Treatment of experimental focal cerebral ischemia with mannitol. Assessment by intracellular brain ph, cortical blood flow, and electroencephalography

Transcription

1 J Neurosurg 66: , 1987 Treatment of experimental focal cerebral ischemia with mannitol Assessment by intracellular brain ph, cortical blood flow, and electroencephalography FREDlUC B. MEYER, M.D., ROBERT E. ANDERSON, B.S., THORALF M. SUNDT, JR., M.D., AND TONY L. YAKSH, PH.D. Departments of Neurosurgery and Pharmacology, Mayo Clinic, Rochester, Minnesota v" Intracellular brain ph, cortical blood flow (CBF), and electrocorticograms were recorded in regions of severe and moderate ischemia in l0 control rabbits and 10 rabbits given mannitol, 1 gm/kg, after occlusion of a major branch of the middle cerebral artery. Pooling the data from all 20 animals, preocclusion CBF was 46.4 _+ 3.6 ml/100 gin/rain and intracellular brain ph was (means + standard error of the means). Although mannitol administration mildly improved CBF in regions of severe ischemia, this increase was not sufficient to prevent metabolic deterioration as assessed by brain ph. However, in regions of moderate ischemia, CBF improved significantly with mannitol and the gradual decline in brain ph observed in control animals was prevented. For example, in the treated moderate ischemia sites 4-hour postocclusion CBF and ph values were 31.8 ml/100 gm/min and , respectively, as compared to control values of 14.3 ml/ 100 gin/rain and 6.75 _ These results suggest that mannitol may be of benefit in stabilizing regions of moderate, but not severe, ischemia after vessel occlusion. KEY WORDS 9 cerebral ischemia 9 brain ph 9 cortical blood flow 9 mannitol 9 electroencephalography 9 rabbit I N a model of transient global ischemia, Ames and colleagues I'1~ demonstrated that after flow restoration there were regions of impaired vascular filling. They postulated that this "no reflow phenomenon" contributed to neuronal death following transient global ischemia. Crowell and Olsson 7 later demonstrated similar microcirculatory obstruction in focal ischemia. Litfie, et al, 2~ showed that this obstruction was related in part to capillary compression by perivascular glial edema. On the basis of its hyperosmolar characteristics, mannitol has been advocated as an adjunctive treatment for acute focal ischemia. Little and Peerless, et al., 23 demonstrated that mannitol improved neurological outcome and infarct size after middle cerebral artery (MCA) occlusion in the cat. Postulated mechanisms included either a reduction in edema or a decrease in blood viscosity, 6,8 each of which would facilitate collateral flow after vessel occlusion. However, other inves- tigators have shown that, although oncotic agents such as albumin improved microcirculatory flow in cats, there were only transient improvements in flow without a reduction in infarct size in squirrel monkeys. 36,37 This may reflect differences in potential collateral flow between carnivores and primates. 13 Recently, the rabbit has been investigated as a model of focal cerebral ischemia. 21 The rabbit brain is supplied by the internal carotid artery without contribution from the external carotid artery as in the dog. 27 There is no rete mirabilis as in the cat. 28 Occlusion of the MCA reliably produces cortical infarction similar to that observed in primates. 21 Therefore, assessment of pharmacological regimens in the rabbit may be more analogous to the true clinical setting in which collateral flow is limited after major vessel occlusion. This current experiment evaluates cortical blood flow (CBF), intracellular brain ph, and electroencephalography (EEG) in rabbits receiving mannitol after MCA occlusion. J. Neurosurg. / Volume 66/January,

