PROGRESS IN BRAIN RESEARCH VOLUME 63 MOLECULAR MECHANISMS OF ISCHEMIC BRAIN DAMAGE EDITED BY K. KOGURE Department of Neurology, Insiitute of Brain Diseases, Tohoku University School of Medicine, Sendai 980, Japan K.-A. HOSSMANN Max-Planck-Institute for Neurological Research, Department of Experimental Neurology, Cologne. FRG B.K.SIESJ~) Laboratory for Experimental Brain Research, Universiiy of Lund, Lund, Sweden and F.A. WELSH Division of Neurosurgery, Hospital of the University of Pennsylvania, Philadelphia, PA 19104, USA , ELSEVIER AMSTERDAM - NEW YOFX - OXFORD 1985 0 1985, Elsevier Science Publishers B.V. (Biomedical Division) All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise without the prior written permission of the publisher, Elsevier Science Publishers B.V./Biomedical Division, P.O. Box 1527, 1000 BM Amsterdam, The Netherlands. Special regulations for readers in the USA: This publication has been registered with the Copyright Clearance Center Inc. (CCC), Salem, Massachusetts. Information can be obtained from the CCC about conditions under which the photocopying of parts of this publication may be made in the USA. All other copyright questions, including photocopying outside of the USA, should be referred to the copyright owner, Elsevier Science Publishers B.V. (Biomedical Division) unless otherwise specified. ISBN 0-444-80654-7 (volume) ISBN 0-444-80104-9 (series) Published by: Elsevier Science Publishers B.V. (Biomedical Division) P.O. Box 211 1000 AE Amsterdam The Netherlands Sole distributors for the USA and Canada: Elsevier Science Publishing Company, Inc. 52 Vanderbilt Avenue New York, NY 10017 USA Library of Congress Cataloging in Publication Data Main entry under title: Molecular mechanisms of ischemic brain damage. (Progress in brain research ; v. 63) Proceedings of a meeting held in 1984. Includes bibliographies and index. 1. Cerebral ischemia--Congresses. 2. Pathology, Molecular--Congresses. I. Kogure, Kyuya. 11. Series. [DNLM: 1. Brain Damage, Chronic--metabolism--congresses. 2. Cerebral Ischemia--metabolism--congresses. W1 PR667J v.63 / WL 355 M718 19841 QP376.P7 VO~.6 3 [RC388.5] 61T.82 s 85-20714 ISBN 0-444-80654-7 (U.S.) [616.8'1] Printed in The Netherlands V List of Contributors K. Abe, Department of Neurology, Institute of Brain Diseases, Tohoku University School of Medicine, 1-1 Seiryo-machi, Sendai 980, Japan D. K. Anderson, The VA Medical Center, Cincinnati, OH, USA H. Arai, Department of Neurology, Institute of Brain Diseases, Tohoku University School of Medicine, 1-1 Seiryo-machi, Sendai 980, Japan A. Barbier, Max-Planck-Institute for Neurological Research, Department of Experimental Neurology, Ost- merheimerstrasse 200, D-5000 Koln 91, FRG M. Baudry, Center for the Neurobiology of Learning and Memory, U.C., Irvine, CA 92717, USA W. Bodsch, Max-Planck-Institute for Neurological Research, Department of Experimental Neurology, Ost- merheimerstrasse 200, D-5000 Koln 91, FRG T.M. Buckholz, Department of Neurology, Yale University School of Medicine, New Haven, CT 06510, USA P. H. Chan, Brain Edema Clinical Research Center, Department of Neurology, University of California School of Medicine, San Francisco, CA 94143, USA S. Chen, Brain Edema Clinical Research Center, Department of Neurology, University of California School of Medicine, San Francisco, CA 94143, USA J.H. Chin, Department of Neurology, Yale University School of Medicine, New Haven, CT 06510, USA N.R. Clandenon, Department of Physiological Chemistry and Neurology, 214 Hamilton Hall, 1645 Neil Avenue, The Ohio State University, Columbus, OH 43210, USA R. CortBs, Preclinical Research, Sandoz Ltd., 386/216, CH-4002 Basle, Switzerland R.J. DeLorenzo, Department of Neurology, Yale University School of Medicine, New Haven, CT 06510, USA P. Demediuk, Department of Physiological Chemistry, 214 Hamilton Hall, 1645 Neil Avenue, The Ohio State University, Columbus, OH 43210, USA R.A. Fishman, Brain Edema Clinical Research Center, Department of Neurology, University of California School of Medicine, San Francisco, CA 94143, USA B. Grosse Ophoff, Max-Planck-Institute for Neurological Research, Department of Experimental Neurology, Ostmerheimerstrasse 200, D-5000 Koln 91, FRG K.