ded in paraffin, and cut into three 3-mm blocks, from which 6-fLm sections were cut and stained with acid fuchsin and cresyl violet for examination by light micros.
Journal of Cerebral Blood Flow and Metabolism 11:949-956 © 1991 The International Society of Cerebral Blood Flow and Metabolism Published by Raven Press, Ltd., New York
A New Method for Producing Temporary Complete Cerebral Ischemia in Rats
*Reizo Shirane, Hiroaki Shimizu, Motonobu Kameyama, and tPhilip R. Weinstein *Department of Neurological Surgery, Tohoku University, Sendai, Japan; and tDepartment of Neurological Surgery, School of Medicine, University of California, San Francisco, California, U.S.A.
Summary: A new model of temporary complete cerebral ischemia was developed and tested in 64 rats. With use of microsurgical techniques, both pterygopalatine and exter nal carotid arteries were occluded and the basilar artery was coagulated to reduce potential collateral CBF during ischemia. After this preliminary five-vessel occlusion, temporary global ischemia was induced by occluding the common carotid arteries (CCAs) with microclips. To val idate the method, CBF was measured autoradiographi cally in 24 anatomical regions at death after 5 min of ischemia or after 15 min of ischemia followed by 5 min of reperfusion. Mean arterial blood pressure and arterial blood gases remained stable under controlled endotra cheal ventilation and anesthesia (halothane, 70% N20, and 30% O2) throughout the CBF experiments, except for a I(}-15% increase in mean arterial blood pressure for 1-5 min after bilateral CCA occlusion. After the initial five vessel occlusion, the EEG did not change, and local CBF
levels were comparable to those in anesthetized non surgical controls. When the CCAs were occluded, the EEG flattened rapidly; after 5 min of ischemia, autoradi ography showed no detectable blood flow in the forebrain and cerebellum. The local CBF levels measured after 15 min of temporary global ischemia and 5 min of reperfu sion demonstrated relatively homogeneous postischemic hyperperfusion; only two of eight rats had several 1- to 3-mm areas of no-reflow. Survival studies showed in creasing motor impairment after 1 0, 15, 30, and 60 min of temporary CCA occlusion. Ischemic neuronal damage was observed histologically in the hippocampus and basal ganglia 24 h after 10 min of temporary ischemia. This model consistently produces reversible complete cerebral ischemia in the rat without the use of intracranial hyper tension, systemic hypotension, hypoxia, or a neck tour niquet. Key Words: Basilar artery-Carotid artery Cerebral blood flow-Cerebral ischemia-Reperfusion.
The metabolic and microcirculatory effects of reperfusion after cerebral ischemia are only par tially understood (Hossmann and Kleihues, 1973). Pathophysiological differences in the metabolic responses to complete and incomplete cerebral ischemia and reperfusion have been described but require further study (Yoshida et aI. , 1985; Rehn crona et aI. , 1982, 1987; Dietrich et aI. , 1987; Steen et aI., 1979). Because of individual variations in vas cular anatomy and collateral blood flow potential, the response to extracranial cerebrovascular occlu sion during incomplete ischemia can vary in most
experimental models, including the standard four vessel occlusion model (Pulsinelli and Brierley, 1979) and the three-vessel (basilar and bilateral common carotid) occlusion method without neck ligation (Kameyama et aI. , 1985) in rats. After bi lateral common carotid (CCA) and vertebral artery occlusion in the rat, some collateral CBF is possible by reversal of flow through the anterior spinal ar tery into the distal vertebral arteries and through extracranial muscular branches that arise distal to the site of carotid occlusion, such as the external carotid and pterygopalatine arteries. To establish a method for creating a reproducible, anatomically uniform reduction of CBF during temporary isch emia, we developed a microsurgical technique of occluding seven vessels to induce reversible com plete cerebral ischemia in rats. This report de scribes the method and its validation by autoradio graphic measurements of local CBF in 24 anatomi cal regions.
