jurnals

Review Article

Heterogeneity in the penumbra

Gregory J del Zoppo1,2*, Frank R Sharp3,4, Wolf-Dieter Heiss5 and Gregory W Albers6

1Department of Medicine (Division of Hematology), University of Washington School of Medicine, Seattle,Washington, USA; 2Department of Neurology, University of Washington School of Medicine, Seattle,Washington, USA; 3Department of Neurology, MIND Institute, University of California at Davis MedicalCenter, Sacramento, California, USA; 4Department of Genetics, MIND Institute, University of Californiaat Davis Medical Center, Sacramento, California, USA; 5Max Planck Institute for Neurological Research,Cologne, Germany; 6Department of Neurology and Neurological Sciences, Stanford University MedicalCenter, Palo Alto, California, USA

Original experimental studies in nonhuman primate models of focal ischemia showed flow-relatedchanges in evoked potentials that suggested a circumferential zone of low regional cerebral bloodflow with normal K+ homeostasis, around a core of permanent injury in the striatum or the cortex.This became the basis for the definition of the ischemic penumbra. Imaging techniques of the timesuggested a homogeneous core of injury, while positing a surrounding ‘penumbral’ region thatcould be salvaged. However, both molecular studies and observations of vascular integrity indicatea more complex and dynamic situation in the ischemic core that also changes with time. Themicrovascular, cellular, and molecular events in the acute setting are compatible with heterogeneityof the injury within the injury center, which at early time points can be described as multiple ‘mini-cores’ associated with multiple ‘mini-penumbras’. These observations suggest the progression ofinjury from many small foci to a homogeneous defect over time after the onset of ischemia. Recentobservations with updated imaging techniques and data processing support these dynamicchanges within the core and the penumbra in humans following focal ischemia.Journal of Cerebral Blood Flow & Metabolism (2011) 31, 1836–1851; doi:10.1038/jcbfm.2011.93; published online 6 July 2011

Keywords: focal ischemia; imaging; ischemic penumbra; metabolic characteristics; microvessel characteristics;molecular characteristics

Introduction

Occlusion of a brain-supplying artery for an exten-ded period of ischemia leads to permanent injury ifreturn of flow is inadequate. Astrup et al (1977, 1981)posited the development of a core of injury destinedfor tissue destruction (infarction), surrounded by a‘penumbra’ of metabolically metastable tissue thathas the potential for full recovery (Figure 1A)(Branston et al, 1974, 1977). This general depiction

of the ‘penumbra’ has greatly influenced the conceptof ischemic stroke following thrombotic and throm-boembolic occlusion of brain-supplying arteries, andour attempts to enhance tissue recovery clinically.However, evolving data regarding cellular andmolecular responses to ischemia, the interrelation-ships of vascular and neuronal activation (the‘neurovascular unit’), experience with models of focalischemia, and progress in acute imaging techniquesin both animals and patients have suggested impor-tant refinements to the ‘penumbra’ concept.

Although the original conceptualization of thepenumbra remains solid, newer data that take intoaccount the time course of injury development,potential interactions of boundary zones to the core,and the vascular ‘territory at risk’ within the strickenhemisphere may provide even greater relevanceto patient outcome than indicated by many acuteintervention clinical trials. Recent work is beginningto address these issues and could point the way tofurther constructive developments in brain injurymanagement.

Received 7 January 2011; revised 8 May 2011; accepted 9 May2011; published online 6 July 2011

*Correspondence: Dr GJ del Zoppo, MD, Department of Medicine (inHematology), Department of Neurology, University of WashingtonSchool of Medicine, Box 359756 at Harborview Medical Center, 325Ninth Avenue, Seattle, WA 98104, USA.E-mail: [email protected]

The work reported here has been supported in part by NIH grants

NS 053716, NS 038710, and NS 036945 to Dr del Zoppo, NIH

grants NS066845 and NS054652 to Dr Sharp, and NIH grants

NS39325 and NS044848 to Dr Albers, and funding by the WDH

Foundation to Dr Heiss.