-A. Hossmann, Max-Planck-Institute for Neurological Research, Department of Experimental Neurology, Ostmerheimerstrasse 200, D-5000 Koln 91, FRG L.A. Horrocks, Department of Physiological Chemistry, 214 Hamilton Hall, 1645 Neil Avenue, The Ohio State University, Columbus, OH 43210, USA Y. Inaba, Department of Neurosurgery, Tokyo Medical and Dental University, 1-5-45 Yushima Bunkyo-ku, Tokyo 113, Japan M. Karobath, Preclinical Research, Sandoz Ltd., 386/216, CH-4002 Basle, Switzerland T. Kirino, Department of Neurosurgery, Teikyo University School of Medicine, 2-1 1-1 Kaga, Itabashi-ku, Tokyo 173, Japan K. Kogure, Department of Neurology, Institute of Brain Diseases, Tohoku University School of Medicine, 1-1 Seiryo-machi, Sendai 980, Japan R.P. Kraig, Cerebrovascular Disease Research Center, Department of Neurology, Cornell University Medical College, 1300 York Avenue, New York, NY 10021, USA S. Longar, Brain Edema Clinical Research Center, Department of Neurology, University of California School of Medicine, San Francisco, CA 94143, USA G. Lynch, Center for the Neurobiology of Learning and Memory, U.C., Irvine, CA 92717, USA E.D. Means, The VA Medical Center, Cincinnati, OH, USA M. Nakano, College of Medical Care and Technology, Gunma University, 3-39- 15 Showa-machi, Meabashi 371, Japan D.G. Nicholls, Neurochemistry Laboratory, Department of Psychiatry, Ninewells Medical School, University of Dundee, Dundee DDI 9SY, Scotland, UK J.M. Palacios, Preclinical Research, Sandoz Ltd., 386/216, CH-4002 Basle, Switzerland F. Plum, Cerebrovascular Disease Research Center, Department of Neurology, Cornell University Medical College, 1300 York Avenue, New York, NY 10021, USA W.A. Pulsinelli, Cerebrovascular Disease Research Center and Department of Neurology, Cornell University Medical College, 1300 York Avenue, New York, NY 10021, USA K. Sano, Department of Neurosurgery, Teikyo University School of Medicine, 2-1 1-1 Kaga, Itabashi-ku, Tokyo 173, Japan VI R.D. Saunders, Department of Physiological Chemistry, 214 Hamilton Hall, 1645 Neil Avenue, The Ohio State University, Columbus, OH 43210, USA W.W. Schlaepfer, Division of Neuropathology, Department of Pathology and Laboratory Medicine, Uni- versity of Pennsylvania Medical School, Philadelphia, PA 19104, USA B.K. Siesjo, Laboratory for Experimental Brain Research, University of Lund, Floor EA-5, Lund Hospital, S-221 85 Lund, Sweden P. Supavilai, Preclinical Research, Sandoz Ltd., 386/216, CH-4002 Basle, Switzerland R. Suzuki, Department of Neurosurgery, Tokyo Medical and Dental University, 1-5-45 Yushima Bunkyo- ku, Tokyo 113, Japan K. Takahashi, Max-Planck-Institute for Neurological Research, Department of Experimental Neurology, Ostmerheimerstrasse 200, D-5000 Koln 91, FRG A. Tamura, Department of Neurosurgery, Teikyo University School of Medicine, 2- 1 1-1 Kaga, Itabashi-ku, Tokyo 173, Japan H.G. Wagner, Laboratory of Neuropathology and Neuroanatomical Sciences, NINCDS, National Institutes of Health, USA F.A. Welsh, Division of Neurosurgery, Hospital of the University of Pennsylvania, Philadelphia, PA 19104, USA T. Wieloch, Laboratory for Experimental Brain Research, University of Lund, S-221 85 Lund, Sweden T. Yamaguchi, Department of Neurosurgery, Tokyo Medical and Dental University, 1-5-45 Yushima Bunk- yo-ku, Tokyo 113, Japan A. Yu, Brain Edema Clinical Center, Department of Neurology, University of California School of Medicine, San Francisco, CA 94143, USA U.