2 F. B. Meyer, et al. Materials and Methods Animal Preparation Twenty-three New Zealand White rabbits, weighing between 3.5 and 4.5 kg each, were anesthetized with 4% inspired halothane, operated on under 2% halothane, and studied under 0.5% halothane. All animals were given 0.15 mg/kg pancuronium bromide to eliminate respiratory effort and were maintained throughout the experiment on a Harvard respirator. Catheters were inserted into the femoral artery and vein for monitoring blood pressure and arterial blood gases, and for the administration of drugs. A tracheostomy was performed and a PE-50 cannula was inserted into the ligated external carotid artery for delivery of umbelliferone into the internal carotid artery. The MCA was exposed through a retro-orbital craniectomy similar to that used in a cat. z~'35 The skin, subcutaneous tissue, and muscle were dissected from the supraorbital ridge and parietal bone. Enucleation of the orbit facilitated removal of the supraorbital ridge with rongeurs. The frontotemporoparietal craniectomy extended medially to the sagittal sinus, laterally to the zygomatic arch, and rostrally to the optic canal. The craniectomy was performed with a high-speed air drill and with the aid of an operating microscope so that no pressure was exerted against the underlying cortex. The dura was removed, and a thin sheet of plastic film (Saran Wrap) was placed on the cerebrum with irrigation to prevent surface dehydration and oxygenation. Total blood loss for the operative procedure did not exceed 5 cc. The MCA bifurcation was identified at the anteroinferior margin of the craniectomy between the frontal and temporal lobes, z2 Occlusion of one of these divisions will yield three zones of flow: 1) a zone of severe ischemia; 2) a zone of normal flow supplied by the division left intact; and 3) a region of moderate ischemia between these two zones, partially supplied by the division left intactfl ',22 This border zone of moderate ischemia was located approximately halfway between the occluded and patent branches and could be identified visually by subtle color changes of the cortex under microscopic observation. Suspected border zones were confirmed by measurement of an immediate 60% reduction in postocclusion CBF as compared to preocclusion flows, with only minimal changes in intracellular brain ph. Immediately following occlusion, severely ischemic regions had an 80% reduction in CBF as compared to preocclusion flows, with a dramatic drop in intracellular brain ph. Data Recording After the operative exposure, the animals were moved to an intravital microscope stage. Measurements of two separate sites on the suprasylvian gyrus were taken at a PaCO2 of 20, 40, 60, and 80 torr by altering the amount of inspired CO2 and the respiratory rate. A normal PaCO2-CBF response curve assured that autoregulation was intact within a normal blood pressure range and that the brain had not been injured. 16 Measurements of a normal intracellular brain ph during this PaCO2 response curve were further evidence of a physiologically intact cerebrum. After the PaCO2 response curve was obtained, two measurements of normal CBF and intracelluiar brain ph at a PaCO2 of 40 torr were made prior to occlusion of the MCA with bipolar cautery. Twenty rabbits were equally divided in two groups: 10 animals were treated with mannitol, 1 gm/kg, given as an intravenous bolus after MCA occlusion and 10 formed an untreated control group. Immediately after vessel occlusion and prior to treatment, identification was made of regions of both severe and moderate ischemia. The x and y coordinates were recorded so that these sites could be successively evaluated each hour. Measurements of intracellular brain ph and CBF at all sites were made for 4 hours after MCA occlusion. Results obtained in similar control animals have been reported previously. 