Journal of Cerebral Blood Flow & Metabolism (2011) 31, 1836–1851& 2011 ISCBFM All rights reserved 0271-678X/11 $32.00

www.jcbfm.com

http://dx.doi.org/10.1038/jcbfm.2011.93

mailto:[email protected]

http://www.jcbfm.com

Evolving Definitions of the Penumbra

Early approaches to modeling focal ischemia inmammals sought to develop discrete regions of neuroninjury by manipulating the cerebral vascular supply.As model development moved toward the moreclinically relevant occlusion of the middle cerebralartery (MCA) in the nonhuman primate, descriptions ofneuron function and integrity were linked to regionalchanges in regional cerebral blood flow (rCBF). Bynecessity, the concept of the ‘penumbra’ has dependedin a large part on the techniques used to visualize it.

From the notion of a circumferential region ofmetabolic disturbance that has not reached thethreshold of permanent injury, the concept hasevolved to include (1) characteristic electrophysio-logical changes, (2) biochemical/molecular altera-tions, (3) responses of the microvasculature, (4)metabolic changes defined by imaging methods,

and (5) observed differences between regions ofabnormal tissue perfusion and diffusion regions onmagnetic resonance imaging (MRI) studies.

In the original electrophysiological concept, the‘ischemic penumbra’ was an area that becameelectrically silent, but could recover function if therCBF was restored in time (Heiss, 1992; Hossmann,1994). An operational definition states that thepenumbra ‘is ischemic tissue which is functionallyimpaired and is at risk for infarction, but has thepotential to be salvagedy. If not salvaged this tissueis progressively recruited into the infarct core, whichwill expand with time into the maximal volumeoriginally at risk (Baron, 1999)’. Imaging of thepenumbra, using positron emission tomography(PET) has allowed measurement of rCBF, cerebralmetabolic rate for oxygen (CMRO2), rCMRglc (regionalcerebral metabolic rate of glucose), and oxygenextraction fraction (OEF). Infarction usually corres-ponds to rCBF decreased below 12 mL/100 g per minand rCMRO2 below 65 mm/100 g per min, whereas thepenumbra has been defined as rCBF decreased to12 to 22 mL/100 mg per min, the rCMRO2 above65 mm/100 g per min, and OEF increased to 50% to90% (Heiss, 2000). The concept of a mismatchbetween areas of blood perfusion and H2O diffusionin the injured territory as revealed by high field-strength magnetic MRI has provided a possible wayto image the ‘penumbra’ in humans.

Using animal models and methods with increasedresolution, alterations in protein synthesis occurwithin and around the regions of injury. In regionsof decreased rCBF, molecular correlates of thepenumbra are: (1) decreased protein synthesis thatcan recover, (2) preserved ATP, (3) synthesis of heat-shock proteins (Hsps), and possibly (4) a successfulunfolded protein response (Sharp et al, 2000). Ifthese molecular criteria are met, the tissue has thecapacity to be salvaged if blood flow is restored.

Once considered to be inert conduits, the micro-vasculature has been shown to respond rapidly anddynamically within the ‘territory at risk’, containingthe ischemic core and regions peripheral to the core.Microvessel adhesion receptor and basal laminamatrix expression can present as ‘mini-cores’ and‘mini-penumbras’ within the early evolving core ofneuron injury, after occlusion of a brain-supplyingartery, in a characteristic time-dependent manner(Tagaya et al, 2001). Given that the microvasculature,which in capillaries consists of the endothelium–basal lamina matrix–astrocyte end-feet complex,serves neurons (as part of a hypothetical ‘neuro-vascular unit’), changes in microvessels and neuronsmust to be connected locally.

Each of these definitions describes the events/conditions in regions of focal ischemic injury withinthe ‘territory at risk’. Not considered by thesedefinitions is whether the processes that determinethe ‘penumbra’ or ‘core’ occur uniformly in adjacentareas over time and whether the active conversion ofthe penumbra to core occurs at the interface of the

Figure 1 The ‘penumbra’. (A) Depiction of ‘the penumbra’ basedon rCBF and electrophysiological studies on the nonhumanprimate (from Symon (1980), with permission). (B) Depiction ofthe relationship among rCBF, electrical failure in the dependentcerebral tissue, and ischemia (taken from a figure in the study byAstrup et al (1977), with permission). rCBF, regional cerebralblood flow.