Received September 27, 1990; revised April 8, 1991; accepted April 9, 1991. Address correspondence and reprint requests to Dr. P. R. Weinstein at Department of Neurological Surgery, % Editorial Office, 1360 Ninth Ave. , Suite 210, San Francisco, CA 94122, U.S.A. Abbreviations used: CCA, common carotid artery; ECA, ex ternal carotid artery.
949
950
R. SHIRANE ET AL. (A)
METHODS Animal preparation and physiological monitoring Male Sprague-Dawley rats weighing 300-400 g were anesthetized in a halothane inhalation chamber, orotra cheally intubated with a 16-gauge angiocatheter, con nected to a rodent respirator (model 683; Harvard Instru ment Co., Cambridge, MA, U.S.A.), and paralyzed by intraperitoneal injection of pancuronium bromide (I mg/kg). Anesthesia was maintained with 1.5% halothane in 70% NzO and 30% O2 during surgery and with 0.7% halothane during ischemia and reperfusion. Bilateral sub cutaneous platinum electrodes were placed over the cal varium for continuous monitoring of a single-channel EEG with a Neurotrac recorder (Interspec, Consho hocken, PA, U.S.A.). Both femoral arteries and a femoral vein were cannulated with PE-50 plastic tubing; the arte rial catheter was connected to a Statham pressure trans ducer for continuous monitoring and recording of sys temic arterial blood pressure with an oscilloscope (Hewlett-Packard Co., Palo Alto, CA, U.S.A.) and a physiological recorder (model 79D; Grass, Quincy, MA, U.S.A.). Rectal temperature was maintained at 37-38°C with a temperature-regulated water jacket. The respira tory rate and volume were adjusted to maintain an arterial carbon dioxide tension of 35-40 mm Hg. Before surgery, all rats received 0.05 ml of atropine (40 mg/ml) by intra peritoneal injection to reduce perioperative respiratory secretions.
Vessel occlusion Dissection and vessel occlusion were performed micro surgically through a midline cervical incision (Fig. I). The omohyoid muscles were retracted bilaterally. Both exter nal carotid arteries (ECAs) and the pterygopalatine branches of the internal carotid artery were occluded by bipolar cauterization with microforceps. The trachea and esophagus were retracted medially to expose the ventral surface of the clivus. A 2 x 2-mm hole was drilled in the midportion of the clivus; the dura and the arachnoid membrane were opened, and the basilar artery was coag ulated at the midpontine level. The bone aperture was then filled with Gelfoam sponge. After the five-vessel occlusion, the CCAs were ex posed carefully to prevent vasospasm and occluded with microclips to produce temporary global ischemia. The success of the occlusion was verified by the absence of arterial pulsation in the distal CCAs. Heparin (200 U/kg) was given 5 min before the onset of ischemia to prevent intravascular coagulation. Reperfusion was verified by observing both the reexpansion of the CCAs and the re turn of arterial pulsations in the internal carotid artery after removal of the microclips.
Autoradiographic measurement of CBF The tissue saturation method described by Sakurada et al. (1978) was used to determine local CBF in 24 anatom ical regions. The radioactive tracer 4-iodo-[N-meth yl-14C]antipyrine, 100 !LCi/kg in 0.85 ml of saline solution, was infused intravenously at a constant rate over 45 s with an infusion pump (model 901; Harvard Instrument Co.). Nine 5-s blood samples were withdrawn from the arterial catheter; 25-!LI aliquots of each sample were an alyzed in a liquid scintillation counter to determine the arterial concentration curve for the isotope. After the blood samples were obtained, the rat was decapitated. The brain was quickly removed, frozen in powdered dry J Cereb Blood Flow Metab, Vol. II, No.6, 1991
(A) and brain illustrating cerebrovascular anatomy. Circle indicates cra niotomy site. Arrow indicates the site of basilar artery occlusion. Potential collateral sources of CSF eliminated by the seven vessel occlusion technique include the external carotid (EG), pterygopalatine (PPA), and anterior spinal arteries. IC, internal carotid artery; CC, common carotid artery; SA, basilar artery; SCA, superior cerebellar artery; AC, anterior cerebral artery; MC, middle cerebral artery; PC, posterior cerebral artery. FIG. 1. Diagram of ventral surface of the rat skull
(B).
