analysis of variance. Significance was assumed when/?<0.05. When significance was reached, Dunnet's test was used to determine which groups differed. Results We found all rats to be normoxic, normocapnic, normotensive, normoglycemic, and without major plasma acid-base disturbances before the induction of ischemia (Table ) and during recirculation. The temperature in the rectum and on the skull remained at 37 C. We found the density gradients to be linear. SG of corresponding tissue samples from the two sides of the brain did not differ in any group, so we calculated the mean SG for brain regions. Data from the 56 rats subjected to 5 minutes of ischemia and from the controls are presented in Figure. In the rats with 0 minutes of recirculation, no significant changes in SG were found in the frontoparietal cortex, the caudoputamen, or the hippocampus compared with control. In the frontoparietal cortex, SG decreased significantly after 5 minutes of ischemia followed by 5 minutes of recirculation compared with control (Figure, top; a nonsignificant change in SG was present after 5 minutes of recirculation). After 30 minutes of recirculation, SG had decreased further and reached the lowest value observed. After 60 and 90 minutes of recirculation, SG was still lower than control but was gradually returning to normal. After 0 minutes of recirculation, no significant difference in SG was observed compared with control. In the caudoputamen and hippocampus, the decreases in SG were similar to that in the frontoparietal cortex, with the lowest value observed after 30 minutes of recirculation. In the caudoputamen and hippocampus, SG then gradually increased, normalizing after 0 minutes of recirculation (Figure, top). The maximal change in SG was similar in these three brain regions. As an example, after 30 minutes of recirculation, the decrease in SG of the frontoparietal cortex was g/cm 3. Five minutes of ischemia followed by 30 or 60 minutes of recirculation did not change SG significantly (Figure 2). Thirty minutes of ischemia followed by 30 minutes of recirculation decreased SG in the frontoparietal cortex, the caudoputamen, and

3 Mellergard et al Postischemic Brain Edema 567 TABLE. Physiologic Parameters Before Ischemia in Rats Parameter Group Control Ischemia 5 min 0 min recirculation 5 min recirculation 5 min recirculation 30 min recirculation 60 min recirculation 90 min recirculation 20 min recirculation 5 min 30 min recirculation 60 min recirculation 30 min 30 min recirculation 60 min recirculation n 9 Rectal temperature fc) 37.2± ± ± ± ± ± ± ± ± ± ± ±0.2 Mean arterial blood pressure (mm Hg) 5±0 ±0 20±20 20±9 2±2 7±2 9±5 ± ±0 Pao 2 (mm Hg) 0±5 05±0 05 ±0 0±2 2± 06±7 3± 0±0 6±2 3±7 ±7 0±0 Paco 2 (mm Hg) ± 35 ±2 37±2 35±0 3± 35 ±2 Arterial ph 7.39± ± ±0.0 7.± ± ± ± ± ±0.0 7.± ± ±0.0 Blood glucose concentration (mmol/) 6.9±.0 7.7±2.2 7.±.0 7.3±.5.0±. 9.±3.0.6±2.2.9±0..9±.2.0± ±0.7.±.2 Data are mean±sd. Downloaded from by on January, 209 the hippocampus (Figure 3, left); the change was similar in magnitude to that caused by 5 minutes of ischemia, but the SG decrease remained unchanged or was slightly more accentuated after 60 minutes of recirculation. Thus, there was an indication that edema following 30 minutes of ischemia is more sustained than that following 5 minutes of ischemia. In the mesencephalon, no significant change in SG was found during recirculation following 5 minutes of ischemia (Figure, bottom). Following 30 minutes of ischemia, there was a tendency toward an increase in SG during 60 minutes of recirculation (Figure 3, right), but this did not reach significance. Discussion Linear density gradients, as first described by Nelson et al, 9 have proved to be reliable for measuring SG and brain edema.",,20,2 Changes in SG are assumed to primarily reflect changes in the water content of brain tissue, 22 and a decrease in SG is usually interpreted as an increase in the amount of water. 222 Clearly, changes in blood content are important for the SG of a tissue 2 ' 23 and should be taken into account during studies of ischemia and reperfusion, during which drastic changes in cerebral blood flow and blood volume occur. However, the postischemic changes in SG we observed are partly inverse to previously mapped postischemic changes in cerebral blood flow, 2 that is, SG decreased during the period of known postischemic /jype/perfusion. Since blood always has a higher density than brain tissue, 2-23 the changes in SG we observed cannot be explained by changes in blood content. It has previously been shown that in the ischemic model we used, blood flow in the forebrain (including the frontoparietal cortex, caudoputamen, and hippocampus) decreases to <5% of normal. 