Cerebral blood flow and edema following carotid occlusion in the gerbil

Neuronal damage following non-lethal but repeated cerebral ischemia in the gerbil

Brief, non-lethal transient forebrain ischemia in the gerbil can injure selectively vulnerable neurons when such ischemia is induced repeatedly. The influence of the number and interval of the ischemic insults on neuronal damage, as well as the time course of damage, following repeated 2-min forebrain ischemia were examined. A single 2-min ischemic insult caused no morphological neuronal damage. A moderate number of hippocampal CA1 neurons were destroyed following two ischemic insults with a 1-h interval, and destruction of almost all CA1 neurons resulted from three or five insults at 1-h intervals. Three and five insults also resulted in moderate to severe damage to the striatum and thalamus, depending on the number of episodes. Although three ischemic insults at 1-h intervals caused severe neuronal damage, this number of insults at 5-min and 4-h intervals caused destruction of relatively few neurons, and no neurons were destroyed at 12-h intervals. Following three ischemic insults at 1-h intervals, damage to the striatum, neocortex, hippocampal CA4 subfield and thalamus was observed at 6-24 h of survival, whereas damage to the hippocampal CA1 subfield appeared at 2-4 days. The results indicate that even a brief non-lethal ischemic insult can produce severe neuronal damage in selectively vulnerable regions when it is induced repeatedly at a certain interval. The severity of neuronal damage was dependent on the number and interval of ischemic episodes.

Rodent models of cerebral ischemia

The use of physiologically regulated, reproducible animal models is crucial to the study of ischemic brain injury--both the mechanisms governing its occurrence and potential therapeutic strategies. Several laboratory rodent species (notably rats and gerbils), which are readily available at relatively low cost, are highly suitable for the investigation of cerebral ischemia and have been widely employed for this purpose. We critically examine and summarize several rodent models of transient global ischemia, resulting in selective neuronal injury within vulnerable brain regions, and focal ischemia, typically giving rise to localized brain infarction. We explore the utility of individual models and emphasize the necessity for meticulous experimental control of those variables that modulate the severity of ischemic brain injury.

MK-801, an excitatory amino acid antagonist, does not improve neurologic outcome following cardiac arrest in cats

The excitatory amino antagonist MK-801 was administered to cats following resuscitation from cardiac arrest to evaluate its effect on neurologic and neuropathologic outcome in a clinically relevant model of complete cerebral ischemia. In 29 cats studied, cardiac arrest (ventricular fibrillation) was maintained for 18 min and resuscitation was successfully performed in 21 cats. Four animals underwent a sham arrest. MK-801 or placebo was administered in a blinded, randomized manner. Beginning at 5 min post resuscitation (PR), MK-801 330 micrograms/kg over 2 min followed by 73 micrograms/kg/h for 10 h or the same volume of placebo was administered. Resuscitated animals remained paralyzed and sedated in an intensive care setting for 24-30 h PR. Neurologic examinations were performed at 2, 4, and 7 days PR by observers blinded to the treatment groups. Seventeen cats were entered into data analysis (nine MK-801-treated and eight placebo-treated). MK-801-treated animals had a significantly greater neurologic deficit score (NDS) rank (0 = normal, 100 = brain death) 2 days PR (mean rank 12.1 vs. 5.6; p = 0.008). This difference is most likely due to ongoing sedative actions of MK-801. There were no significant differences in NDS rank at 4 (10.3, MK-801 vs. 7.5, placebo) and 7 (9.6, MK-801 vs. 8.3, placebo) days PR. There were no significant differences in frontal cortex, hippocampus, occipital cortex, or cerebellar neuropathology between groups. Sham-arrested cats had normal neurologic and neuropathologic evaluations. In the circumstance of complete cerebral ischemia as employed in the current study, MK-801 had no beneficial effect upon neurologic or neuropathologic outcome.

Phenylephrine-induced hypertension reduces ischemia following middle cerebral artery occlusion in rats

We studied the influence of phenylephrine-induced hypertension on the area of ischemia during brief middle cerebral artery occlusion. Rats were anesthetized with 1.2 minimal alveolar concentration (MAC) isoflurane, and the middle cerebral artery was occluded via a subtemporal craniectomy. Immediately thereafter, in one group (n = 9) arterial blood pressure was increased 30-35 mm Hg above the preocclusion level by intravenous infusion of phenylephrine. In a second, control, group (n = 10) there was no manipulation of blood pressure. Local cerebral blood flow was determined autoradiographically 15 minutes after occlusion. The areas (expressed as a percentage of the total coronal cross-sectional area) in which local cerebral blood flow decreased to three ranges (0-6 ml/100 g/min [rapid neuronal death probable], 6-15 ml/100 g/min [delayed neuronal death probable], and 15-23 ml/100 g/min [electrophysiologic dysfunction with prolonged survival probable]) were measured. The areas in which local cerebral blood flow decreased to the two more severely ischemic ranges were smaller in the phenylephrine group than in the control group. For example, in the coronal section in the center of the middle cerebral artery distribution, local cerebral blood flow was 0-6 ml/100 g/min in 6.7 +/- 1.4% of the section in normotensive rats but was in that range in only 1.7 +/- 0.6% of the section during phenylephrine-induced hypertension (p less than 0.05). For the 6-15 ml/100 g/min range, the areas were 6.8 +/- 0.8% and 3.8 +/- 0.7%, respectively (p less than 0.05). For the 15-23 ml/100 g/min range, there were no differences between groups.(ABSTRACT TRUNCATED AT 250 WORDS)

Blood-brain barrier sodium transport limits development of brain edema during partial ischemia in gerbils

Sodium derived from the blood is known to accumulate in brain tissue during the early stages of incomplete ischemia. Our present studies were undertaken to determine the relation between blood-brain barrier sodium transport and the development of ischemic brain edema. Incomplete cerebral ischemia was produced in gerbils by ligation of the left common carotid artery under ether anesthesia. Following recovery from the anesthetic, the gerbis were evaluated for the presence of neurologic symptoms and were divided into symptomatic (n = 77) and asymptomatic (n = 94) groups. Tissue water, sodium, and potassium contents, tissue plasma volume, and brain uptake of 22Na were measured in both groups 1.5, 3, 6, 12, and 24 hours after carotid ligation. There was a progressive accumulation of sodium and water in the ipsilateral cerebral cortex of the symptomatic group compared with either the corresponding contralateral cortex of the same gerbils or with the asymptomatic group. Net changes in brain sodium and potassium concentrations appeared to be the main determinants of fluid accumulation. Brain edema was not due to opening of the blood-brain barrier because the unidirectional transport of 22Na remained low and was even reduced by 35-55% in the ischemic cortex. Nevertheless, this sodium transport activity appeared to be rate-limiting in the development of brain edema during the first 3 hours of ischemia because the rate of sodium accumulation in the tissue was the same as the rate of 22Na transport from the blood to the brain. We conclude that blood-brain barrier sodium transport is an important factor in the formation of ischemic brain edema.