Heterogeneity in the penumbraGJ del Zoppo et al

1837

Journal of Cerebral Blood Flow & Metabolism (2011) 31, 1836–1851

core. Recent data have suggested that the core ofischemic injury develops heterogeneously, and withtime coalesces dynamically into a homogenous core.These data are consistent with the original conceptsof the penumbra and core, but recognize the dynamiccomplex heterogeneous processes involved. Thisrefinement has potential therapeutic implications.

The Ischemic Penumbra and theIschemic Core

The ‘ischemic penumbra’ was initially defined byphysiologic experiments performed in large animalssubjected to focal cerebral ischemia as (1) a regionaround a central core of low rCBF and physiologicsilence, (2) a boundary region with release ofintracellular K+, and (3) a region of metabolicinstability that could be recovered.

Symon et al (1980) described the penumbra as atissue of evolving injury around a homogeneouscentral core destined for infarction. Symon et al(1974) showed that 3 years after the rapid reductionin rCBF in the MCA territory in the nonhumanprimate (Papio sp), a residual core of low rCBFremained that was congruent with the acute CBFreduction. Regions of CBF peripheral to the ‘core’had returned to normal levels. Branston et al (1974),using the same nonhuman primate model, relatedthe loss in evoked potential (EP) amplitude tosignificant reductions in local CBF. Here, the rCBFthreshold for 50% reduction in EP amplitude wasB18 mL/100 g per min, when regional flow wasreduced by phlebotomy and induced hypotension.

The functional threshold for neurologic dysfunc-tion progressed from mild paresis at 22 mL/100 g permin to complete paralysis at 8 mL/100 g per min.Hence, the presence of a boundary zone in which theEP amplitude was sensitive to blood pressure/flowcould be defined. Reduction in EP amplitudecorrelated with cellular K+ release into the surround-ing tissue. When measured with H2 clearanceelectrodes, rCBF reduction could be observed atwhich the EP amplitude remained at 50% normaland K+ extravasation occurred in the ‘core’, but notin the peripheral zones (Symon et al, 1974; Branstonet al, 1977). Further studies have shown a differentialsensitivity of K+ and Ca2 + extravasation when rCBFranged from 8 to 15 mL/100 g per min (Harris andSymon, 1984); EP amplitude decreased when rCBFdropped below 15 mL/100 g per min even whenK+ extravasation was not observed (Branston et al,1977). On the basis of model studies, flow rates of12 mL/100 g per min lasting for 2 to 3 hours led tolarge infarctions, but individual cells became necro-tic after shorter intervals of time and at higher levelsof residual flow. The potential for irreversibledamage or recovery then seemed determined notonly by the level of residual flow but also by theduration of flow disturbance for a given region.

These observations form the basis for the familiartemporal relationship among rCBF, electrical function,and brain tissue integrity characterizing regions ofpermanent injury and of reversible injury fromischemia onset (Figure 1B) (Jones et al, 1981). Thissynthesis implies that the core is homogeneous, andexpands as a wave front with time into a penumbra oftissue that is at least initially not doomed to infarction.The range of perfusion between those limits—a rCBFlevel below which neuronal function is impaired and alower threshold below which irreversible membranefailure and morphologic damage occur—typifies the‘ischemic penumbra’ (Astrup et al, 1981).

Separately, experiments using multiple electrodes incats subjected to focal cerebral ischemia, which simul-taneously assessed single-cell activity and local flow,showed pockets of low flow and cortical neuronvulnerability in closely adjacent cortical areas (Rosneret al, 1986). The altered single-cell activity with groupedor regular discharges at flow levels just above thethreshold correlated with the gradual appearance offunctional deficits (Figure 2). Spontaneous neuronalactivity and EPs were restored when blood flow wasreestablished (Heiss et al, 1976; Heiss and Rosner, 1983).