TEMPORARY COMPLETE CEREBRAL ISCHEMIA IN RATS
951
TABLE 1. Physiological findings immediately before death in Groups A-D Body temperature (OC)
Group A (control, n B (5-VO, n
=
37.1 37.3 37.2 37.4
6)
=
6)
C (7-VO, n 8) D (l5-min TCI, n =
=
Values are means
8) ±
±0.2 ±0.5 ±0.2 ±0.3
MABP (mmHg)
(mmHg)
91 ±7 98 ± 14 140 ±25 94 ±17
121 ±20 116 ± 15 109 ±27 118 ±17
PaOZ
PaCOZ
(mmHg) 37.2 35.8 38.1 38.3
pH
±3.9 ±2.2 ±2.7 ±4.4
7.40 ±0.05 7.48 ± 0.05 7.39 ±0.09 7.41 ± 0.17
SD. VO, vessel occlusion; TCI, temporary complete ischemia.
ice, and cut into sections 20 fLm thick with a Hacker 5030 cryostat (Bright Instrument Co., Huntingdon, England). The sections were mounted on glass coverslips, dried on a hot plate, placed in an x-ray cassette, and exposed to Kodak SB-5 x-ray film for 7 days. The films were ana lyzed with a videodigitizer image analysis system (Imag ing Research, Peterborough, Ontario, Canada) to deter mine regional optical density, from which the CBF was calculated as described by Sakurada et al. (1978). In each anatomical region, CBF was determined from three adja cent points in each structure on three adjacent sections in both hemispheres. The values presented are the averages of the 18 CBF determinations for each region.
Experimental groups CBF studies were performed 2 h after the induction of anesthesia in nonoperated, nonischemic control rats (Group A, n 6); 5 min after five-vessel occlusion (Group B, n 6); 5 min after complete ischemia was induced by bilateral CCA (i.e., seven-vessel) occlusion (Group C, n 8); and 5 min after removal of the CCA clips following 15 min of complete ischemia (Group D, n 8). Neurological function was evaluated in survival studies by observing motor activity and respiratory function for 24 h after recovery from anesthesia in five groups of rats: =
=
=
=
sham-operated control rats that underwent vessel expo sure without occlusion (Group E, n 4); rats that under went vessel dissection and permanent basilar artery oc clusion only (Group F, n 4); rats that underwent 15-min five-vessel occlusion (Group G, n 4); rats that under went 15 min (n 4), 30 min (n 4), or 60 min (n 4) of temporary ischemia (seven-vessel occlusion) followed by 6 h of reperfusion (Group H); and rats that underwent lO-min temporary bilateral ECA and CCA occlusion after permanent basilar and bilateral pterygopalatine artery oc clusions (Group I, n 12). Motor activity was scored as follows: 0, normal; 1, symmetrical movement, but the rat could be pushed over; 2, quadriparesis with reduced spontaneous or stimulated movement; 3, no spontaneous movement. Histological studies were performed in Group I rats that underwent lO-min ischemia induced by seven-vessel occlusion followed by 24 h of reperfusion. After 24 h, the rats were anesthetized, and perfusion-fixation was per formed by intracardiac injection of heparinized normal saline solution followed by 10% formalin in 0.1 M phos phate buffer (pH 7.4). The brains were removed, embed ded in paraffin, and cut into three 3-mm blocks, from which 6-fLm sections were cut and stained with acid fuchsin and cresyl violet for examination by light micros copy. =
=
=
=
=
=
=
TABLE 2. Regional CBF levels 5 min after onset of TCI or reperfusion in rats in Groups A-D
Structure Frontal cortex Sensorimotor cortex Parietal cortex Auditory cortex Visual cortex Subcortical white matter Corpus callosum Capsula interna Caudoputamen Septal nucleus Thalamus Hypothalamus Amygdala Hippocampus Superior colliculus Inferior colliculus Lateral geniculate body Medial geniculate body Red nucleus Cerebellum Pons Medulla Spinal gray Spinal white
Group A (control, n
=
179 156 158 150 157 54
± 17 ±25 ±27 ±32 ±22 ±14
53 ±5 49 ± 3 174 ±39 97 160 84 87 93 146 174 144 132 127
± 14 ±27 ±7 ±10 ±12 ±10 ±7 ±22 ±21 ± 5
82 ± 15 107 ±12 108 ±22 91 ±15 32 ± 7
6)
Group B (5-VO, n 6)
Group C (7-VO, n 8)
168 ±29 166 ± 20
0 0 0
383
0 0 0 0 0 0 0 0 0 0 0
365 337
0 0 0 0
249 ±42 317 ± 71
=
151 183 178 55 55 53 176 118 166 95 88 108 157 180 129 127 144
±27 ±39 ±49 ±12 ±7 ±8 ±25 ±27 ±8 ±15 ±12 ±13 ±22 ±59 ±20 ±15 ±44
81 ±20 109 ±17 111 ±10 100 ±19 36 ±5