Neuronal damage following repeated brief ischemia in the gerbil

The effect of repetition of brief ischemia, which causes no morphological brain damage when given as a single insult, was studied. Two-minute forebrain ischemia was induced in gerbils singly and 3 or 5 times at 60-min intervals. Although 2-min ischemia induced no neuronal damage, 3 or 5 repeated ischemic insults caused neuronal damage in the selectively vulnerable regions, the severity being dependent on the number of episodes.

Is postischaemic water accumulation related to delayed postischaemic hypoperfusion in rat brain?

The effect of reversible cerebral ischaemia on brain oedema development was studied with a gravimetric method. Cerebral blood flow changes after ischaemia were correlated with alterations in brain specific gravity. Forebrain ischaemia (15 min) was induced in rats by reversible bilateral ligation of both carotid arteries plus induction of controlled hypotension to 50 mm Hg. The specific gravity of different brain structures was determined in a Percoll column up to 24 h after ischaemia. In addition, regional cerebral blood flow was measured by 14C-iodoantipyrine autoradiography. Cerebral ischaemia resulted in reduction of cerebral blood flow to less than 1% of normal in cortical structures and the caudatoputamen. One hour after the end of ischaemia blood flows were still reduced to 30-50% of the control level indicative of delayed postischaemic hypoperfusion. Specific gravity in cortex and hypothalamus reached a maximal decrease 10 min after the end of the ischaemia, and was still significantly reduced at 1 h, while it was normal again 6 hrs later. Regression analysis between regional cerebral blood flows and the corresponding specific gravities were made at various time points, but no significant correlations could be established. Other mechanisms, like vasoconstriction, rheologic or metabolic factors may be causative for the delayed postischaemic hypoperfusion.

Effect of dihydroergotoxine mesylate (Hydergine) on delayed neuronal death in the gerbil hippocampus

The CA 1 neurons in the gerbil hippocampus exhibiting necrosis with delayed onset following 5 min ischemia were reduced markedly by the systemic administration of dihydroergotoxine mesylate (Hydergine; HYG). Immediately after 5 min of forebrain ischemia, the animals were injected intraperitoneally with HYG. Seven days after ischemia, perfusion-fixed brains were processed by conventional histology. The number of neurons per millimeter in the CA 1 pyramidal cell layer were calculated and they were labelled neuronal density. In the control group, the neuronal density was 66.03 +/- 7.37 (mean +/- SEM), in the vehicle group, it was 11.25 +/- 4.93. The neuronal density in the HYG group was 69.19 +/- 6.49. The difference in the neuronal density between the HYG group and the control group was not statistically significant. These data indicate that HYG protects on the CA 1 neurons, and this suggest that the suppression of adrenoceptors by this drugs may be the main mechanism of action. This morphologic outcome may explain the functional amelioration of mental impairment by HYG.

Differences in mortality rate between abrupt and progressive carotid ligation in the gerbil: role of endogenous angiotensin II

Studies have shown that in comparison to rapid occlusion of a vessel, gradual occlusion produces less severe tissue ischemia due to a more effective development of collateral circulation. As other studies have shown that collateral circulation can be enhanced by stimulation of the endogenous renin-angiotensin II system, it was hypothesized that this system is involved in the mechanism of protection against ischemia that obtains during gradual vascular occlusion. To test this hypothesis, mortality rates were evaluated in gerbils subjected to gradual vascular occlusion by means of progressive carotid ligation while simultaneously infused with inhibitors of the renin-angiotensin II cascade--enalaprilat or saralasin. Groups of animals with either abrupt or progressive carotid ligation infused with saline served as controls. Results showed that (1) in saline-infused animals, there was a significant decrease in the mortality rate of progressive-ligated animals when compared to abrupt-ligated animals, and (2) administration of either enalaprilat or saralasin to progressive-ligated animals resulted in mortality rates that were indistinguishable from those of saline-infused abrupt-ligated animals. These results suggest that the endogenous renin-angiotensin system is indeed involved in an adaptive mechanism that occurs during progressive ligation of the carotid artery, and more specifically, that the relatively benign effect of progressive carotid ligation may be due to the action of angiotensin II to stimulate the development of collateral circulation and reduce the severity of focal brain ischemia.

Computer-regulated constant pressure ischemia in the rat: the animal model

A system permitting computer control of partial ischemia in the normotensive rat brain was developed. Right carotid cannulation and bilateral subclavian artery occlusion made the input of blood to the brain dependent solely on left carotid artery flow. Perfusion pressure was controlled by partial compression of this artery with a balloon. The system can produce a range of partial ischemic states maintaining perfusion pressures from 4 to 20 mm Hg. The adequacy of the servo-control system was evaluated in greater detail at requested perfusion pressures of 7 and 12 mm Hg in 14 male Sprague-Dawley rats (300-450 g). Experimentally obtained cerebral perfusion pressures of 6.84 (SD = 0.25, n = 7) and 11.72 (SD = 0.89, n = 7) mm Hg, respectively, demonstrate the efficacy of the system. CBFs were concurrently measured at four separate bilaterally symmetric anatomic sites (cortex, hippocampus, thalamus, and substantia nigra). Significant intra- and interhemispheric differences were found to exist, with regional flows monitored ipsilaterally to the carotid balloon exceeding those of the opposite hemisphere. In summary, this acute model of cerebral ischemia permits control of perfusion pressure over the entire critical partial ischemic range.

Experimental model for repetitive ischemic attacks in the gerbil: the cumulative effect of repeated ischemic insults

An experimental model for repeated ischemic attacks, which allows easy induction of cerebral ischemia of any desired duration and frequency, has been developed in the gerbil. With this procedure, a pronounced cumulative effect on development of edema and tissue injury was observed using 3 separate, 5-min bilateral occlusions of the common carotid arteries spaced at various time intervals. This effect was most evident when the occlusions were carried out at 1-h intervals, i.e., during the period of marked postischemic hypoperfusion. Such animals, killed after 24 h of recirculation, showed significantly more severe edema and brain tissue injury in the areas exposed to ischemia than was observed in animals killed 24 h after single 5- or 15-min occlusions. The changes of regional CBF, assayed with a [3H]nicotine method, indicated a relatively rapid onset of hypoperfusion of similar degree after each release of arterial occlusion. The hypoperfusion recovered significantly within 6 h of recirculation following either single or multiple occlusions, and no residual hypoperfusion was observed in animals which, when killed at 24 h, showed severe edema and brain tissue injury. This model should prove useful in elucidating the pathophysiological mechanisms operative in repetitive cerebral ischemia.