Metabolic Characteristics of thePenumbra and Core

Energy demands of the central nervous system arehigh and almost entirely provided by oxidative

Figure 2 Response of tissue to the duration of rCBF reduction,and development of the ‘penumbra’. Diagram of CBF thresholdsrequired for the preservation of function and morphology of braintissue. The activity of individual neurons is blocked when flowdecreases below a certain threshold (dashed line) and returnswhen flow is increased again above this threshold. The upperrecordings are from a single neuron before, during, and afterreversible MCA occlusion. The fate of a single cell depends onthe duration for which CBF is impaired below a certain level. Thecurved solid line demarcates conditions for structurally damagedfrom a functionally impaired, but morphologically intact tissue,the ‘penumbra’. The dashed line distinguishes tissue ‘not at risk’from the functionally impaired tissue. MCA, middle cerebralartery; rCBF, regional cerebral blood flow.


1838


glucose metabolism. Increases in glucose consump-tion (and rCBF) evoked by functional activation aremost prominent in synapse-rich regions, i.e., whichcontain axonal terminals, dendritic processes, andastrocytic processes that envelope the synapses andmicrovasculature. Energy requirements for func-tional activation are mostly caused by stimulationof Na+K+-ATPase activity to restore the ionicgradients across neuron cell membranes and mem-brane potentials that are altered by the spike activity(Sokoloff, 1999).

From its conceptual foundation in experiments inthe nonhuman primate, further work on metabolicaberrations in the ‘at-risk’ tissue proceeded insmaller animal models of focal cerebral ischemia.Loss of bioluminescent ATP and tissue acidosisoccur in the ischemic core and penumbra (Astrupet al, 1977; Hossman and Mies, 2007) (AJ Strong,personal communication). The relationships amongischemic perfusion, functional impairment, bio-chemical disturbances, tissue damage, and durationof critical perfusion were shown in PET imagingstudies of cats following MCA occlusion (Heiss et al,1994, 1997). The PET imaging studies have suggestedthree phases of injury progression:

(1) At flows below the threshold of energy metabolism(B20% to 30% of preocclusion values), ‘acute’ischemic injury is usually established withinminutes after the onset of ischemia (Branstonet al, 1974, 1977; Astrup et al, 1977, 1981).

(2) During the subsequent ‘subacute’ phase (flowsranging from 25% to 50% of preocclusion values),the core expands into the penumbra, defined asareas of decreased CBF and O2 metabolism, butincreased OEF. In most models, the core hasconsumed the entire penumbra within 4 to6 hours, whereas in some animal models and insome humans (Heiss et al, 1992), the extension ofthe core into the penumbra can require > 24 hours.This expansion is postulated to be attributed inpart to peri-infarct spreading depression, and otherfactors related to the failure to restore CBF. Peri-infarct spreading depressions are initiated at theborder of the infarct core and spread over theipsilateral hemisphere. During spreading depres-sion, the metabolic rate of the tissue markedlyincreases in response to the activated ion exchangepumps (Somjen, 2001), but is not associated withan increased rCBF in ischemic areas of the brain(review in the study by Strong (2009)). Thus,the increased metabolic workload coupled withlimited O2 supply leads to transient episodes ofworse hypoxia and stepwise increases in lactatewith each depolarization, loss of ionic gradients,and other molecular disturbances that increasecell death (Mies et al, 1993; Hossmann, 2006).

(3) A ‘delayed’ phase of injury evolution occurs thatcan last for several days to weeks. Using multi-parametric imaging techniques for differentiationbetween the core and the penumbra, by 1 hour

after proximal MCA occlusion in the cat, thepenumbra approximately predicts the size of thefinal infarct if flow is not restored (Heiss et al,1994). However, after 3 hours, > 50% of thepenumbra progresses to infarction, and between6 and 8 hours, almost all of the penumbradisappears and is converted into the irreversiblydamaged infarct core (see Figure 5 below).

The flow values for irreversible damage and forfunctional impairment from experimental models ofischemia in cats correspond to those observed inhumans. Considerable variability exists (damagebelow 4.8 to 17.3 mL/100 g per min, penumbra below14.1 to 35.4 mL/100 g per min), which depends onthe definition of damage/penumbra, the methodsused for flow determination, the time of measure-ment after stroke, and whether occlusion waspermanent or temporary (for review see the studyby Heiss (2000)).