21'22 Three rabbits with a sham craniectomy but without MCA occlusion served as time controls, and intracellular brain ph and CBF were assessed after a normal PaCOz response curve during a 4-hour experimental period. Electrocorticograms were recorded in all animals by placement of gold electrodes at the margins of the craniectomy. Arterial blood gases were measured hourly, and blood pressure was continuously recorded. The animals were sacrificed with a lethal injection of potassium chloride. Intracellular Brain ph and CBF Measurements The use of umbelliferone for measuring intracellular brain ph and CBF in vivo has been described elsewhere. 3,31-33 Umbelliferone is a lipid-soluble, ph-sensitive fluorescent indicator that freely diffuses across the blood-brain barrier. It is nontoxic and isolated to the intracellular compartment. Its ionic and molecular forms are fluorophors which have different fluorescent characteristics. 9 The ratio of the ionic and molecular forms at 370 and 340 nm excitation, respectively, is ph-sensitive. It is possible to create a nomogram relating ph to the ratio of the 450 nm fluorescent curves of the indicator at 370 and 340 nm excitation. Cortical blood flow was determined by the washout curve of the molecular form of umbelliferone over a 60-second period, beginning after the arterial spike, using the 1-minute initial slope index. 3 Since the bloodbrain partition coefficient of umbelliferone is 1.0 (RE Anderson, unpublished data), its clearance is a measure of perfusion through the intracellular compartment and in effect is focal CBF. 2'21'~3 It is thought to be dependent on diffusion from capillary to glial cell to neuron. 2'3'-33 Since this technique measures flow only in the outer 50 um of the cortex, it eliminates the "look through" phenomenon associated with measurements of blood flow obtained with xenon- 133 during ischemia. ~ 110 J. Neurosurg. / Volume 66/January, 1987

3 Treatment of experimental cerebral ischemia with mannitol Instrumentation for Measuring ph The microspectrofluorometer used in this experiment was equipped with optics for bright field illumination which permitted: low-intensity excitation energy; highefficiency recording from an 80-urn avascular area of cortex; a high-speed filter wheel with four interference filters that allowed synchronization of the emission recording system at 450 nm with the 370- and 340- nm excitation bands; and an emission recording system consisting of a high-sensitivity thermoelectrically cooled photomultiplier tube attached to a high-efficiency grating monochromator. Fluorescent emission signals were amplified by a cascaded electrometer amplifier and directed into a photodemodulator synchronized with the filter wheel. The fluorescence washout curves were recorded on a dual strip-chart recorder. The instrumentation is fully detailed in previous reports. 32,33 Statistical Analysis Ten sites of severe ischemia in the control animals were compared with 10 sites of severe ischemia in the animals treated with mannitol for each hourly measurement. Likewise, 10 sites of moderate ischemia in the control animals were compared to 10 sites of moderate ischemia in mannitol-treated animals. The statistical significance was calculated by a paired t-test at each hourly measurement. The deviation from mean value is expressed as standard error of the mean. Results Stability of the Preparation In the three rabbits without MCA occlusion there was no significant decline (paired t-test) in either intracellular brain ph or CBF for 4 hours after a normal PaCO2 response curve. The initial intracellular brain ph was and at 4 hours it was Initial CBF was ml/100 gm/min and at 4 hours it was 43.0 _+ 4.0 ml/100 gm/min. Intracellular Brain ph and CBF in Severe Ischemia Pooling data from all 20 animals with MCA occlusion, preocclusion CBF was ml/100 gm/min and intracellular brain ph was 7.01 _ 0.04 (Table 1 and Fig. 1). In the 10 control animals, CBF at the severe ischemia sites immediately after occlusion was ml/100 gm/min. Within 10 minutes, intracellular brain ph was 6.65 _ These sites continued to demonstrate reductions in both intracellular brain ph and CBF over the ensuing 4 hours. Four hours after occlusion, the intracellular brain ph was 6.10 _ 