The effect of hypoxia on traumatic head injury in rats: alterations in neurologic function, brain edema, and cerebral blood flow

We evaluated the effects of early posttraumatic hypoxia on neurologic function, magnetic resonance images (MRI), brain tissue specific gravities, and cerebral blood flow (CBF) in head-injured rats. By itself, an hypoxic insult (PaO2 40 mm Hg for 30 min) had little effect on any measure of cerebral function. After temporal fluid-percussion impact injury, however, hypoxia significantly increased morbidity. Of rats subjected to impact (4.9 +/- 0.3 atm) plus hypoxia, 71% had motor weakness contralateral to the impact side 24 h after injury, while only 29% of rats subjected to impact alone had demonstrable weakness (p less than 0.05). Lesions observed on MR images 24 h after injury were restricted to the impact site in rats with impact injury alone, but extensive areas with longer T1 relaxation times were observed throughout the ipsilateral cortex in rats with impact injury and hypoxic insult. Brain tissue specific gravity measurements indicated that much more widespread and severe edema developed in rats with impact injury and hypoxia. [14C]Iodoantipyrine autoradiography performed 24 h after injury showed that there was extensive hypoperfusion of the entire ipsilateral cortex in rats with impact injury and hypoxia. These results show that large areas of impact-injured brain are extremely vulnerable to secondary insults that can irreparably damage neural tissue, and provide experimental evidence for the observed adverse effects of hypoxia on outcome after human head injury.

Experimental intracerebral mass: time-related effects on local cerebral blood flow

Cerebral blood flow (CBF) was measured at different times during the first 150 minutes following an experimental space-occupying lesion produced with a 50-microliter microballoon in rats. Local CBF was measured with the carbon-14-labeled iodoantipyrine quantitative autoradiographic technique. A region of local ischemia developed around the mass, while the remote effects of the mass were minimal. The focal ischemic lesion enlarged with time, and simulated removal of the lesion within this design did not alleviate the ischemia.

Effects of traumatic brain injury on cerebral high-energy phosphates and pH: a 31P magnetic resonance spectroscopy study

Traumatic injuries to the CNS produce tissue damage both through mechanical disruption and through more delayed autodestructive processes. Delayed events include various biochemical changes whose nature and time course remain to be fully elucidated. Magnetic resonance spectroscopy (MRS) techniques permit repeated, noninvasive measurement of biochemical changes in the same animal. Using phosphorus MRS, we have examined certain biochemical responses of rats over an 8-h period following lateralized brain injury (1.5-2.5 atmospheres) using a standardized fluid-percussion model recently developed in our laboratory. Following injury, the ratio of phosphocreatine to inorganic phosphate (PCr/Pi) showed a biphasic decline: The first decline reached its nadir (4.8 +/- 0.4 to 2.8 +/- 0.7) by 40 min post-trauma with recovery by 100 min, followed by a second decline by 2 h that persisted for the remaining 6-h observation period (mean 2.5 +/- 0.5). The first, but not the second, decrease in PCr/Pi was associated with tissue acidosis (pH 7.10 +/- 0.03 to 6.86 +/- 0.11). No changes in ATP occurred at any time during the injury observation period. Such changes may be indicative of altered mitochondrial energy production following brain injury, which may account for the reduced capacity of the cell to recover from traumatic injury.

Acute cerebral ischaemia: concurrent changes in cerebral blood flow, energy metabolites, pH, and lactate measured with hydrogen clearance and 31P and 1H nuclear magnetic resonance spectroscopy. II. Changes during ischaemia

CBF has been measured with the hydrogen clearance technique in the two cerebral hemispheres of the gerbil under halothane anesthesia. This has been correlated with changes in local pH, tissue lactate, and phosphorus energy metabolites measured in the same animals with 1H and 31P nuclear magnetic resonance spectroscopy. We demonstrate a threshold flow value for the metabolic changes associated with energy failure at a level similar to the values previously reported for electrical failure and tissue water accumulation, but higher than that associated with breakdown of extracellular potassium homeostasis.

Quantitative analysis of glucose after transient ischemia in the gerbil hippocampus by light and electron microscope radioautography

Changes in glucose uptake in the gerbil hippocampus were studied by high-resolution [3H]2-deoxyglucose radioautography under sham and postischemic conditions. Sections of dorsal hippocampi were fixed by chemical fixatives or rapid-freezing and freeze-substitution techniques. Light and electron microscope radioautograms showed that the cell soma of each CA1 neuron subjected to transient ischemia revealed various degrees of glucose uptake. In the neuropil of the CA1 stratum radiatum, glucose uptake was higher in the thin dendrites of the ischemic group. Cell damage due to transsynaptic stimulation is suggested by these results.

Acute cerebral ischaemia: concurrent changes in cerebral blood flow, energy metabolites, pH, and lactate measured with hydrogen clearance and 31P and 1H nuclear magnetic resonance spectroscopy. I. Methodology

CBF has been measured with the hydrogen clearance technique in the two cerebral hemispheres of the gerbil under halothane anaesthesia. This has been correlated with changes in local pH, tissue lactate, and phosphorus energy metabolites measured in the same animals with 1H and 31P nuclear magnetic resonance (NMR) spectroscopy. The NMR measurements were made with two surface coils, one on each hemisphere. This article describes the experimental details and shows that in acute unilateral or bilateral forebrain ischaemia metabolic changes can be monitored by NMR with no significant interhemispheric cross talk. The metabolic effects of reperfusion are also shown. The model allows the definition of the time course of the metabolic consequences of regional ischaemia and reperfusion in individual laboratory animals.

Greater disturbance of water and ion homeostasis in the periphery of experimental focal cerebral ischemia

Differences in disturbed water and ion homeostasis in the periphery and the center of focal cerebral ischemia were investigated. Focal cerebral ischemia was induced by occlusion of the right common carotid artery in gerbils. Water and electrolyte content were determined using punched-out samples. In 2 h of ischemia, water content of the cerebral cortex was 79.0 +/- 0.3, 82.0 +/- 0.4, and 80.7 +/- 0.4% (means +/- SE) for the nonischemic region, the periphery, and the center of the ischemic region, respectively (significantly different). Na+ content was increased and K+ content was decreased most prominently in the periphery of the ischemic region. K+ depletion and exogenous Ca2+ accumulation in the peripheral region were visualized by K+ staining and 45Ca autoradiography, respectively. Thus, water and electrolyte changes in the periphery of ischemia were different from those in the center. Brain edema was developed initially in the marginal region of the focal cerebral ischemia.