Molecular Characteristics of theIschemic Penumbra and the Core

During focal ischemia, the molecular events thatcorrespond best to the physiologic and imagingdefinitions of the penumbra are: (1) regions ofdecreased protein synthesis with preserved ATPand (2) regions of Hsp70 induction in neurons. Manymolecular events occur well outside the clinicallydefined penumbra of ischemia-induced injury thatdoes not lead to detectable infarction. These areassociated with vascular changes during noninjur-ious decreases in blood flow, inflammation with lossof cells and their processes in the core, and synaptic/brain reorganization.

Kleihues and Hossmann (1971) and Hossmann(1993) first described decreased cerebral proteinsynthesis following focal cerebral ischemia. As rCBFdecreases to B30% to 50% of baseline (Mies et al,1990), blockade of translation and decreased proteinsynthesis are observed (Dienel et al, 1980; Kiesslinget al, 1986; Bergstedt et al, 1993; Hossmann, 1994).Within the core region of ischemic injury, proteinsynthesis decreases early and is associated with ATPloss and irreversible translation blockade, whereasin the ‘penumbra’, protein synthesis is initiallydepressed, ATP remains normal, and then proteinsynthesis recovers over time (Hossmann, 1993, 1994).

The Unfolded Protein Response

The decrease in protein synthesis in the ischemiccore is mediated in part by the unfolded proteinresponse within the cell endoplasmic reticulum.Translation blockade proceeds through initiation,maintenance, and termination (DeGracia and Hu,2007). Initiation of the translation blockade in partinvolves the PKR-like endoplasmic reticulum kinasethat phosphorylates eukaryotic initiation factor-2a


1839


and blocks translation in the endoplasmic reticulum.If eukaryotic initiation factor-2a is phosphorylated,the initiation complex is disrupted and translationis slowed or blocked (Hu and Wieloch, 1993). Thisis associated with disaggregation of polysomes,followed by the appearance of stress granules(DeGracia and Hu, 2007). The PKR-like endoplasmicreticulum kinase knockouts only show early trans-lation blocks after ischemia, so that eukaryoticinitiation factor-2a is related to the early, but notprolonged, blockade of protein synthesis followingischemia. After the initial translation blockade, othertranscription factors, including eukaryotic initiationfactor-4(E,F,G), likely maintain decreased proteinsynthesis for 4 to 24 hours and perhaps longer. Thismaintenance phase is followed either by (1) recoveryof translation or by (2) permanent translation block-ade and cell death, which constitute the terminationphase (DeGracia et al, 2008). The proximate stimulifor the unfolded protein response in cerebral ische-mia are still unknown, but may include increasedCa+ 2 in the endoplasmic reticulum, impaired amino-acid or glucose delivery, abnormal glycosylationof proteins, and other factors (Paschen, 1996, 1998;Kaufman et al, 2002, 2004; Schroder and Kaufman,2005). It has been suggested by several groups thatprotein synthesis is arrested following ischemiato prevent formation of partially denatured orimproperly processed proteins, which is probablyprotective. How protein synthesis resumes in thepenumbra that will be salvaged and why proteinsynthesis remains suppressed in regions of penum-bra that infarct are still not understood, but are understudy.

Heat-Shock Protein Induction

Another major molecular responder to ischemia isthe family of Hsps. The Hsps, expressed in everyliving cell and organism, are induced in cells duringperiods of ischemia and other stresses that produceintercellular denatured proteins (Lindquist andCraig, 1988; Craig et al, 1993; Lindquist and Kim,1996; Massa et al, 1996; Yenari, 2002).