=
0 0 0 105 ±12 116 ± 12 34 ±2
Group D (l5-min TCI) ±
71
361 ±60 411 ± 81 ± ±
33 47
162 ± 18 141 ±27 119 ±15 351 ± 64 213 ±25 328 ±32 164 ± 10 158 ±51 276 ±49
401 ± 93 421 ±76 336 ±64 177 188 93 82 58
±10 ±25 ± 15 ±21 ±10
Values are means ±SD, expressed as mll100 glmin. TCI, temporary complete ischemia; VO, vessel occlusion. J Cereb Blood Flow Metab, Vol. 11, No.6, 1991
952
R. SHIRANE ET AL.
Statistical analysis The paired t test was used to determine the statistical significance of differences in all mean local CBP values between Groups A and B.
RESULTS
The results of physiological monitoring immedi ately before CBF studies in Groups A-D are pre sented in Table 1. The pH and blood gas values in each group remained normal throughout the exper iment. The mean systemic arterial blood pressure increased slightly after basilar artery occlusion, in creased by 10-- 15% after occlusion of the CCAs, and then returned to the baseline level within 1-5 min. Bleeding was encountered during dissection of the basilar artery after the dura was opened only in three rats, which were excluded from the study. The EEG recordings demonstrated predominantly an 8- to 10-Hz pattern that did not change during the initial surgical preparation or after five-vessel oc clusion in any group. Within 15 s after bilateral CCA occlusion, EEG activity disappeared in all rats in Groups C and D. The EEG remained isoelectric until the end of each experiment. After the onset of reperfusion in Group D, no EEG recovery or sei zure activity was observed. The EEG was not re corded after reversal of anesthesia in the survival studies. The calculated local CBF values are presented in Table 2. There were no significant differences in local CBF values between control rats that were anesthetized only (Group A) and those that under went five-vessel occlusion (Group B) (see Fig. 2). In rats subjected to seven-vessel occlusion (Group C), no isotope activity was detected in the forebrain or cerebellum (Fig. 3). Isotope uptake was visible only below the level of the basilar artery occlusion in the medulla and spinal cord. In all eight rats sub jected to 15 min of temporary global ischemia followed by 5 min of reperfusion (Group D), diffuse postischemic hyperperfusion was observed (Fig. 4). Two of these rats had several 1- to 3-mm areas of no-reflow in the hippocampus and in the thalamus. Neurological observation after recovery from an esthesia revealed no motor deficits in sham operated rats or in those subjected only to perma nent basilar artery or five-vessel occlusion (Groups E, F, and G). Despite the administration of atro pine, all rats in the five-vessel temporary occlusion group, but none in the sham or basilar artery occlu sion group, had intermittent partial respiratory ob struction by tracheal secretions and died with re spiratory insufficiency 2-6 h postoperatively. In
J Cereb Blood Flow Metab, Vol. II, No.6, 1991
Group H (temporary global ischemia for 15, 30, or 60 min after seven-vessel occlusion), 15 min of ischemia produced Grade 2 motor deficits in three rats and Grade 1 deficits in one; 30 min of ischemia produced Grade 2 deficits in all four rats. The sur vival times in these eight rats (2-6 h) were related to the severity of upper respiratory obstruction after recovery from anesthesia. However, no fatal respi ratory obstruction was observed in Group I rats, in which the seven-vessel occlusion procedure was modified by performing temporary rather than per manent ECA and pterygopalatine artery occlusion; 10 of 12 rats survived 24 h with an initial Grade 2 deficit that persisted in 7 and resolved to Grade I in 3 at the time of death, Rats subjected to 60 min of temporary ischemia had Grade 3 deficits and did not recover motor activity, reflexes, or adequate respi ratory function after reversal of anesthesia postop eratively, Histological studies obtained at death 24 h after 10-min temporary ischemia in Group I demon strated severe ischemic neuronal damage in the cor tex, hippocampus, and basal ganglia in seven rats and moderate damage in three (Fig,S).