2 In our study, 5 minutes of severe forebrain ischemia resulted in no observable regional edema during recirculation. When the ischemic period was prolonged to 5 minutes, edema developed and was of similar severity in the frontoparietal cortex, the caudoputamen, and the hippocampus. Edema developed gradually and reached a maximum after 30 minutes of recirculation. The edema then gradually resolved in these three regions; SG was normalized after 0 minutes of recirculation. Following 30 minutes of ischemia, edema was more sustained in time but was not more prominent in magnitude than after a 5-minute ischemic insult. Thus, 5 minutes of forebrain ischemia seems to be adequate for studying early reversible postischemic brain edema. In the mesencephalon, where blood flow is approximately 50% of normal during this ischemic insult, 2 no changes in brain water content were noted after 5 minutes of ischemia. However, following 30 minutes of ischemia, SG increased progressively after 30 and 60 minutes of recirculation. The most likely explanation for this is an increased relative blood content Postischemic hypoperfusion may cause a relative redistribution of blood to brain regions less severely affected by the ischemic insult, such as the mesencephalon. All extra fluid in the brain is ultimately derived from blood; hence, if there is no blood flow, there will be no increase in tissue water content. In fact, in our study SG did not change in any forebrain region in rats decapitated immediately following

4 56 Stroke Vol 20, No, November 99 Specific gravity (g/crra Specific gravity (g/cm,03 -i,0 -i,02,0 -,00,039 - I,03,037,036,035,03 Ctrl 90 0,03 Ctrl 30 Recirculauon (min) Downloaded from by on January, 209 FIGURE 2. Bar graph of mean ± SEM specific gravity in four brain regions of control rats (ctrl, n=9) and of rats subjected to 5 minutes of ischemia followed by 30 or 60 minutes of recirculation (n= in each group). Values are average of both hemispheres. CP, caudoputamen; CX, frontoparietal cortex; HIP, hippocampus; MES, mesencephalon. No significant changes in specific gravity were observed.,03 Ctrl 90 0 FIGURE. Bar graphs of mean±sem specific gravity in four brain regions of control rats (Ctrl, n=9) and of rats subjected to 5 minutes of ischemia followed by 0, 5, 5, 30, 60, 90, or 0 minutes of recirculation (n= in each group). Values are average of both hemispheres. Top: CP, caudoputamen; CX, frontoparietal cortex; HIP, hippocampus. Similar decreases in specific gravity were seen in all three regions, with maximal change noted after 30 minutes of recirculation. p<0.05, p<0.0 different from control. Bottom: Mesencephalon. No significant changes in specific gravity were observed during recirculation. ischemia, which confirms that the ischemic model we used produces very severe forebrain ischemia.6-2 Our findings are complementary to previous results. In our laboratory, Warner et al25 showed that forebrain ischemia in rats with comparably severe insults led to biphasic edema. The first sample was taken after 90 minutes of recirculation, and these authors then observed an edema that persisted for up to 6 hours of recirculation before complete normalization; after 2 hours, edema recurred. We have shown that the initial edema noted by Warner et al25 is probably most pronounced after 30 minutes of recirculation and that after 90 minutes of recovery this edema is already partially resolved. Available information on edema development during the early recirculation period following an ischemic insult is sparse. Avery and coworkers26 exposed gerbils, animals with an incomplete circle of Willis, to very profound ischemia (60 minutes of bilateral carotid artery occlusion), which caused 9% to die after <2 hours of recirculation. The authors showed a rapid decrease of SG after 5 minutes of recirculation in all regions investigated (including the cortex, corpus striatum, thalamus, and hippocampus).26 The change in SG progressed, being maximal after -2 hours of recirculation. After 3 hours, SG returned to normal in all regions except the hippocampus. Using another model of very severe ischemia in rats (four-vessel occlusion for 30 minutes), Petito et al27 showed a transient % increase in total brain water content after 5 and 30 minutes of recirculation. Although still elevated, brain water content was returning to normal after 60 minutes of recovery. These authors, however, did not detect any regional edema in the striatum, middle cerebral artery cortex, or hippocampus at any postischemic interval.27 Finally, Todd and coworkers,2'29 using