Now, Hsp70, the major inducible Hsp, is expressedat almost undetectable levels in normal brain cells.After heat stress, ischemia, and other stresses, Hsp70is massively induced and becomes the most abun-dant protein in the cell because it binds denaturedproteins directly and attempts to refold them(Beckmann et al, 1990, 1992; Welch, 1993; Welchand Brown, 1996). Denatured proteins induce heat-shock transcription factors that bind to heat-shockelements in the Hsp genes and induce Hsp40, Hsp70,Hsp90, and others. At this point, the injured cellappears to make a decision: Hsp70 can promoteprotein refolding and cell survival or allow proteindegradation and progression to cell death (Hohfeldet al, 2001). Importantly, cells that synthesize Hsp70mRNA but cannot synthesize Hsp70 protein likely

die. Most cells that express both Hsp70 mRNA andHsp70 protein appear to survive.

After MCA occlusion, Hsp70 mRNA is expressedthroughout the MCA territory, both within the infarc-tion and in regions adjacent to the infarction (Kinouchiet al, 1993a, b, 1994). Hata et al (1998) have shown thatHsp70 transcription (RNA) and Hsp70 translation(protein) occur in regions of decreased protein synth-esis, but preserved ATP. This is the area where bloodflow is 18% to 20% normal and in which infarctionwould occur if reperfusion did not occur (Hata et al,2000a, b, 1998). Under conditions in which Hsp70transcripts are induced in the MCA distribution, theHsp70 protein is only synthesized in neurons peri-pheral to the ischemic core, whereas there is no or verylittle Hsp70 protein synthesized in the core (Kinouchiet al, 1993a). Thus, the ischemic core corresponds to aregion of either no Hsp70 mRNA induction or induc-tion of Hsp70 mRNA, but without translation into theHsp70 protein. This contrasts with the molecularevents within the penumbra: regions of Hsp70 mRNAinduction and Hsp70 protein synthesis in the regionsof denatured proteins where cells can effectively refoldproteins (Figure 3). This also correlates with differen-tial cellular vulnerability to ischemia because bloodvessels that synthesize Hsp70 mRNA and Hsp70protein may well survive in areas of infraction whereneurons and glia succumb. Although the regions ofHsp70 mRNA and Hsp70 protein induction at 24 hoursappear to correlate well with the clinical definition ofthe penumbra, it is not known whether the volume oftissue destined to infarction corresponds exactly to thatrepresented by the Hsp70 protein, because there is noknown independent molecular marker of tissue at riskfor infarction.

The degree and duration of ischemia affectthe regional expression of Hsp70. For example,10 minutes of focal ischemia in the MCA distribution(using the suture model), which does not infarct ratbrain, induces Hsp70 protein in neurons throughoutthe MCA distribution 4 to 24 hours later. Thus, theregions of Hsp70 protein induction indicate thepenumbra because this is the region ‘at risk’ forinfarction, but which did not infarct because bloodflow was restored after 10 minutes of focal ischemia(Zhan et al, 2008). In contrast, at 24 hours after3-hour (suture-induced) MCA occlusion infarctionoccurs in the MCA territory, and the small area ofHsp70 induction indicates a small penumbra locatedbetween the middle and anterior, and the posteriorcerebral arteries (Zhan et al, 2008). The expressionof Hsp70 in neurons outside areas of infarction ispresumed to protect these cells from further proteindenaturation, as overexpression of the Hsp70 proteinin transgenic mice markedly protects the brainagainst infarction (Rajdev et al, 2000).

Heterogeneity of the Molecular Responses

The ‘boundary’ of the penumbra reflects thedegree and duration of decreased cerebral perfusion.


1840


Outside this region, the hypoperfused ischemic braintissue contains neurons, glia, and vascular elementstogether with inflammatory cells that are associatedwith gliosis, synaptic pruning, and other structuraland molecular changes presaging recovery afterstroke.