FIG. 2, Autoradiograms from anesthetized nonsurgical con trol rats (Group A).
TEMPORARY COMPLETE CEREBRAL ISCHEMIA IN RATS
FIG. 3. Autoradiograms obtained after seven-vessel occlu sion (Group C), showing no uptake of radioisotope in the forebrain and cerebellum. Blood flow in the medulla is pre served.
DISCUSSION
CBF is often restored after ischemia during cir culatory arrest, temporary surgical cerebrovascular occlusion, and strokes followed by thrombolysis or development of compensatory collateral blood flow. An experimental model of temporary cerebral ischemia that causes an anatomically uniform re duction in CBF would therefore be useful for stud ies of the response to reperfusion after ischemia. For example, in rats, depending on the severity and duration of ischemia, delayed secondary reactions during reperfusion may prevent the restoration of energy metabolism and maintenance of cellular integrity (Yoshida et aI., 1985). However, using in vivo nuclear magnetic resonance spectroscopy to study rats subjected to temporary forebrain ischemia by the standard four-vessel occlusion method (Pulsinelli and Brierley, 1979), we found significant variability as well as fluctuation over time in the metabolite levels during both ischemia and reperfusion in individual rats (Andrews et aI., 1987; Richards et aI., 1987). The seven-vessel occlusion model reduces CBF
953
FIG. 4. Autoradiograms obtained after 15 min of temporary complete ischemia and 5 min of reperfusion (Group 0) d em onstrate reperfusion in the forebrain and cerebellum and several punctate areas of no-reflow. These sections showed no evidence of "bubble" artifact.
more completely and reproducibly than previous models. This should minimize individual variation in the metabolic response to temporary ischemia without altering other systemic or intracranial phys iological parameters. For example, the intracranial hypertension method (Ljunggren et aI., 1974) may cause unexpected hemodilution due to absorption of continuously infused artificial cerebrospinal fluid from the cisterna magna. Adding systemic hypoten sion to bilateral carotid artery occlusion (Kagstrom et al., 1983; Yoshida et al., 1985) may affect cardiac function and systemic pH; these factors may further alter brain metabolism during reperfusion. Applying a neck tourniquet (Nemoto et al., 1977) may cause systemic or intracranial hypertension or cervical vessel thrombosis that impairs reperfusion. Our method, a modification of the three-vessel occlu sion model described by Kameyama et al. (1985), more consistently produced uniform levels of com plete ischemia without significantly altering sys temic circulatory function. Since no CBF in the forebrain was detected autoradiographically in any rat subjected to seven-vessel occlusion, we assume
J Cereb Blood Flow Metab. Vol. ll. No.6. 1991
954
R. SHIRANE ET AL. (A)
FIG. 5. Photomicrographs of cortex (A) and CA1 sector of hippocampus (8) in 6-j-Lm brain sections stained with acid fuchsin and cresyl violet 24 h after 10 min of temporary complete ischemia. In A and B, note ischemic neuronal damage on the left compared with normal adjacent tissue on the right. Original magnification, x2S.