0.10 and CBF was 5.0 _ 1.5 ml/100 gm/min. Compared to the preocclusion values, the reduction in both CBF and intracellular brain ph at each measured interval in the control animals was statistically significant (p < 0.001). In the 10 mannitol-treated animals, immediate postocclusion CBF at the severe ischemia sites was 13.2 _ 1.6 ml/100 gm/min and brain ph was 6.58 _ TABLE 1 Measurements in severe ischemia sites with and without treatment * Cortical Blood Flow Time of Intracellular Brain ph (ml/100 gm/min) Measurement (min) Control Treated Control Treated preocclusion _ _ _ _+ 2.5 postocclusion _ _ _ _ _+ 4.4 * Data are means _+ standard error of the means for 10 rabbits in the control (untreated) group and 10 in the group treated with mannitol. 6~ CBF 40 ml/ 100 g/ rain 20 ~... ~-... _}....~ 0 MCA,, ' ' Occlus on Mannitol,, IV bolus b-----_.,, \ 6.8 Intr a- i~71 - cellular 6.6 brain ph M ]I" Control, occlusion T ~ ~ II Control, no OCCIUsLon J I 5.a I I I//I I I I Time (min) FIG. 1. Graph of intracellular brain ph and cortical blood flow (CBF) in severely ischemic sites in 10 control animals (solid lines) and severely ischemic sites in 10 animals treated with an intravenous (IV) bolus of mannitol (dashed lines). Pooling data from all 20 animals, CBF obtained before middle cerebral artery (MCA) occlusion was 46.4 _+ 3.6 ml/100 gin/ rain and intracellular brain ph was _ Immediately postocclusion, CBF was 12.0 _+ 2.5 ml/100 gin/rain in the control group and 13.2 _+ 1.6 ml/100 gin/rain in the treated group, and brain ph was 6.65 _ in the control group and in the treated group. The control group demonstrated a progressive decline in CBF and brain ph over the ensuing 4 hours. Although mannitol treatment stabilized CBF, this improvement was not sufficient to significantly alter intracellular brain ph. Three animals without MCA occlusion (dotted lines) served as time controls, Results at these sites prior to infusion of mannitol compared well with those in the control group. After the immediate postocclusion measurements, mannitol (1 gm/kg) was given intravenously. One hour after occlusion (45 minutes after treatment) CBF was 15.2 J. Neurosurg. / Volume 66/January,

4 F. B. Meyer, et al. TABLE 2 Measurements of systemic parameters in control and treated groups Parameter Group Preocclusion mean arterial blood pres- control sure (mm Hg) treated _+ 4.0 PaCO2 (torr) control treated PaO2 (torr) control _+ 7.8 treated arterial ph control treated Time Postoeclusion 15 Min 60 Min 120 Min 180 Min 240 Min _ _ _ _ TABLE 3 Measurements in moderate ischemia sites with and without treatment* Time of Cortical Blood Flow Measure- Intracellular Brain ph (ml/100 gm/min) ment (min) Control Treated Control Treated preocclusion _ _ _ _ postocclusion _ _ _ 5.2 * Data are means _+ standard error of the means for 10 rabbits in the control (untreated) group and l0 in the group treated with mannitol. 3.1 ml/100 gm/min as compared to 9.0 _+ 1.5 ml/100 gm/min in the control group, and at 4 hours it was 16.4 _+ 4.4 ml/100 gm/min. This improvement in CBF was statistically significant at each hourly measurement as compared to the controls (p < 0.01). Despite this improvement in microcirculatory flow, there were no significant changes in intracellular brain ph. Four hours after occlusion, intracellular brain ph was 6.23 _ in the treated animals as compared to 6.10 _ in the control group. There is some suggestion that mannitol stabilized brain ph from 2 to 4 hours, albeit at a significant level of acidosis. There was no significant difference in systemic parameters between the control and treated groups (Table 2). Intracellular Brain ph and CBF in Moderate Ischemia In the 10 control animals, initial postocclusion CBF at the moderate ischemia sites was ml/100 gm/min or approximately 40% of preocclusion flows. This was associated with an intracellular brain ph of 6.95 _ While the reduction in flow was significant (p < 0.001), the drop in ph was