In contrast to a central zone of hypoperfused brainat risk for infarction, the pattern and evolution of thepenumbra can be quite variable. For example, insome models of cerebral infarction, there are nozones that correspond to decreased protein synthesisand Hsp70 protein—and thus there is no penumbra.In other models, there are small islands of theinfarcted brain (‘mini-cores’) surrounded by neuronsand glial cells that express Hsp70 (‘mini-penum-bras’) (Figure 3). The ‘most-stressed’ cells at themargins of these islands are sometimes neurons andsometimes microglia. The features that determinethis heterogeneity are unknown, but likely reflect adynamic interplay between adjacent microvascularbeds and complex interactions between the micro-vessels and the adjacent glia and neurons that differover regions of 30 to 40 mm. How this relates toinfarction without the intervening surviving tissue isstill unknown. Indeed, although neurons are morevulnerable than glia to ischemia, there is eventuallya discrete demarcation between the infarcted brainand the surviving brain where both viable neuronsand glia exist at the margin and where theircompanions have perished on the infarcted side ofthe margin. The factors that define such a sharpmargin are unknown, but may reflect interdepen-dent survival growth factors responsible for thesurvival of all members of a given neurovascularunit, or regions lacking adequate microvascularsupply.

The data in Figure 3 strongly support the ideaof ‘mini-penumbras’ and ‘mini-cores’ based on the

pattern of stress gene response. Histologic dataobtained from the study by Hughes et al (2010) alsosupport this concept because they found patchyneuronal loss and associated microglial activation inthe salvaged cortical penumbra after brief MCAocclusion. In addition, we have shown very patchyareas of microinfarction in the striatum and occa-sionally in the cortex after 5- and 10-minute periodsof focal ischemia produced with the suture techni-que in rats (Zhan et al, 2008). These ‘mini-cores’were microscopic and defined by contiguous lossof NeuN-immunostained neurons associated withcontiguous loss of glial fibrillary acidic protein(GFAP)-stained astrocytes in the same region.

Vascular Characteristics of theIschemic Penumbra and the Core

Heterogeneity in the development of cortical neuroninjury has been shown in a rodent model of focalcerebral ischemia (Dawson and Hallenbeck, 1996).One provocative view has suggested that vascularand hemostatic responses vary depending on thespecific organ, although little information was avail-able about the cerebral vasculature (Rosenberg andAird, 1999). It is now known that cellular andmolecular responses of cerebral microvessels to focalischemia are heterogeneously distributed in spaceand time very early in the ischemic territory.

After MCA occlusion in the nonhuman primate,neuron injury and the microvessel responses arerapid and are distributed in a characteristic topo-graphical arrangement with regard to the develop-ment of the ischemic core. Tagaya et al (1997) haveshown that in the evolving core early after injury(defined by evidence of DNA strand breaks or repair),80.0%±6.6% of injured cells were neurons at

penumbra

mini-core

mini-penumbra

*

*

Figure 3 Development of ‘mini-penumbra’ and ‘mini-core’. (A) Coronal section of the adult rat brain that shows Hsp70 protein(stain) 24 hours after occlusion of the right middle cerebral artery (MCA) for 10 minutes. Hsp70 immunoreactivity in the cortex, basalganglia, and ventral thalamus and dorsal hypothalamus in the MCA distribution must be noted. Hsp70 staining delineates the entirepenumbra, because this region of brain did not infarct, but would have infarcted if MCA occlusion had persisted for 2 hours. (B) Theregion of the cortex within the box in panel A shows a central area of little Hsp70 expression surrounded by a large number ofHsp70-stained cells that are predominantly microglia (two are denoted with green stars). For the purposes of this review, we havedesignated the central area that represents a possible microscopic infarct as the ‘mini-core’ and the surrounding area of Hsp70-stained glia and some neurons as the ‘mini-penumbra’. This figure is adapted from the study by Zhan et al (2008). Hsp70, heat-shock protein 70.


1841


2 hours after MCA occlusion, but there is a muchlower proportion of injured endothelial cells at latertimes ( < 2% of cells at 27 hours). This is consistentwith the notion that neurons are most sensitive toischemia, whereas vascular endothelial cells aremost resistant. However, microvessels within theterritories of injury are rapidly reactive, participatingin inflammatory cell adhesion, and displayingchanges in barrier permeability and matrix integrity.These observations indicate heterogeneity in cellvulnerability to injury within the developing coreand adjacent penumbra. Indeed, in the primatecorpus striatum after proximal MCA occlusion,microvessel alterations and injury within the regionsof the ischemic core are distributed heterogeneously(Tagaya et al, 2001).