that ischemia was complete during occlusion even though very low levels of collateral flow are theo retically possible through distal muscular and meningeal branches of the occluded external caJ Cereb Blood Flow Metab, Vol. 11, No.6, 1991
rotid and pterygopalatine arteries. Previously re ported results with the three-vessel (bilateral CCA and basilar artery) occlusion method indicate that complete ischemia was documented in all rats stud-
TEMPORARY COMPLETE CEREBRAL ISCHEMIA IN RATS
ied only when neck ligation with a circumferential suture was added (Kameyama et aI., 1985). Prelim inary studies of the three-vessel occlusion model with in vivo nuclear magnetic resonance spectros copy demonstrated excessive variability in the re sponse of energy metabolism to ischemia and reper fusion. Although the seven-vessel occlusion method requires expertise in microvascular sur gery, an experienced researcher can complete the procedure in 20 min through a 3-cm midline incision with negligible blood loss. This approach obviates the need for the additional posterior cervical inci sion required for the four-vessel occlusion tech nique (Pulsinelli and Brierley, 1979). The flow pattern in the early reperfusion phase after 15 min of ischemia was predominantly that of postischemic hyperperfusion; in some areas, corti cal CBF increased by > 100% of control values. However, limitations of the autoradiographic method may reduce the quantitative accuracy of CBF measurements during reperfusion (Tomita and Gotoh, 1981). In another study (Kagstrom et aI., 1983), CBF autoradiography consistently showed postischemic no-reflow (Ames et aI., 1968) after> 10 min of com plete temporary forebrain ischemia produced by four-vessel occlusion, neck tourniquet, and hy potension in rats. In our study, only two of eight rats had small regions of no-reflow after 15 min of temporary ischemia. Experiments comparing these two models are needed to determine whether or not our results are related to use of the seven-vessel occlusion method, which produces more complete temporary global ischemia without systemic hy potension. Survival after permanent five-vessel occlusion and temporary CCA occlusion was severely limited by respiratory insufficiency caused by partial upper airway obstruction in our initial recovery experi ments. This did not occur in sham-operated rats or those subjected only to basilar artery occlusion and it was not prevented by administration of atropine to reduce secretions or by a trial of chloral hydrate anesthesia without endotracheal intubation. We therefore suspect that laryngeal edema due to per manent ECA and pterygopalatine artery occlusion was responsible. Use of temporary ECA and ptery gopalatine occlusion has eliminated this problem and must be used for survival studies with this model. Initial observations after recovery from anesthe sia did not demonstrate neurological impairment in rats subjected only to uncomplicated basilar artery (Group F) or five-vessel (Group G) occlusion. How ever, microsurgical exposure of the basilar artery
955
for coagulation and transection of the main trunk without occurrence of hemorrhage or injury to per forating branches is a technically precise procedure that requires practice but can be learned by inves tigators experienced in microsurgery in small ani mals. CBF in unoperated anesthetized controls (Group A) was similar to that in rats subjected to five-vessel occlusion only (Group B). From these data and neurological observations in Groups F and G, we conclude that if ventilation is adequate, col lateral blood flow is adequate and neurological function is normal after five-vessel or basilar artery occlusion. We therefore assume that cerebral ischemia is limited to the period of CCA occlusion. Observations in survival studies (Group I) indi cate that alert rats respond to visual and auditory stimuli, indicating that complete cephalic ischemia did not impair retinal or cochlear function. A 10-min temporary occlusion time was used in this group since only five of nine rats survived 24 h after 15min seven-vessel occlusion in preliminary studies. No therapeutic measures such as elevation of sys temic blood pressure or administration of mannitol and bicarbonate (Hossmann and Kleihues, 1973) were evaluated to enhance reperfusion in this study. In vivo magnetic resonance spectroscopy studies have shown that 15 min of ischemia induced by our seven-vessel occlusion model rapidly produces uni form and reversible lactic acidosis and depletion of phosphate energy metabolites (Shirane et aI., 1988; Chang et aI., 1990). The metabolic effects of reper fusion or other potentially protective manipulations as well as the delayed neurochemical reactions that occur after complete ischemia can be evaluated with this model. Acknowledgment: This work was supported by NIH grant ROJ NS 22022 and by the Veterans Administration Research Service. The authors thank Cheryl Christensen for manuscript preparation and Stephen Ordway for edi torial assistance.