not. For the first 3 hours, CBF was stable but then fell significantly at the 4th hour to 14.3 _+ 3.5 ml/100 gm/min (p < 0.01). Intracellular brain ph slowly worsened at each hourly measurement (p < 0.05) and at 4 hours it was 6.75 _ (p < 0.01) (Fig. 2 and Table 3). In the 10 mannitol-treated animals, initial postocclusion CBF at the moderate ischemia sites was ml/100 gm/min, approximately a 60% reduction of preocclusion flows. This was associated with an intracellular brain ph of 6.85 _+ 0.04, slightly lower than in the control group. However, both parameters at these sites compared well to those at moderate ischemia sites of the control animals. Mannitol had no effect on CBF at the 1st hour ( ml/100 gm/min versus 19.5 _+ 2.0 ml/100 gm/min in the control group). However, in contrast to the progressive fall in ph observed in the control animals during the subsequent 3 hours, CBF and brain ph remained unexpectedly stable. Thus, at the 4th hour, CBF was ml/100 gm/min in the treated group versus 14.3 _+ 3.5 ml/100 gm/min in the control group (p < 0.001), and brain ph was 6.89 _ versus 6.75 _ (p < 0.01), respectively. Electroencephalographic Correlates Baseline electrocorticograms in rabbits consist of a 6- to 10-Hz rhythm. Since the EEG leads were located at the edges of the craniectomy, they summated electrical activity across both normal and ischemic cortex. Of the l0 control animals, eight demonstrated loss of amplitude or a slow 1- to 2-Hz rhythm immediately after occlusion. In these eight rabbits, the EEG continued to demonstrate progressive loss of amplitude and frequency over the next 4 hours? 3 Of the 10 mannitol-treated animals, seven showed initial EEG changes after occlusion consisting of a slow 1- to 2-Hz rhythm or loss of amplitude. Although none of these seven animals improved to a normal rhythm, four maintained the postocclusion EEG without further deterioration. Discussion Previous evaluation of this model has shown that sites of severe ischemia are evolving infarcts, while moderate ischemia sites appear analogous to ischemic penumbras. 21 Although mannitol improved microcir- 112 J. Neurosurg. / Volume 66/January, 1987

5 Treatment of experimental cerebral ischemia with mannitol culatory flow in severe ischemia sites, this increase was not sufficient to facilitate metabolic recovery as assessed by intracellular brain ph. In severe ischemia sites, evaluation of brain ph after vessel occlusion is a reflection of production of both lactic acid and free fatty acid. The acidosis observed in the first 30 minutes reflects the four- to eightfold increase in lactic acid concentrations from anaerobic metabolism. 34 During ongoing ischemia at flows of approximately 10 ml/100 gm/min or less, there is a flux of calcium into the ischemic neurony Alterations in calcium homeostasis are secondary to loss of adenosine triphosphate for the Na-K pump which causes depolarization of voltage-sensitive calcium channels. This rise in intracellular calcium precipitates several metabolic cascades including the activation of phospholipase A and C, which attack cellular membranes liberating free fatty acids.15'29'4~ Therefore, serial measurements of intracellular brain ph will reflect both lactic acid and free fatty acid concentrations. This acidosis has been shown to enhance glial edema, denature proteins, suppress energy generation, and create an optimal milieu for free radical formation '30,39,4~ Previous analysis of CBF-intracellular brain ph data has suggested that under light halothane anesthesia there is a threshold of ph failure that ranges from 12 to 15 ml/100 gm/min. 21 This would correspond closely with the threshold of ionic failure. 