-J.P. Zimmerman, Division of Neuropathology, Department of Pathology and Laboratory Medicine, Uni- versity of Pennsylvania Medical School, Philadelphia, PA 19104, USA VII Preface The clinical importance of cerebral ischemia is reflected in a large and constantly grow- ing literature, dealing with both, clinical and experimental aspects of ischemic states. For a number of years most publications, and virtually all symposia and workshops devoted to the subject, have dealt with the relationships between reduction in blood flow, metabolic perturbation, and functional or structural brain damage. By necessity, the bulk of the new information has been of a descriptive nature. During the last few years, research within the field has been partly redirected. The most important change is the emphasis on mechanisms of ischemic damage, and par- ticularly on its cellular and molecular aspects. It was in attempt to focus attention on this new, rapidly developing, and exciting research that the Sendai Forum '84 was or- ganized. The topics chosen were those in which development is now rapid. Our starting point was the widely recognized but poorly understood phenomenon of selective neu- ronal vulnerability. In an attempt to explain this selectivity, as contrasted to necrotic destruction of glial cells and vascular endothelium as well, the program was centered on a perturbed metabolism of excitatory transmitters, of calcium, and of H+, and also on important target molecules such as proteins and lipids. The organization of this symposium was made possible by a generous grant from A.G. Sandoz in Basel, and supported by the Southern Tohoku Research Institute for Neurosciences. K. Kogure K.-A. Hossmann B.K. Siesjo F.A. Welsh K. Kogure. K.-A. Hossmann. 8. K. Siesjo and F. A. Welsh (Eds.), Progress in Brain Research, Vol. 63. 0 1985 Elsevier Science Publishers B.V. (Biomedical Division) 3 Post-ischemic resuscitation of the brain: selective vulnerability versus global resistance K.-A. Hossmann Max-Planck-Institute for Neurological Research, Department of Experimental Neurology, Ostmerheimer Str. 200. 0-5000 Koln 91, FRG Introduction If this hypothesis is true, prolongation of the re- vival time of the brain should be possible under ex- Despite considerable progress in the understanding perimental conditions in which the no-reflow phe- of the pathophysiology of cerebral ischemia, the nomenon is prevented. We have tested this hypoth- determination of the revival time of the brain re- esis in our laboratory. It turned out in fact that mains a topic of great controversy. In the classic appropriate treatment of post-ischemic recircula- literature animal experiments and clinical observa- tion disturbances led to recovery of energy metab- tions of cardiac arrest suggested that cerebral ische- olism and neuronal excitability after complete cer- mia of more than 4-5 minutes results in irreversible ebro-circulatory arrest of as long as 1 hour at nor- brain damage. This view was challenged by Hirsch mal body temperature (Hossmann and Sato, et al. (1957) who observed recovery of EEG activity 1970a,b). Systematic exploration of the limits of after 8-10 minutes of ischemia when cerebral blood central nervous system revivability further revealed flow was arrested without interfering with cardiac that functions of considerable complexity such as function, i.e. by inflating a pneumatic cuff around protein synthesis (Kleihues and Hossmann, 1971, the animal’s neck. They concluded that the shorter 1973) or EEG activity (Hossmann and Zimmer- revival time of the brain after cardiac arrest was mann, 1974) returned, and that - occasionally - due to the fact that the recovery time of the heart even restitution of neurological function occurred had to be added to the ischemia of the brain. In after these long periods of ischemia (Hossmann et 1968, Ames et al. discovered that even without car- al., in preparation). diac insufficiency post-ischemic recirculation is a However, neurological recovery is not predict- limiting factor of cerebral recovery. With the same able after prolonged ischemia, and an increasing tourniquet model as previously used by Hirsch and number of delayed post-ischemic disturbances have colleagues (1957), they noticed that after more than been identified. These include post-ischemic hypo- 7.5 minutes of ischemia a substantial volume of the perfusion and dysregulation of blood flow (Hoss- brain was not recirculated, a phenomenon they mann et al., 