Experimental intracerebral hemorrhage: effects of a temporary mass lesion

Late pathophysiological events after the production and subsequent removal of an intracerebral mass were investigated using a mechanical microballoon model to simulate intracerebral hemorrhage. Immediately following balloon inflation in the caudate nucleus of rats, there was a significant increase in intracranial pressure to 14 +/- 1 mm Hg (mean +/- standard error of the mean), accompanied by a reduction in cerebral blood flow (CBF) in the ipsilateral frontal cortex, as measured by the hydrogen-clearance technique. Carbon-14-iodoantipyrine autoradiography revealed a significant reduction in the CBF of the ipsilateral caudate nucleus 4 hours after balloon inflation: 31% of the caudate nucleus had a CBF of less than 20 ml X 100 gm-1 X min-1 compared to only 1% in the sham-treated control group (balloon insertion without inflation). The rats with an intracerebral mass exhibited a significant increase in the volume of ischemic damage in the ipsilateral caudate nucleus (17.1% of total volume) compared to only 1.7% in the sham-treated group; however, there was no evidence of cerebral edema. Ischemic damage and reduced CBF persisted for 4 hours after transient inflation of a microballoon in the caudate nucleus. This suggests that ischemic damage occurs at the time of formation of the lesion and is not prevented by its early removal.

Morphometric evaluation of brain infarcts in rats and gerbils

The Levine rat preparation, the gerbil stroke model, and appropriate control animals were used to determine if the 2,3,5-triphenyltetrazolium chloride (TTC) would selectively identify noninfarcted versus infarcted cerebral tissue. The TTC is frequently used to quantify infarcted myocardial tissue and has been shown to have great specificity, reproducibility, and efficacy. The TTC produces a red product upon reaction with the respiratory enzymes (dehydrogenases) present in non-infarcted tissues. Irreversibly damaged tissues, lacking dehydrogenases, do not form red reaction products. Six gerbil brains and seven rat brains were incubated with the TTC, and the unreacted areas were macroscopically identified. The brains were fixed and sectioned for routine hematoxylin and eosin staining to determine the specificity of the TTC. The TTC was found to react selectively only with non-infarcted cerebral tissue. The gross brain sections were evaluated by macroscopic morphometric analysis, and the unreacted area was always ipsilateral to ligation and correlated with histologic identification of infarct. The brains from neurologically intact animals demonstrated neither macroscopic nor histological evidence of infarction. This technique allows macroscopic quantification of infarct size by planimetry. The average area of infarct for the neurologically impaired rats was 34.7% and it was 31.4% for the impaired gerbils. The percentage of surface area of each infarcted slice was found to correlate with the severity of the neurologic deficit. We conclude that TTC staining is effective for macroscopically delineating cerebral infarcts in rats and gerbils, thus permitting quantification of infarct size.

Recirculation after cerebral ischemia. Simultaneous measurement of cerebral bloodflow, brain edema, cerebrovascular permeability and cortical EEG in the rat

The 4-vessel occlusion rat model of cerebral ischemia was modified to permit the simultaneous measurement of cerebral blood flow (hydrogen clearance), brain edema (specific gravity), cerebrovascular permeability (14C-AIB) and electrocardiogram. Surgery was performed in one stage in the anesthetised, paralysed and ventilated rat and severe hemispheric ischemia was produced in all animals. Electrode implantation did not alter cortical specific gravity or Ki for 14C-AIB. During 4-vessel occlusion mean cortical CBF was 5.8 +/- 1.4 ml-1 100 g-1 min. and this was associated with an isoelectric ECoG; 15 min of ischemia produced a significant reduction in mean cortical specific gravity (increase in brain edema). Following 15 min ischemia, 180 min of recirculation were permitted. Post-ischemic blood flow showed an immediate hyperemia (CBF = 202 +/- 12 ml-1 100 g-1 min.) followed by hypoperfusion (CBF = 58 +/- 8 ml-1 100 g-1 min). There was an early further decrease in cortical specific gravity. Further recirculation led to a significant increase in cortical specific gravity (resolution of brain edema). The transfer constant (Ki) for 14C-AIB was not altered at any stage in recirculation. This appears to be a model of pure cytotoxic edema until 180 min recirculation after 15 min cerebral ischemia. Recirculation permitted return of cortical electrical activity.

Regional cerebral ischemia in the gerbil: measurement of regional cerebral blood flow by quantitative autoradiography

Alterations in the regional CBF after occlusion of the posterior communicating, middle cerebral, or common carotid artery were investigated in the gerbil with a quantitative autoradiographic technique using [14C]iodoantipyrine. Occlusion of the posterior communicating artery produced severe ischemia in the ipsilateral hippocampus, thalamus, and dorsal mesencephalon. Occlusion of the middle cerebral artery produced severe ischemia in the ipsilateral rostral and central cerebral cortex and lateral caudate-putamen. Occlusion of the common carotid artery produced ipsilateral hemispheric ischemia of variable degrees. The distribution and degree of cerebral ischemia produced by occlusion of one of these arteries correlated closely to the arterial territory and the extent of collateral blood supply. Since the areas affected after occlusion of the posterior communicating or middle cerebral artery differ, those models will be useful for the comparative investigation of the ischemia-related cerebral pathophysiology associated with different sites of primary lesion.

A reversible type of neuronal injury following ischemia in the gerbil hippocampus

The Mongolian gerbil is known to develop delayed neuronal death in the hippocampus following brief forebrain ischemia (Brain Res 239: 57-69, 1982). The effect of pentobarbital on this slow process of neuronal damage was examined. Immediately following 5 min of bilateral carotid occlusion, pentobarbital (10, 20, or 40 mg/kg) was injected. The control animals received saline injection. Seven days following ischemic insult, animals were perfusion-fixed and the neuronal density in the hippocampal CA1 subfield was counted. Most of the neurons in the CA1 sector survived ischemic insult when pentobarbital was given, whereas most of control group neurons were lost without the treatment. The average neuronal density of 20 mg/kg group was 168.2 +/- 12.3 (SEM) per 1 mm linear length of the CA1 subfield. The density in 40 mg/kg group was 181.1 +/- 14.9. The neuronal density in the whole control group was 34.3 +/- 5.1. The density of unoperated normal gerbils was 212.3 +/- 3.9. This result indicates that the neuronal damage of "delayed neuronal death" is reversible. On the other hand, when pentobarbital was injected 1 hr following ischemia, it showed no effect. The cell change in the CA1 sector, reversible at the initial stage, seems to rapidly become irreversible, while neurons still remain intact morphologically.