5 Mellerg&rd et al Postischemic Brain Edema FIGURE 3.,02,0,00,039,03,037,036,035,03 - ip i! ct gravity (g/cni 3) c 0 c Specific gravity (g/cm 3 ),03 -, U H 22 I 3C) Recirculau t»n (mm) II II,0,00-,039,03,037,036,035,03 - T I I HI Bar graphs 569 of mean ±SEMspecific gravity in four brain regions of control rats (ctrl, n=9j and of rats subjected to 30 minutes of ischemia followed by 30 or 60 minutes of recirculation (n= in each group). Values are average of both hemispheres. Left: CP, caudoputamen; CX, frontoparietal cortex; HIP, hippocampus. Similar decreases in specific gravity are seen in all three regions, of magnitude similar to that seen after 5 minutes of ischemia (Figure ) but sustained to at least 60 minutes of recirculation. p<0.05, p<0.0 different from control. Right: Mesencephalon. No significant changes in specific gravity were observed, but there was tendency to progressive increase in specific gravity for up to 60 minutes of recirculation. HHH MM 63 Downloaded from by on January, 209 four-vessel occlusion in rats for 5, 30, or 60 minutes, suggested a decrease in mean cortical SG after 30 minutes of recirculation followed by normalization after 0 minutes of recirculation. Conditions for the development of brain edema clearly exist during the early recirculation period following severe ischemia. Experiments have shown that severe ischemia is accompanied by a rapid decrease in extracellular Na +, Cl~, and Ca2+ contents, concomitant with an increase in cortical impedance.30"3 This is generally interpreted as reflecting a cellular influx of Na+ and Cl~ together with osmotically obliged water, leading among other things to a decrease in extracellular space. Cerebral ischemia is also known to be accompanied by a significant increase in tissue osmolality Since no (or very little) loss of osmoles into the circulating blood takes place during ischemia, an osmotic pressure gradient develops between the blood and the brain. It is also generally accepted that transient incomplete cerebral ischemia is followed by a "hyper-hypo-perfusion" sequence.2 When recirculation begins, this early hyperperfusion together with the osmotic pressure gradient may thus induce a rapid shift of water from the blood to the brain, which promotes the development of edema. However, the nature of the extra osmoles remains to be shown.37 Furthermore, there must be clarification of the forces by which water is translocated from the blood to the extracellular fluid (ECF), that is, how the electrolyte composition of the ECF is altered during the immediate postischemic period. clrl 30 KeciauldUon (nun) 60 Since brain edema following cerebral ischemia has some specific properties, the term ischemic edema has been proposed to replace both of the more commonly used terms cytotoxic/cellular edema and vasogenic edema.2 Ischemic edema is initially primarily cytotoxic, whereas vasogenic edema, with leakage of high-molecular-weight indicators of bloodbrain barrier dysfunction, is a late phenomenon that develops hours or days after the early edema.2-3-0'3-39 The early transient edema shown in our study is probably of the so-called cytotoxic type, that is, it primarily reflects the intracellular accumulation of water. In summary, following 5 minutes of severe forebrain ischemia, no regional edema was observed during the early recirculation period. Fifteen minutes of ischemia was followed by transient brain edema in the frontoparietal cortex, the caudoputamen, and the hippocampus. This edema reached a maximum after approximately 30 minutes of recirculation and normalized after 0 minutes of recirculation. Following 30 minutes of ischemia, the development of edema was similar, albeit more sustained in time. It cannot be excluded that this early postischemic edema, although resolved within some hours, has an impact on metabolically perturbed cells during recirculation and thereby on the final outcome of the ischemic insult. Whether of pathologic significance or not, this edema is one parameter to take into account when considering the postischemic neurochemical events that contribute to ischemic cell necrosis.