4'5 Therefore, the mild improvement in CBF in severe ischemia sites, although statistically significant when compared to the controls, was not great enough to permit normal energy production. Hence, there was persistent lactic acid formation along with probable continued free fatty acid production. In the moderate ischemia sites, mannitol offered a protective effect by stabilizing intracellular brain ph and CBF at the 4th hour. Prior to this current experiment, it appeared reasonable to conclude that the deterioration in these moderate control sites was related in part to fatigue of collaterals by progressive edema choking off residual flow. 2~ From that perspective, mannitol protected microcirculatory flow to these regions, thereby facilitating normal basal metabolism. It has been postulated that mannitol works through osmolar dehydration. This would reduce cerebral edema and improve microcirculatory patency. Little ~7-=9 demonstrated significant improvement in neurological outcome, neuronal alterations, and capillary obstruction by mannitol in cats. Our current results are not as dramatic as those reported by Little and others. 38'42 The difference may be related to the animal model used. In cats, the propensity of collateral flow after MCA occlusion is significant. Therefore, reduction in edema would facilitate collateral flow. However, in rabbits, collateral flow after MCA occlusion is less. Prevention of microcirculatory obstruction by attenuating cerebral edema would be expected to have a less positive effect. This would explain why the severe ischemia sites continued to deteriorate after the loss of primary perfusion. The adjacent moderately ischemic sites were more proximal CBF 60,0 \_+ roll ljr min 20 MCA 0 - Occ us on P Mannitol ~. 7,1 - V bous... Intra- "... cellular 6.9 brain ph Mannit 6.7 _ -- Control, occlusion... Control, no occlusion 6.6 I I 5"" [/610 I I I Time (min) FIG. 2. Graph of intracellular brain ph and cortical blood flow (CBF) in moderately ischemic sites in 10 control animals (solid lines) and moderately ischemic sites in 10 animals treated with an intravenous (IV) bolus of mannitol (dashed lines). After occlusion of the middle cerebral artery (MCA), CBF was 21.0 _+ 2.0 ml/100 gin/rain in the control group and ml/100 gm/min in the treated group while brain ph was 6.95 _ 0.05 and , respectively. With infusion of mannitol, there was a progressive improvement in CBF: flow at the 4th hour postocclusion was ml/ 100 gm/min as compared to 14.3 _+ 3.5 ml/100 gm/min in the control group (p < 0.001). This resulted in gradual improvement in intracellular brain ph as opposed to the gradual decline observed in controls: at the 4th hour postocclusion, brain ph was in the treated group as compared to _ 0.06 in the control group (p < 0.01). Three animals without MCA occlusion (dotted lines) served as time controls. to the source of collateral flow and therefore would demonstrate greater beneficial effects. These results are compatible with early studies by Sundt, eta/., 36"37 who treated both cats and monkeys with osmolar dehydrating agents like albumin. In cats, the CBF, infarct size, and neurological outcome were improved. However, albumin did not influence infarct size or outcome in primates. Monkeys, like rabbits, have limited leptomeningeal collateral flow after MCA occlusion. Therefore, attenuating edema-induced microcirculatory obstruction would have less beneficial effects. One.other possible explanation is that our rabbit model involved an open-skull preparation as compared to the closed preparation used by Little. The decompression that would occur from a craniectomy might have masked more positive results. It is important to briefly examine the importance of cerebral edema. Ischemic edema is a combination of both early cytotoxic and late vasogenic edema. 14 Cytotoxic edema can be seen within 30 minutes of vessel occlusion under light microscopy and first occurs in perivascular glia. However, Little, et al.,2~ demonstrated that neuronal injury precedes this glial edema, and that this glial edema then caused microcirculatory obstruction. Therefore, in core regions of ischemia, glial edema J. Neurosurg. / Volume 66/January,