Three-dimensional metabolic mapping of the freeze-trapped brain: effects of ischemia in the Mongolian gerbil

The effects of unilateral carotid artery (left or right) occlusion on the three-dimensional metabolic mapping were studied in the Mongolian gerbil (Meriones unguiculatus). The brain was freeze-trapped by the liquid nitrogen funnel technique and was analyzed for the two-dimensional distribution of the reduced pyridine nucleotides and oxidized flavoprotein using a time-sharing micro-light guide surface fluorometer/reflectometer. The results could be summarized as follows. Bilateral carotid artery occlusion induced uniform effects in terms of redox state in the cerebral cortex and the olfactory bulbs. After unilateral carotid occlusion, the redox state of the ischemic ipsilateral hemisphere was higher and was not affected by blood volume changes as evaluated from the reflectance signal scanned in several brains. The narrow band of tissue near the midline of the ischemic hemisphere connected to the contralateral hemisphere blood supply appeared in a large number of gerbils. The anomaly in the redox state of the intracellular space also appeared in subcortical structures, as seen in all the two-dimensional mappings measured. It seems that blood circulation to the two olfactory bulbs is connected to the same blood vessel. It is suggested that unilateral carotid artery-occluded gerbils may show large variability inside the same hemisphere and therefore results must be evaluated very carefully.

Development of cyclooxygenase and lipoxygenase metabolites of arachidonic acid after transient cerebral ischemia

Vasoactive arachidonic acid metabolites are postulated to play a role in the pathogenesis of cerebral ischemia. In order to characterize the local generation of cyclooxygenase and lipoxygenase metabolites of arachidonic acid in transient ischemia with reperfusion, Mongolian gerbils were studied for regional cerebral blood flow (CBF), using the hydrogen clearance technique, and for cerebral levels of the thromboxane metabolite TXB2, and prostaglandins 6-keto-PGF1 alpha and PGE2, as well as the leukotriene LTB4. The gerbils were anesthetized with pentobarbital, and half of the animals were pretreated with the cyclooxygenase inhibitor indomethacin. All received 10 or 20 minutes of dense forebrain ischemia followed by reperfusion of 10 minutes, 50 minutes, or 100 minutes. A separate control group received no ischemic lesion. Regional CBF decreased significantly from 23.7 +/- 2.6 to 4.3 +/- 1.7 cc/100 gm/min during ischemia (p less than 0.01). Reperfusion resulted in initially normal flows (22.5 +/- 5.1 cc/100 gm/min) followed by a progressive hypoperfusion (11.3 +/- 2.7 cc/100 gm/min). All metabolites showed parallel significant (p less than 0.05) increases after transient ischemia and reperfusion compared to baseline levels (values (in pg/mg protein) were: TXB2 45.5 +/- 7.1 vs 23.3 +/- 3.6; 6-keto-PGF1 alpha 262.8 +/- 47.9 vs 175.8 +/- 26.8; PGE2 256.5 +/- 35.6 vs 112.5 +/- 11.2; and LTB4 37.8 +/- 4.6 vs 24.6 +/- 6). These levels were all significantly decreased (p less than 0.05) by pretreatment with indomethacin except for the leukotriene LTB4, which was increased. Transient cerebral ischemia results in a reperfusion abnormality and the local generation of cyclooxygenase products, which are reduced by pretreatment with indomethacin; however, cyclooxygenase inhibition may result in increased substrate availability for the lipoxygenase system. Studies of such an interaction may lead to new understandings of the pharmacological modification of detrimental vascular changes after transient cerebral ischemia.

Angiotensin II decreases mortality rate in gerbils with unilateral carotid ligation

Evidence indicates that after vascular occlusion, infusion of angiotensin II restores blood supply to ischemic tissues by stimulating the development of collateral circulation through a mechanism independent of the mechanical effects of increased blood pressure. To test this effect in focal cerebral ischemia, angiotensin II was intravenously administered for four hours to gerbils immediately after unilateral carotid ligation. Three different pressor doses, 50, 250, and 500 ng/kg/min, were used, and mortality rate was evaluated at 1 and 2 days after vascular occlusion. Two additional groups similarly prepared were infused either with saline or with the pressor agent metaraminol. There was a significant inverse relationship between the infusion dose of angiotensin II and mortality: the greater the infusion dose of angiotensin II, the lower the mortality rate. Infusion of metaraminol, at the dose chosen to mimic the pressor effect of the highest angiotensin II dose, yielded a mortality rate which was statistically indistinguishable from that obtained with saline infusion. It is concluded that the mortality rate after unilateral carotid occlusion is significantly reduced by intravenous administration of angiotensin II through mechanisms unrelated to its hypertensive action. Evidence suggests that this may occur by the enhancement of the development of collateral circulation and therefore the reduction of the severity of brain ischemia.

Prolonged focal cerebral edema associated with partial status epilepticus

Following several days of partial status epilepticus, three patients developed striking focal cerebral edema as demonstrated by computed axial tomography (CT) scan. An angiogram done in one patient showed a capillary blush and early cortical draining veins in the corresponding area. All patients developed severe focal neurological deficit which resolved as the edema improved, and this was demonstrated on serial CT scans (at 6 months, 1 month, and 2 months, respectively). In the first patient, an underlying tumor, and in the second and third, vascular occlusions, were suspected because we were not aware that edema due to status epilepticus could produce changes of such intensity and duration. The neurological disability after the partial status was long-lasting but reversed completely in our patients. Maximal radiological changes occurred in the area of maximal epileptic discharge. Minimal atrophic changes persisted in two of the three patients. The clinical, CT scan, and angiographic findings suggest that partial status epilepticus can be associated with abnormal vascular permeability leading to prolonged focal cerebral edema. Similar pathophysiology of lesser intensity may be responsible for shorter postictal neurological deficits. Awareness of this clinical and radiological entity should avoid misdiagnosis of cerebral tumor or infarction.