6 Downloaded from by on January, Stroke Vol 20, No, November 99 Acknowledgment The skilled technical assistance by Lena Evergren is gratefully acknowledged. References. Klatzo I: Neuropathological aspects of brain edema. / Neuropathol Exp Neurol 967;26:- 2. Katzman R, Clasen R, Klatzo I, Meyer JS, Pappius HM, Waltz AG: Report of the Joint Committee for Stroke Resources. IV. Brain edema in stroke. Stroke 977;: Ito U, Ohno K, Nakamura R, Suganuma F, Inaba Y: Brain edema during ischemia and after restoration of blood flow: Measurement of water, sodium, potassium content and plasma permeability. Stroke 979;0: Iannotti F, Hoff J: Ischemic brain edema with and without reperfusion: An experimental study in gerbils. Stroke 93; : Hossman K-A: Treatment of experimental cerebral ischemia. J Cereb Blood Flow Metab 92;2: Hossman K-A: Postischemic resuscitation of the brain: Selective vulnerability versus global resistance. Prog Brain Res 95;63: van Harreveld A: A mechanism for fluid shifts specific for the central nervous system. Top Probl Psychiatry Neurol 970;0: Rothman C, Olney J: Glutamate and the pathophysiology of hypoxic/ischemic brain damage. Ann Neurol 96;9:05 9. Schuier FJ, Hossman K-A: Experimental brain infarcts in cats: II. Ischemic brain edema. Stroke 90;ll: Uyama O, Okamura N, Yanase M, Narita M, Kawabata K, Sugita M: Quantitative evaluation of vascular permeability in the gerbil brain after transient ischemia using Evans blue fluorescence. J Cereb Blood Flow Metab 9;:22-2. Kuroiwa T, Cahn R, Juhler M, Goping G, Campbell G, Klatzo I: Role of extracellular proteins in the dynamics of vasogenic brain edema. Ada Neuropathol (Berl) 95;66:3-2. Baethman A: Pathophysiological and pathochemical aspects of cerebral edema. Neurosurg Rev 97;l: Kogure K, Busto R, Scheinberg P: The role of hydrostatic pressure in ischemic brain edema. Ann Neurol 9;9: Chan PH, Fishman RA: Brain edema, in Lajtha A (ed): Handbook of Neurochemistry. New York, Plenum Publishing Corp, 95, vol 0, pp Siesjo BK: Acidosis and ischemic brain damage. Neurochem Pathol 9;9:3-6. Smith M-L, Auer RN, Siesjo BK: The density and distribution of ischemic brain injury in the rat following 2-0 min of forebrain ischemia. Ada Neuropathol (Berl) 9;6: Tengvar C, Forssen M, Hultstrom D, Pertoft H, Petersson A: Measurement of edema in the nervous system. Use of Percoll density gradients for determination of specific gravity in cerebral cortex and white matter under normal conditions and in experimental cytotoxic brain edema. Ada Neuropathol (Berl) 92;57:3-50. Tengvar C, Hultstrom D, Olsson Y: An improved Percoll density gradient for measurements of experimental brain edema. Ada Neuropathol (Berl) 93;6: Nelson SR, Mantz M-L, Maxwell JA: Use of specific gravity in the measurement of cerebral edema. /Appl Physiol 97; 30: Marmarou A, Tanaka K, Shulman K: An improved gravimetric measure of cerebral edema. J