6 F. B. Meyer, et al. is a minor factor in infarct evolution. However, in regions of marginal flow where basal metabolism is temporarily intact, progressive glial edema will be a critical factor. This current study supports the contention that mannitol may be of benefit in moderate ischemia regions. The late vasogenic edema is secondary to ischemic endothelial damage with increased capillary permeability. Mannitol would be less beneficial in treating late vasogenic edema since the presence of ischemic endothelial damage would imply the coexistence of irreversible neuronal damage. Furthermore, mannitol could have detrimental effects by leaking across the blood-brain barrier with a rebound increase in edema. Conclusions Although mannitol mildly improved CBF in core regions of ischemia, this elevation was not sufficient to alter metabolism as assessed by intracellular brain ph. However, mannitol was bepeficial in stabilizing zones of moderate ischemia (ischemic penumbras), presumably by reducing cytotoxic edema that resulted in improved CBF. It should be emphasized that while the present study shows only a mild but significant modification by mannitol in contrast to the reports of others, differences in species and the attendant variations in potential collateral flow, the methods of measuring outcome (CBF, brain ph, infarct size, neurological recovery), and the experimental preparation used render comparison difficult. However, the overall positive findings across this range of preparations support the beneficial effects of mannitol and emphasize the importance of edema in the deterioration observed during acute focal cerebral ischemia. References 1. Ames A III, Wright RL, Kowada M, et al: Cerebral ischemia. II. The no-reflow phenomenon. Am J Pathol 52: , Anderson RE, Michenfelder JD, Sundt TM Jr: Brain intraceuular ph, blood flow, and blood-brain barrier differences with barbiturate and halothane anesthesia in cats. Anesthesiology 52: , Anderson RE, Sundt TM Jr: Brain ph in focal cerebral ischemia and the protective effects of barbiturate anesthesia. J Cereb Blood Flow Metab 3: , Astrup J, Siesj6 BK, Symon L: Thresholds in cerebral ischemia -- the ischemic penumbra. Stroke 12: , Astrup J, Symon L, Branston NM, et al: Cortical evoked potential and extracellular K and H + at critical levels of brain ischemia. Stroke 8:51-57, Cantu RC, Ames A III: Experimental prevention of cerebral vasculature obstruction produced by ischemia. J Neurosurg 30:50-54, Crowell RM, Olsson Y: Impaired microvascular filling after focal cerebral ischemia in monkeys. J Neurosurg 36: , Crowell RM, Olsson Y: Impaired microvascular filling after focal cerebral ischemia in the monkey. Modification by treatment. Neurology 22: , Fink OW, Koehler WR: ph effects of fluorescence of umbeuiferone. Anal Chem 42: , Fischer EG, Ames A III, Hedley-White ET, et al: Reassessment of cerebral capillary changes in acute global ischemia and their relationship to the "no-reflow phenomenon." Stroke 8:36-39, Hanson EJ Jr, Anderson RE, Sundt TM Jr: Comparison of S5krypton and 133xenon cerebral blood flow measurements before, during, and following focal, incomplete ischemia in the squirrel monkey. Cite Res 36:18-26, Harris RJ, Symon L, Branston NM, et al: Changes in extracellular calcium activity in cerebral ischemia. J Cereb Blood Flow Metab 2: , Hossmann KA, Schuier FJ: Experimental brain infarcts in cats. I. Pathophysiological observations. Stroke 11: , Klatzo I: Neuropathological aspects of brain edema. J Neuropathol Exp Neurol 26: I- 14, Kuwashima J, Nakamura K, Fujitani B, et al" Relationship between cerebral energy failure and free fatty acid accumulation following prolonged brain ischemia. Jpn J Pharmacol 28: , Lassen NA: The luxury-perfusion syndrome and its possible relation to acute metabolic acidosis localised within the brain. Lancet 2: , Little JR: Modification of acute focal ischemia by treatment with mannitol. Stroke 9:4-9, Little JR: Modification of acute focal ischemia by