Brain specific gravity and CT scan density measurements after human head injury

White matter specific gravity was measured using the microgravimetric method in 20 comatose patients with diffuse head injury who were undergoing intracranial pressure (ICP) monitoring, and in 19 patients with focal injuries who were undergoing evacuation of contusions or intracerebral hematomas. Computerized tomography (CT) density readings were obtained for each site of white matter sampling by locating the sampling site on the preoperative CT scan. A significant correlation was found between the specific gravity values and the CT density numbers (r = 0.775; p less than 0.001). Patients with focal injuries demonstrated reduced perifocal specific gravity, suggesting brain edema. The mean specific gravity in patients with diffuse injury was within the normal range. In 10 of 12 patients in whom the specific gravity was above the normal range, the CT density was also above the normal range. These data suggest that cerebral vascular engorgement is the cause of the high specific gravity. Six (60%) of this small subgroup of 10 patients also demonstrated a high ICP.

Indomethacin-mediated improvement following middle cerebral artery occlusion in cats. Effects of anesthesia

Focal cerebral ischemia initiates multiple detrimental effects in the brain. Chief among these are the regional development of ischemic edema, decreased local perfusion, altered neuronal function, and eventual infarction. To determine if pretreatment with the cyclo-oxygenase inhibitor, indomethacin, would result in improvement in these parameters, adult cats were given indomethacin or control solvent (4 mg/kg intraperitoneally twice daily) and were studied for periods up to 24 hours after right middle cerebral artery occlusion. The interaction of anesthetic agents with indomethacin was also examined in separate groups of experimental animals using pentobarbital and ketamine. In cats allowed to recover from pentobarbital anesthesia, indomethacin reduced gray and white matter edema at 6 and 24 hours after occlusion (p less than 0.05). This was noted in densely areas (indomethacin = 84.3%, control = 87.5%), "penumbra" regions (indomethacin = 82.5%, control = 85.3%), and in nonischemic zones (indomethacin = 81.5%, control = 82.3%) at 24 hours. Somatosensory evoked potential amplitude and central latency were also improved in the indomethacin group (p less than 0.05), as was cerebral perfusion (p less than 0.05). In animals anesthetized with continuous ketamine administration, cerebral edema and perfusion as well as evoked potentials were not significantly improved in any region by indomethacin. Regional cerebral blood flow in the group was increased by indomethacin in the nonischemic opposite hemisphere (indomethacin = 64.7 cc/100 gm/min, control = 48.5 cc/100 gm/min, p less than 0.05), but not in the penumbra region of the ischemic hemisphere (indomethacin = 15.0 cc/100 gm/min, control = 18.6 cc/100 gm/min, p less than 0.05), when measured 4 hours after occlusion. This suggested a steal phenomenon. Beneficial effects of indomethacin were evident in the presence of pentobarbital, but not after ketamine anesthesia. This suggests a synergism dependent on decreased arachidonic acid production from pentobarbital-stabilized membranes coupled with diminished production of cyclic endoperoxides from available arachidonate due to inhibition of cyclo-oxygenase with indomethacin.

The role of cerebral blood volume changes in brain specific-gravity measurements

Cerebral blood volume (CBV) was calculated in gerbils from specific-gravity (SG) changes between normal and saline-perfused brains. Furthermore, changes in CBV were investigated during ischemia using carbon-14-labeled dextran (MW 70,000) as an intravascular marker. Both data were used to evaluate the possible error due to a change in CBV on the measurement of ischemic brain edema by the SG method. The methodological error found was 0.0004 for a 100% CBV change. This error is insignificant, being less than the standard deviation in the SG measured for the gerbil cortex (SG 1.0494 +/- 0.0006). Thus, CBV changes are not responsible for the SG variations observed during the first phase of ischemia. These variations are better explained as an increase of brain water content during ischemia.

Regional blood-brain barrier permeability changes after restoration of blood flow in postischemic gerbil brains: a quantitative study

A quantitative technique utilising [14C]alpha-aminoisobutyric acid as a tracer was used to study cerebrovascular permeability in 22 Mongolian gerbils. Seven other animals were used to measure cerebral blood volumes. Global cerebral ischaemia was produced by temporary bilateral carotid artery occlusion (60 min) in 16 gerbils that were sacrificed at 1, 2, and 3 h following reperfusion. The blood-to-brain transfer constant was significantly increased after 2 h of reperfusion in the ischaemic zones and also in structures, like the cerebellum, not supplied by the carotid artery and not ischaemic during the vessel occlusion. The blood-brain barrier (BBB) alterations were coincident with the onset of ischaemia--induced seizures that were accompanied by sudden "spikes" of systemic blood pressure. Epilepsy may play an important role in the development of BBB damage in this ischaemic model, and this factor must be considered in the interpretation of BBB damage data in gerbils.

CBF and time thresholds for the formation of ischemic cerebral edema, and effect of reperfusion in baboons

Ischemic cerebral edema has been studied in 41 baboons, with regional cerebral blood flow (CBF) determined by hydrogen clearance, and edema measured by microgravimetry. A threshold of ischemia has been identified for baboon cortex and subcortical white matter, which has to be crossed before edema formation begins. This threshold is 40.5% of normal CBF in cortex, and 34.4% of normal flow in subcortical white matter. A time threshold has also been determined, and the baboon brain can withstand 30 minutes of ischemia of the middle cerebral artery without significant edema formation. Reperfusion of ischemic brain has no effect on tissue water if the ischemic flow and time thresholds have not been crossed. Reperfusion of cortex, where water has begun to accumulate, exacerbates the water accumulation in proportion to the extent of the reperfusion. If these results are applicable to man, restoration of flow should not be attempted after an ischemic insult that reduces flow to less than 40% of normal unless it can be accomplished within 30 minutes of the insult. Provided CBF can be restored to above the 40% threshold within 30 minutes, reversal of the neurological deficit and prevention of ischemic edema can be expected.

Brain tissue pressure in focal cerebral ischemia

Twenty-three anesthetized cats underwent permanent middle cerebral artery occlusion in a study of the relationships of regional cerebral blood flow, ventricular fluid pressure, brain tissue pressure, and ischemic edema formation. A pressure gradient of 8 mm Hg developed between ischemic tissue and normally perfused tissue during a 4-hour observation period after occlusion. Brain water accumulated as tissue pressure rose, while blood flow in the same area fell. The data suggest, but do not prove, that ischemic brain edema causes tissue pressure to rise focally, and that blood flow to the ischemic zone is compromised further by the resultant hydrostatic pressure gradient.