1973; Nemoto et al., 1975), permeabil- called “no-reflow”. They also observed that evoked ity changes of the blood-brain barrier (It0 et al., potentials of isolated retina, maintained in vitro, 1976; Petito et al., 1982), calcium-mediated pro- recovered after much longer periods of oxygen and cesses (Hossmann et al., 1983; Dienel, 1984; Simon glucose deprivation than in the living animal after et al., 1984b), and even secondary depression of circulatory arrest (Ames and Gurian, 1963). They previously recovered metabolic activity (Mies et al., therefore speculated that recirculation disturbances 1983; Pulsinelli and Duffy, 1983; Bodsch and Tak- may be responsible for the short revival time of the ahashi, 1984). Ito et al. (1925) coined the term brain. “maturation phenomenon” for the gradual devel- 4 opment of such disturbances. It was also pointed with selective vulnerability are discussed in detail in out that the delay of “maturation” increased in- this volume by several authors (Bodsch et al., 1985; versely with the duration of ischemia (It0 et al., Kirino et al., 1985; Pulsinelli, 1985b; Wieloch, 1985; 1979, and that in certain “selectively vulnerable Siesjo, 1985; Suzuki et al., 1985). It is, therefore, areas” of hippocampus delayed neuronal death sufficient to mention that the following regions are could occur as long as 4 days after an ischemic generally considered to be selectively vulnerable to period of as short as 5 minutes (Kirino, 1982; Su- ischemia: the limbic system, in particular pyramidal zuki et al., 1983a). cells of CA 1 subfield of hippocampus, Purkinje cells These observations caused the pendulum to of cerebellum, small and medium-sized neurons of swing back, and focus interest more and more on striatum, and layers 3, 5 and 6 of cortex. The lower the deleterious effects of short-lasting ischemia. limit of selective injury to CA1 neurons or Purkinje However, in many recent studies little attention has cells is less than 5 minutes (Kirino, 1982; Gurvitch been given to post-ischemic complicating side fac- et al., 1972); however, the phenomenon of selective tors, and methods are again being used for the pro- vulnerability has also been studied after longer duction of ischemia which previously have been periods of ischemia such as 10 min tourniquet shown to delay or prevent the post-ischemic recov- ischemia (Diemer and Siemkowicz, 1981a) or 30 ery process. These include tourniquet ischemia or min four-vessel occlusion in the rat (Pulsinelli et al., severe incomplete ischemia. Studying mechanisms 1982a). of ischemic brain damage in such models makes it In acute experiments, identification of selective extremely difficult to differentiate between phenom- vulnerability requires certain methodological pre- ena caused by the primary ischemic impact from cautions. As pointed out by Klatzo and co-workers, those resulting from unsuccessfully treated post- the histological manifestation or “maturation” of ischemic complicating side effects. Similarly, the cell damage may require up to several days, the mechanisms responsible for selective vulnerability shorter or milder the ischemia the longer the inter- should be clearly differentiated from those occur- val (It0 et al., 1975; Bubis et al., 1976). Similarly, ring in the more resistant parts of the brain. This certain epiphenomena of irreversible cell damage differentiation is of particular importance for the such as reduction of energy-producing metabolism establishment of therapeutic procedures because (Mies et al., 1983) or breakdown of the blood-brain the same approach which may help to ameliorate barrier (It0 et al., 1976) appear after a “matura- damage in the selectively vulnerable areas might tion” interval which may be longer than the usual further impair the lesion in the resistant regions, duration of an acute animal experiment. For this and vice versa. At the present, many of the mech- reason, failure to detect histological or biochemical anisms discussed are controversial, and there are lesions shortly after ischemia does not exclude