Neurosurg 92; 56: Shigeno T, Brock M, Shigeno S, Fritschka E, Cervos- Navaro J: The determination of brain water content: Microgravimetry versus dry-weighing method. J Neurosurg 92; 57: Bothe H-W, Bodsche W, Hossman K-A: Relationship between specific gravity, water content, and serum protein extravasation in various types of vasogenic brain edema. Ada Neuropathol (Berl) 9;6: Shigeno T, Fritschka E, Shigeno S, Schramm J, Cervos- Navarro J, Brock M: Cerebral edema following experimental subarachnoid hemorrhage. / Cereb Blood Flow Metab 9; l(suppl ):S55-S Kagstrom E, Smith M-L, Siesjo BK: Recirculation in the rat brain following incomplete ischemia. / Cereb Blood Flow Metab 93;3: Warner DS, Smith M-L, Siesjo BK: Ischemia in normo- and hyperglycemic rats: Effects on brain water and electrolytes. Stroke 97;: Avery S, Crockard A, Russel RR: Evolution and resolution of edema following severe temporary cerebral ischemia in the gerbil. J Psychol Neurol Neurosurg 9;7: Petito CK, Pulsinelli WA, Jacobson G, Plum F: Edema and vascular permeability in cerebral ischemia: Comparison between ischemic neuronal damage and infarction. J Neuropathol Exp Neurol 92;: Todd NV, Picozzi P, Crockard HA, Russell RR: Reperfusion after cerebral ischemia: Influence of duration of ischemia. Stroke 96;7: Todd NV, Picozzi P, Crockard HA, Russell RR: Duration of ischemia influences the development and resolution of ischemic brain edema. Stroke 96;7: Hossman K-A, Sakaki S, Zimmerman V: Cation activities in reversible ischemia of the cat brain. Stroke 977;:77-3. Hansen AJ, Zeuthen T: Extracellular ion concentrations during spreading depression and ischemia in the rat brain cortex. Ada Physiol Scand 9;3: Harris RJ, Symon L, Branston NM, Bayhan M: Changes in extracellular calcium activity in cerebral ischaemia. / Cereb Blood Flow Metab 9;l: Hansen AJ: Effect of anoxia on ion distribution in the brain. Physiol Rev 95;65:0-3. von Hanwehr R, Smith M-L, Siesjo BK: Extra- and intracellular ph during near-complete forebrain ischemia in the rat. / Neurochem 96;6: Hossman K-A, Tagaki S: Osmolality of brain in cerebral ischemia. Exp Neurol 976;5: Hatashita S, Hoff JT, Slamat SM: Ischemic brain edema and the osmotic gradient between blood and brain. J Cereb Blood Flow Metab 9;: Jennings RB: Myocardial ischemia revisited. The osmole load, membrane damage, and reperfusion. / Mol Cell Cardiol 96;: O'Brien MD, Jordan MM, Waltz AG: Ischemic cerebral edema and the blood-brain barrier: Distribution of pertechnate, albumin, sodium, and antipyrine in brains of cats after occlusion of the middle cerebral artery. Arch Neurol 97; 30: Suzuki R, Yamaguchi T, Kirina T, Orzi F, Klatzo I: The effect of 5-minute ischemia in Mongolian gerbils: I. Bloodbrain barrier, cerebral blood flow, and local cerebral glucose utilization changes. Ada Neuropathol (Berl) 93;57:3-50 KEY WORDS brain edema cerebral ischemia rats