Relationship between hemispheric cerebral blood flow, central conduction time, and clinical grade in aneurysmal subarachnoid hemorrhage

The relationship between central conduction time (CCT) and hemispheric cerebral blood flow (CBF) has been examined in 20 patients presenting with subarachnoid hemorrhage. A total of 63 combined CCT/CBF recordings were performed at various times throughout the hospital course of these patients, and the findings were correlated to clinical status. The initial-slope index of the CBF (CBF isi) was found to correlate well with clinical grade, and a gradation in flow was noted between the different neurological grades. Patients in Grades I and II (Hunt and Hess classification) had the highest flows (mean CBF isi = 47.2 +/- 8.1); Grade III patients had intermediate flows (mean CBF isi = 39.6 +/- 7.8); and Grade IV patients had the lowest flows (mean CBF isi = 32.0 +/- 6.4). While CCT tended to become increasingly prolonged with worsening grade, a significant difference could not be demonstrated between Grade I, II, and III patients. Only when Grade IV status was reached was the CCT significantly prolonged. When CBF isi and CCT were examined, a threshold relationship was noted between CBF isi and CCT prolongation. At flow values above 30, little change was noted in CCT, and CCT remained in the normal range. However, at flow values below 30, CCT became increasingly prolonged as blood flow diminished. The degree of CCT prolongation appeared to be directly proportional to the degree of blood flow diminution at flows below threshold.

Regional studies of changes in brain fatty acids following experimental ischaemia and reperfusion in the gerbil

Regional studies of brain phospholipid metabolism were carried out during a period of ischaemia induced in the gerbil by bilateral carotid occlusion for 60 min. The associated changes in free fatty acids (FFAs) during this period and following recirculation for up to 180 min were noted. Following ischaemia there was a generalised rise in the levels of all FFAs with no selective release of either the unsaturated (arachidonic and docosahexaenoic) or saturated (palmitic and stearic) fatty acids. There were no observed differences between the brain regions studied, which is in contrast to previously reported observations for prostaglandins. There was also no indication of any specific phospholipid fraction being involved in FFA release. This would indicate that the release of FFAs from phospholipids is a nonspecific event, probably due to the action of hydrolytic lipases. Restoration of the circulation resulted in a short, sharp increase (within 5 min) in all FFAs, but in contrast to the observations during ischaemia alone there was a relatively larger rise in the unsaturated FFAs as compared to the saturated FFAs. Following this increase there was a gradual general decline in all FFA levels until 180 min of reperfusion. Since there was no preferential depletion of unsaturated FFAs during reperfusion, when free radical attack is considered to be at its maximum, it is our opinion that free radical peroxidation is unlikely to explain the pathology described in our model.

Prostaglandin synthesis and oedema formation during reperfusion following experimental brain ischaemia in the gerbil

Regional brain tissue prostaglandin (PG) levels have been measured during ischaemia produced by bilateral carotid occlusion for 1 hour and following restoration of flow. In the normal gerbil, the frontal cortical levels of PGF2 alpha were: 6.7 +/- 1.3 pg/mg and for PGE2: 6.4 +/- 1.1 pg/mg of brain tissue protein. Following 1 hour of ischaemia PGF2 alpha rose to 50.4 +/- 8.3 pg/mg whilst there was only a slight rise in PGE2 (10.7 +/- 1.6 pg/mg). Post ischaemic values for parietal and occipital areas were somewhat higher, but showed the same trend. Within 15 minutes of the restoration of flow there was a massive increase in PGF2 alpha levels which reached a peak at 2 hours (300 pg/mg) and then subsided to control values. PGE2 levels did not change for the first 30 minutes of recirculation, but then rose for the rest of the period of observation. The pattern of cytotoxic oedema resembled PGF2 alpha closely while the Evans blue staining (vasogenic oedema) was similar in time to the PGE2 pattern.

Evolution and resolution of oedema following severe temporary cerebral ischaemia in the gerbil

Regional cerebral blood flow (rCBF) and oedema following profound temporary ischaemia were studied in the gerbil. Ninety-four per cent of animals died within 24 hours of reperfusion; 50% by 4 hours. Regional differences in oedema (specific gravity method), Evans blue (EB) staining and rCBF (hydrogen clearance technique) occurred. Oedema developed during arterial occlusion, being inversely proportional to residual flow and was markedly exacerbated during reperfusion. Reperfusion hyperaemia was maximal in the parietal and hippocampal regions (ischaemic rCBF 4 ml 100 g-1 min-1). Oedema was disappearing in all areas by 3 hours of reperfusion and autoregulation returned in the occipital region (mean ischaemia rCBF 8 ml 100 g-1 min-1). EB staining and haemorrhage appeared in the thalamus (rCBF 10 ml 100 g-1 min-1) as oedema was decreasing. It is suggested that the amount of oedema and hyperaemia during reperfusion are dependent on the severity of the ischaemia. Areas of moderate ischaemia (8-10 ml 100 g-1 min-1) show little hyperaemia and greater oedema resolution during reperfusion as compared to areas of severe ischaemia (circa 4 ml 100 g-1 min-1) where there is marked hyperaemia with less oedema resolution. Early in the reperfusion period, oedema is not associated with EB staining and indicates a cytotoxic mechanism. The vasogenic component, with macroscopic haemorrhages and leakage of EB occurs later. In this model it is concluded that the early cytotoxic oedema formation and hyperaemia are phenomena with little bearing on mortality, which correlates better with later vasogenic changes.

Common carotid artery stump pressure in the gerbil stroke model

In 106 slightly anaesthetised adult mongolian gerbils one common carotid artery (CCA) was ligated and the blood pressure in the distal and in the proximal stump was monitored for 8 minutes. The mean distal CCA stump pressure of the 39 nonsurvivors was 15 (+/- 6) mm Hg, that of the 25 survivors with retinocerebral infarcts was 25 (+/- 6) mm Hg, and that of the 42 intact survivors was 31 (+/- 7) mm Hg. The corresponding mean arterial blood pressures (MABP), as measured in the proximal CCA stump, were 81 (+/- 12) mm Hg, 84 (+/- 13) mm Hg, and 87 (+/- 11) mm Hg, respectively. There were no differences between the samples concerning sex, body weight, rectal temperature, arterial blood gases, arterial pH, and haematocrit. Measurements in a second series of 10 awake gerbils showed that the mean values of MABP, heart rate, and respiratory rate of the nonsurvivors were less than those of the survivors during 90 minutes after CCA ligation. It is inferred that in the mongolian gerbil the lower threshold of the arterial blood pressure for the development of brain infarcts ranges within 22 and 25 mm Hg, that is, within the values found in monkeys and cats. The longlasting depression of respiration and circulation in the nonsurvivors is considered to be related to the phenomenon of diaschisis .