the few studies in which a distinct differentiation be- possibility that such changes may develop at a later tween vulnerable and resistant areas has been made. time. In the following, the available information is re- The clinical symptoms of injury to the selectively viewed and discussed with respect to this particular vulnerable brain regions have been poorly investi- problem . gated. Isolated lesions of hippocampal subfield CAI do not seem to provoke any major deficits, at Limits of reversibility least not in lower mammals such as gerbils and rats. Bothe et al. (1983) noticed a slight deterioration of (a) Selectively vulnerable areas avoidance reaction during the maturation phase which disappeared by the time although histologi- The topographical pattern and the morphological, cal lesions became manifest. Similarly, Volpe et al. biochemical and functional processes associated (1984) observed only a slight impairment of learn- 5 ing behavior after selective lesion of CAI and stria- tions are those in which the progression of clinical tum. Injury of the selectively vulnerable areas, in recovery has been followed, using intensive care for consequence, is compatible with life and seems to the prevention of complicating factors. In such produce relatively mild neurological deficits. studies complete neurological recovery consistently occurred after 12-1 5 rnin cerebro-circulatory arrest (6) Resistant areas of the brain (Crowell and Smith, 1956; Safar et al., 1976; Nem- oto et al., 1977). However, occasionally complete All regions which are not selectively vulnerable, recovery or recovery with minor neurological defi- may be tentatively defined as “potentially resist- cits was reported after 15-18 rnin (Gilston, 1979), ant”. As has been pointed out above, the no-reflow after 2&24 min (Miller and Myers, 1970), after 28 phenomenon may prevent recovery after ischemia min (Makarenko, 1972), after 30 rnin (Volpe et al., as short as 7.5 min but appropriate treatment of 1984) or even after 60 rnin ischemia (Hossmann et this complication can result in progressive restora- al., in preparation). In the latter series of experi- tion of complex brain functions after complete cer- ments the classical pattern of hippocampal injury ebro-circulatory arrest of up to 1 hour:The limits was present but viable neurons were detected in of reversibility in resistant areas, in consequence, most Other regions of the brain including layers 3 can only be explored after optimizing treatment of and 5 of cortex and the Purkinje cell layer of cere- post-ischemic complicating side effects and, there- bellum (Kleihues and Hossmann, 1973). Global re- fore, require special technical precautions (see be- sistance of the brain to ischemia, in consequence, low). Whenever these are taken, and when the core exceeds by far the lower limits of damage in the temperature and the density of ischemia are care- selectively vulnerable regions. In the following, I fully documented to make sure that ischemia is shall discuss how far known mechanisms of ische- complete and normothermic, positive evidence of mic brain damage may account for this difference. post-ischemic recovery is more meaningful than demonstration of its absence. Mechanisms of ischemic brain damage Another factor which has to be taken into ac- count, is the latency of the recovery. On the one (a) Factors related to disturbances of recirculation hand, the latency of post-ischemic recovery depends on the duration of ischemia and, on the other, on Following a period of ischemia two types of recir- the complexity of the biochemical or functional culation disturbances can be distinguished: the no- process. Shortlasting experiments, therefore, bear reflow phenomenon and the (delayed) post- the risk that a lesion may be erroneously considered ischemic hypoperfusion syndrome. The no-reflow to be irreversible because the latency period for re- phenomenon (Ames et al., 1968) is the combined re- covery has not elapsed. However, transient recov- sult of increased viscosity of (stagnant) blood