REFERENCES Ames A III, Wright RL, Kowada M, Thurston JM, Majno G (1968) Cerebral ischemia. II. The no-reflow phenomenon. Am J Pathol 152:437-453 Andrews BT, Weinstein PW, Keniry M, Pereira B (1987) Se quential in vivo measurement of cerebral intracellular me tabolites with phosphorus-3l nuclear magnetic resonance spectroscopy during global cerebral ischemia and reperfu sion in rats. Neurosurgery 21:699-708 Chang LH, Shirane R, Weinstein PR, James TL (1990) Cerebral metabolite dynamics during temporary complete ischemia in rats monitored by time-shared lH and 31p NMR. Magn Res Med 13:6-13 Dietrich WD, Busto R, Yoshida S, Ginsberg MD (1987) Histo pathological and hemodynamic consequences of complete
J Cereb Blood Flow Metab, Vol. 11, No.6, 1991
956
R. SHIRANE ET AL. versus incomplete ischemia in the rat. J Cereb Blood Flow Metab 7:300-308
Hossmann K-A, Kleihues P (1973) Reversibility of ischemic brain damage. Arch Neurol 29:375-384 Kagstrom E, Smith ML, Siesj6 BK (1983) Local cerebral blood flow in the recovery period following complete cerebral ischemia in the rat. J Cereb Blood Flow Metab 3:170-182 Kameyama M, Suzuki J, Shirane R, Ogawa A (1985) A new model of bilateral hemispheric ischemia in the rat: three vessel occlusion model. Stroke 16:489-493 Ljunggren B, Ratcheson RA, Siesj6 BK (1974) Cerebral meta bolic state following complete compression ischemia. Brain Res 73:291-307 Nemoto EM, Bleyaert AL, Stezoski SW, Moossy J, Rao GR, Safar P (1977) Global brain ischemia. A reproducible mon key model. Stroke 8:558-564 Pulsinelli WA, Brierley JB (1979) A new method of bilateral hemispheric ischemia in the unanesthetized rat. Stroke 10:267-272 Rehncrona S, Westerberg E, Akesson B, Siesj6 BK (1982) Brain cortical fatty acids and phospholipids during and following complete and severe incomplete ischemia. J Neurochem 38:84--93 Rehncrona S, Mela L, Siesj6 BK (1987) Recovery of brain mi-
J Cereb Blood Flow Metab, Vol. 11, No.6, 1991
tochondrial function in the rat after complete and incomplete cerebral ischemia. Stroke 10:437-446 Richards TL, Keniry M, Weinstein PR, Pereira BM, Andrews BT, Murphy EJ, James TL (1987) Measurement of lactate accumulation by in vivo proton NMR spectroscopy during global cerebral ischemia in rats. Magn Res Med 5:353-357 Sakurada 0, Kennedy C, Jehle J, Brown JD, Carbin GL, Sokoloff L (1978) Measurement of local cerebral blood flow and 14C-iodoantipyrine. Am J Physiol 234:H59-H66 Shirane R, Chang LH, Weinstein PR, James TL (1988) Sequen tial changes of brain metabolism during temporary global ischemia in the rat detected by simultaneous IH and 31p magnetic resonance spectroscopy. In: Advances in Surgery for Cerebral Stroke (Suzuki J, ed), Tokyo, Springer-Verlag, pp 633-638 Steen PA, Michenfelder JF, Milde JH (1979) Incomplete versus complete cerebral ischemia: improved outcome with a min imal blood flow. Ann NeuroI6:389-398 Tomita M, Gotoh F (1981) Local cerebral blood flow values as estimated with diffusible tracers: validity of assumptions in normal and ischemic tissue. J Cereb Blood Flow Metab 1:403-411 Yoshida S, Busto R, Martinez E, Scheinberg P, Ginsberg MD (1985) Regional brain energy metabolism after complete ver sus incomplete ischemia in the rat in the absence of severe lactic acidosis. J Cereb Blood Flow Metab 5:490-501