treatment with mannitol and high-dose dexamethasone. J Neurosurg 49: , Little JR: Treatment of acute focal cerebral ischemia with intermittent, low dose mannitol. Neurosurgery 5: , Little JR, Kerr FWL, Sundt TM Jr: Microcirculatory obstruction in focal cerebral ischemia. Relationship to neuronal alterations. Mayo Clin Proc 50: , Meyer FB, Anderson RE, Sundt TM Jr, et al: Intracellular brain ph, indicator tissue perfusion, electroencephalography, and histology in severe and moderate focal cortical ischemia in the rabbit. J Cereb Blood Flow Metab 6: 71-78, Meyer FB, Anderson RE, Yaksh TL, et al: Effect of nimodipine on intracellular brain ph, cortical blood flow, and EEG in experimental focal cerebral ischemia. J Neurosurg 64: , Peerless S J, Ishikawa R, Hunter IG, et al: Protective effect of Fluosol-DA in acute cerebral ischemia. Stroke 12: , Plum F: What causes infarction in ischemic brain? The Robert Wartenberg Lecture. Neurology 33: , Rehncrona S, Rosrn I, Siesj6 BK: Brain lactic acidosis and ischemic cell damage: I. Biochemistry and neurophysiology. J Cereb Blood Flow Metab 1: , Rehncrona S, Rosrn I, Siesj6 BK: Excessive cellular acidosis: an important mechanism of neuronal damage in the brain? Acta Physiol Scand 110: , Sagawa K, Guyton AC: Pressure-flow relationships in isolated canine cerebral circulation. Am J Physiol 200: , Scremin OU, Sonnenschein RR, Rubinstein EH: Cerebrovascular anatomy and blood flow measurements in the rabbit. J Cereb Blood Flow Metab 2:55-66, Siesj6 BK: Cerebral circulation and metabolism. J Neurosurg 60: , Siesj6 BK, Bendek G, Koide T, et al: Influence of acidosis on lipid peroxidation in brain tissues in vitro. J Cereb Blood Flow Metab 5: , Sundt TM Jr, Anderson RE" Intracellular brain ph and 114 J. Neurosurg. / Volume 66/January, 1987

7 Treatment of experimental cerebral ischemia with mannitol the pathway of a fat soluble ph indicator across the bloodbrain barrier. Brain Res 186: , Sundt TM Jr, Anderson RE: Umbelliferone as an intracellular ph-sensitive fluorescent indicator and bloodbrain barrier probe: instrumentation, calibration, and analysis. J Neurophysioi 414:60-75, Sundt TM Jr, Anderson RE, Van Dyke RA: Brain ph measurements using a diffusible, lipid soluble ph sensitive fluorescent indicator. J Neuroehem 31: , Sundt TM Jr, Michenfelder JD: Focal transient cerebral ischemia in the squirrel monkey. Effect on brain adenosine triphosphate and lactate levels with electrocorticographic and pathologic correlation. Cire Res 30: , Sundt TM Jr, Waltz AG: Experimental cerebral infarction: retro-orbital extradural approach for occluding the middle cerebral artery. Proe Staff Meet Mayo Clin 41: , Sundt TM Jr, Waltz AG: Hemodilution and anticoagulation. Effects on the microvasculature and microcirculation of the cerebral cortex after arterial occlusion. Neurology 17: , Sundt TM Jr, Waltz AG, Sayre GP: Experimental cerebral infarction: modification by treatment with hemodiluting, hemoconcentrating, and dehydrating agents. J Neurosurg 26:46-56, Suzuki J, Yoshimoto T, Kodama N, et al: A new therapeutic method for acute brain infarction: revascularization following the administration of mannitol and perfluoroehemicals -- a preliminary report. Surg Neurol 17: , Welsh FA, Ginsberg MD, Rieder W, et al: Deleterious effect of glucose pretreatment on recovery from diffuse cerebral ischemia in the cat. II. Regional metabolite levels. Stroke 11: , Welsh FA, O'Connor M J, Marcy VR, et al: Factors limiting regeneration of ATP following temporary ischemia in cat brain. Stroke 13: , Yanagihara T, McCall JT: Ionic shift in cerebral ischemia. Life Sei 30: , Yoshimoto T, Sakamoto T, Watanabe T, et al: Experimental cerebral infarction. Part 3: Protective effect of mannitol in thalamic infarction in dogs. Stroke 9: , 1978 Manuscript received June 23, Address reprint requests to: Fredric B. Meyer, M.D., Department of Neurosurgery, Mayo Clinic, Rochester, Minnesota J. Neurosurg. / Volume 66/January,