ery during maturation of an irreversible lesion is (Fischer, 1973), microcirculatory compression by also possible and in this case the demonstration of swollen perivascular glial cells (Arsenio-Nunes et recovery may be erroneously interpreted as an in- al., 1973), formation of endothelial microvilli (Diet- dicator of post-ischemic revival, although the same rich et al., 1984), increased intracranial pressure function may disappear at a later time. (Zimmermann et al., 1975), post-ischemic hypoten- With these considerations in mind, it becomes sion (Cantu et al., 1969) and disseminated intra- understandable that the reported revival times of vascular coagulopathy (Hossmann and Hossmann, various brain functions may vary considerably 1977). No-reflow is a limiting factor for post- from one laboratory to another, and that the dis- ischemic resuscitation in most regions of the brain, crepancies are more pronounced the more complex as evidenced by the close correlation of this distur- the function (Table 1). The most relevant observa- bance with histological lesions (Ginsberg and 6 TABLE 1 Revival times* of brain functions after complete ischemia in normothermia Function Method Species Duration Reference (min) Energy-producing metabolism 4-vessel occlusion rat 30 Name et a]., 1984 CSF compression rat 30 Nordstrom et al., 1978 decapitation guinea pig 45 Okada, 1974 decapitation rat 60 Ikr6nyi et al., 1976 arterial inflow occlusion cat 60 Hossmann et al., 1976 Protein biosynthesis isolated retina rabbit 20 Ames and Nesbett, 1983a 4-vessel occlusion rat 30 Dienel et al., 1980 arterial inflow occlusion cat 60 Kleihues and Hossmann, 1971 Evoked potentials decapitation guinea pig 45 Okada, 1974 compression ischemia retina rabbit 60 Foulds and Johnson, 1974 isolated head dog 60 Sobotka and Gebert, 1971 arterial inflow occlusion cat 60 Hossmann et al., 1983 EEG activity occlusion aorta cat 15 Ten Cate and Horsten, 1952 cardiac arrest cat 30 Hossmann and Hossmann, 1973 isolated head dog 30 Hirsch et al., 1975 arterial inflow occlusion monkey 60 Hossmann and Zimmerrnann, 1974 Clinical recovery cardiac arrest dog 10-15 Crowell and Smith, 1956 cardiac arrest dog 12 Safar et al., 1976 tourniquet ischemia monkey 15 Nemoto et al., 1977 exsanguination man 15-18 Gilston, 1979 systemic circulatory arrest monkey 20-24 Miller and Myers, 1970 drowning dog 28 Makarenko, 1972 4-vessel occlusion rabbit 30 Kolata, 1979 arterial inflow occlusion cat 60 Hossmann et al., in preparation * Revival time is defined as the longest duration of ischemia compatible with recovery of the function under investigation. Myers, 1972). The extent of no-reflow depends on chemia is longer than 10 min (Kigstrom et al., the type and duration of ischemia. It increases with 1983). time, and seems to be most pronounced when brain In our experience homogeneous blood reperfu- vessels are filled with blood, i.e. after incomplete sion after prolonged ischemia is possible only when ischemia or when the venous outflow is obstructed. special methodological requirements are fulfilled. This is the reason that after only 15 min of tour- Ischemia must be induced by arterial inflow occlu- niquet ischemia recirculation may fail in up to 95% sion and it must be complete. In large laboratory of brain volume (Ames et al., 1968) -because this animals like cats and monkeys, intrathoracal type of ischemia is both incomplete and stagnant. clamping of innominate, left subclavian and both However, even without venous obstruction small mammary arteries under induced hypotension is areas of no-reflow are a consistent finding when is- best suited for this purpose (Hossmann and Zim- I mermann, 1974). Furthermore, recirculation must flow are still present. This may explain why several be initiated by a hypertensive “flush” in order to authors have observed irreversible brain damage reach the critical capillary opening pressure (Fisch- although blood flow was not reduced (Harrison et er et al., 1979), and osmolality of blood must be al., 1975; Levy et al., 1975). On the other