Selective vulnerability in the gerbil hippocampus following transient ischemia

Following brief ischemia, the Mongolian gerbil is reported to develop unusual hippocampal cell injury (Brain Res 239:57--69, 1982). To further clarify this hippocampal vulnerability, gerbils were subjected to ischemia for 3, 5, 10, 20, and 30 min by bilateral occlusion of the common carotid arteries. They were perfusion-fixed after varying intervals of survival time ranging from 3 h up to 7 days. Following brief ischemia (5--10 min), about 90% of the animals developed typical hippocampal damage. The lesion was present throughout the extent of the dorsal hippocampus, whereas damage outside the hippocampus was not observed. Each sector of the hippocampus showed different types of cell reaction to ischemia. Ischemia cell change was seen in scattered CA4 neurons , and reactive change was found in CA2, whereas CA1 pyramidal cells developed a strikingly slow cell death process. Ischemia for 3 min did not produce hippocampal lesion in most cases. Following prolonged ischemia (20--30 min), brain injury had a wide variety in its extent and distribution. These results revealed that the gerbil brief ischemia model can serve as an excellent, reliable model to study the long-known hippocampal selective vulnerability to ischemia. Delayed neuronal death in CA1 pyramidal cells was confirmed after varying degrees of ischemic insult. These findings demonstrated that the pathology of neuronal injury following brief ischemia was by no means uniform nor simple.

Why are infant gerbils more resistant than adults to cerebral infarction after carotid ligation?

Younger gerbils have been found to be more resistant than adults to cerebral infarction after carotid ligation. In this study, the perfused cerebral area after bilateral common carotid occlusion was evaluated in infant, young, and adult Mongolian gerbils by the carbon black perfusion method to assess the existence and significance of collateral blood vessels between the vertebrobasilar and carotid circulations. Nineteen gerbils were divided into three groups (i.e., infant, young and adult gerbils aged 3-4, 5-7, and 10-17 weeks, respectively). After bilateral common carotid artery occlusion, carbon black was injected directly into the left ventricle by cardiac puncture through the closed thorax. In five of eight infant gerbils, the whole brain was perfused by carbon black, while in the remaining three, only the cerebellum and brainstem were stained well, and marked bilateral cerebral pallor was observed. On the other hand, carbon black did not perfuse the brain region supplied by the carotid arteries, both in young and adult gerbils (11 animals in total). These results suggest that infant gerbils might have a more highly developed network of collateral blood vessels between the vertebrobasilar and carotid circulations, and the existence of such a significant network might be the basis for the fact that infant gerbils are resistant to cerebral infarction following carotid ligation. We propose that gerbils should be used as a stroke model only when they are 5 weeks old or older.

Brain free fatty acids, edema, and mortality in gerbils subjected to transient, bilateral ischemia, and effect of barbiturate anesthesia

Brain free fatty acids (FFAs) and brain water content were measured in gerbils subjected to transient, bilateral cerebral ischemia under brief halothane anesthesia (nontreated group) and pentobarbital anesthesia (treated group). Mortality in the two groups was also evaluated. In nontreated animals, both saturated and mono- and polyunsaturated FFAs increased approximately 12-fold in total at the end of a 30-min period of ischemia; during recirculation, the level of free arachidonic acid dropped rapidly, while other FFAs gradually decreased to their preischemic levels in 90 min. In treated animals, the levels of total FFAs were lower than the nontreated group during ischemia, but higher at 90 min of reflow, and the decrease in the rate of free arachidonic acid was slower in the early period of reflow. Water content increased progressively during ischemia and recirculation with no extravasation of serum protein, but the values were consistently lower in the treated group. None of the nontreated animals survived for 2 weeks; in contrast, survival was 37.5% in the treated group. It is suggested that barbiturate protection from transient cerebral ischemia may be mediated by the attenuation of both membrane phospholipid hydrolysis during ischemia and postischemic peroxidation of accumulated free arachidonic acid.

Biochemical changes during graded brain ischemia in gerbils. Part 1. Global biochemical alterations

In Mongolian gerbils (Meriones unguiculatus) cerebral ischemia was produced by occlusion of the right common and the left external carotid arteries. Gerbils were classified according to their neurological appearance as "symptom-negative" (8 animals), "mild symptoms" (unilateral hemiparesis, 10 animals) and "severe symptoms" (hemiparesis and rolling seizures, 8 animals). Two hours after vascular occlusion various substrates and enzymes related to the energy-producing metabolism, were assessed in tissue samples from both hemispheres. In symptom-negative animals, the only change was a slight decrease of glycolytic intermediates in the right hemisphere. In animals with mild symptoms, the right hemisphere additionally exhibited an impairment of the redox and energy state and an enhancement of the activity of most enzymes of the glycolytic pathway, except hexokinase. In animals with severe symptoms, these changes were even more pronounced and affected--to a lesser degree--also the left hemisphere. The results obtained demonstrate that the neurological appearance of the animals after vascular occlusion correlates with the biochemical alterations and, therefore, can be used for estimating the density of graded ischemia.

A 31P nuclear magnetic resonance in vivo study of cerebral ischaemia in the gerbil

We have used the noninvasive method of 31phosphorus nuclear magnetic resonance (31P NMR) in vivo to follow changes in phosphorous metabolite concentrations and the intracellular pH in the right and left hemispheres and in the cerebellum of gerbil brains after the occlusion of the right carotid artery. Spatial resolution over the brain was possible using surface coils. Ligation, which is know to cause ischaemia in this species in the ipsilateral hemisphere, resulted in the diminution of phosphocreatine and adenine nucleotides and a decrease in tissue pH. Less acidification occurred in the contralateral hemisphere and in the cerebellum. The high-energy metabolite concentrations, phosphocreatine and adenosine triphosphate (ATP), declined in unison in the ischaemic region, in marked contrast to the sequence of events in skeletal muscle, in which phosphocreatine buffers against an immediate fall in ATP concentration. In a separate series of gerbils, 31P NMR spectra were followed for exactly 1 h after carotid ligation. The animals were then sacrificed and brain grey matter specific gravity was rapidly measured to assess the development of oedema. There was a clear correlation between abnormality of spectra and the presence of oedema. It cannot, however, be confidently asserted that a normal spectrum is never seen in oedematous gerbil brains. 31P NMR spectra specific gravity and histological changes shown by light microscopy have been correlated and show that useful signals are received from a depth of at least 4 mm or more from the 10-mm diameter coil.