hand, the increased for prevention of osmotic influx of fluid degree of hyperemia is a function of vascular pat- from the blood into the brain (Hossmann and Tak- ency; it is therefore not surprising that functional agi, 1976). In our experiments of up to 1-h ischemia recovery after ischemia proceeds most rapidly in we raise systolic arterial pressure with catechol- those cases in which reactive hyperemia is most pro- amines to about 180 mm Hg immediately before nounced (Hossmann et al., 1973). releasing the vessel clamps, and increase blood 0s- It should be noted that prevention of no-reflow molality by about 30 mOsm with hypertonic solu- promotes recovery only in the resistant but not in tions, also before the beginning of recirculation. the selectively vulnerable areas of the brain. The Finally, because disseminated post-ischemic coagu- no-reflow phenomenon, in consequence, is not a lation and post-ischemic hypotension are enhanced limiting factor for ischemic damage in these re- at low arterial pH, the acid equivalents released gions. This conclusion is supported by the obser- during the early recirculation period must be buf- vation that selective vulnerability is seen after fered by controlled infusion of Tris buffer or so- ischemia of only 5 min, i.e. after a period which is dium bicarbonate, and arterial pCOz must be kept too short to evoke a no-reflow phenomenon. close to normal by controlled ventilation. For this reason blood gas analyses must be carried out at The post-ischemic hypoperjiision syndrome is a con- short intervals, and endtidal COz must be moni- sistent complication of post-ischemic recirculation tored continuously. With this procedure recircula- even under those conditions in which a no-reflow tion rates up to 350% of control have been ob- phenomenon is absent or in which it has been suc- served after 30 min and up to 250% of control after cessfully treated, and it develops after the phase of 60 min global ischemia (Hossmann et al., 1973). reactive hyperemia has ceased (Hossmann et al., After shorter periods of ischemia adequate recir- 1973; Nemoto et al., 1975; Miller et al., 1980). The culation may be obtained with less drastic therapy degree of post-ischemic hypoperfusion seems to be but it is unlikely that a no-reflow phenomenon can the more pronounced the shorter the duration of bd prevented without vigorous prophylaxis when ischemia: after 5 min ischemia in gerbils flow de- ischemia exceeds a duration of more than 10 min. creased to about 30% (Cahn et al., 1985), after 30 Once the circulation has been reinitiated, reactive min ischemia in rats to 50% (Pulsinelli et al., 1982b) hyperemia continues at normal blood pressure be- and after 60 min ischemia in cats to 70% of control cause the blood viscosity of streaming blood ab- (Van den Kerckhoff et al., 1983). Post-ischemic hy- ruptly decreases (Schmid-Schonbein, 1977) and be- poperfusion affects both the vulnerable and the re- cause vascular tone is temporarily reduced after is- sistant areas of the brain (Pulsinelli et al., 1982b). chemia (Takagi et al., 1977). Reoxygenation of the It is characterized by a dissociation between sup- brain results in rapid resolution of post-ischemic pressed COz reactivity and maintained autoregu- brain swelling and normalization of intracranial lation (Hossmann et al., 1973; Nemoto et al., 1975) pressure (llossmann, 1976). However, since the no- leading to vascular constriction (Siernkowicz, 1980; -reflow phenomenon is heterogenous, postischemic White et al., 1983), and uncoupling of blood flow reactive hyperemia is heterogenous as well (Gins- and metabolic activity (Levy and Duffy, 1977; berg and Myers, 1972; KBgstrom et al., 1983). Mea- Hossmann, 1979). Post-ischemic respiratory insuf- surement of global cerebral blood flow after is- ficiency aggravates the effects of hypoperfusion be- chemia, in consequence, may reveal normal or even cause it further reduces oxygen availability (Hoss- increased flow rates although focal areas of no-re- mann and Hossmann, 1977). It cannot be amelior-