Microvessel Matrix Receptor Responses

This heterogeneity in the evolution of injury is alsoshown by the response of integrin a1b1 expression bymicrovessel endothelial cells in regions of theterritory at risk that become the core. Integrin a1b1

is a common receptor for endothelial cell adhesion tolaminin, collagen IV, and perlecan in the microvesselbasal lamina. The b1 integrins on the cerebral endo-thelium, and the integrin a6b4 as well as ab-dystro-glycan on astrocyte end-feet all decrease significantlywithin 2 hours after MCA occlusion (Wagner et al,1997; Abumiya et al, 1999; Tagaya et al, 2001).Tagaya et al (2001) have shown that the distributionsand density of microvessels suffering loss of a1 andb1 subunits were heterogeneously distributed, in apattern that depended on the vascular supply.Microvessels in the ischemic striatum expressedincreased b1 subunit transcripts in boundariesaround central regions devoid of b1 integrin trans-cription. To help explain these observations, in vitrointegrin b1 subunit mRNA expression increasedwhen murine brain endothelial cells (grown oncollagen type IV, laminin, or perlecan) were exposedto moderate oxygen–glucose deprivation (experimen-tal ischemia) (Milner et al, 2008). Hence, the regionsof endothelial cell response in the striatum of thenonhuman primate can be interpreted as multiple‘mini-penumbras’ made up of areas of b1 integrinmRNA upregulation on the edges of each core regionand absent b1 integrin expression in ‘mini-cores’because of the failure of transcription and translationin the areas of infarction (Figure 4A) (Tagaya et al,1997, 2001).

Angiogenesis

Another example of vascular response heterogeneityto focal ischemia is the increased expression ofintegrin avb3 and vascular endothelial growth factor(VEGF) by cerebral microvessels within the ischemiccore during MCA occlusion (Abumiya et al, 1999).Vascular endothelial growth factor expression is

responsible for the upregulation of integrin avb3,and stimulates proliferation, migration, and in-creased permeability of the endothelium (Sengeret al, 1996). Abumiya et al have shown that duringstriatal ischemia, VEGF mRNA was detectable innoncapillary microvessels in relationship to the avb3

expression. Colocalization studies, involving micro-vessels reconstructed over 100 mm of their length inregions of dUTP incorporation (the ‘core’), showeda significant colocalization among proliferating cellnuclear antigen (PCNA), the integrin avb3, and VEGFalong the length of activated microvessels (Abumiyaet al, 1999). Integrin avb3 was upregulated mainly on7.5- to 30.0-mm diameter microvessels within thecore region. Microvessel-associated PCNA, VEGF,and integrin avb3 were all upregulated in this region(Figure 4B). Using a hierarchical log-linear model forthe categorical data analysis, tests for pairwiseinteractions at different times showed that theinteraction with time was not significant. Therefore,coexpression of PCNA, VEGF, and integrin avb3 bymicrovessels within the ischemic core indicated ahighly significant interaction that was independentof time, and was therefore heterogeneous within thecore regions.

An interesting observation in this respect suggeststhat in some settings, cerebral microvessels in aterritory could be variously primed for response,such that their response is heterogeneous. Ruetzleret al (2001) have shown segmental heterogeneity ofcentral nervous system vascular responses in nor-moxic spontaneously hypertensive stroke-prone ratsexposed to lipopolysaccharide, in which circulardecoration of brain tissue by manganese superoxidedismutase was seen around some microvesselsand not others. This is possibly relevant to the obser-vations made by Mabuchi et al (2005) of an ordered,but heterogeneously distributed, neuron injurywithin 2 hours of MCA occlusion that is related tothe microvessel supply (the ‘neurovascular unit’).Of importance, these observations are not simplyexplained by the dependence of local injury on thefall-off of O2 diffusion as applied to cerebralcapillaries (the Krogh cylinder), but suggest othermechanisms (Quistorff et al, 1977; Mabuchi et al,2005). Whether the coordinated expression of PCNA,integrin avb3, and VEGF by cerebral microvesselsduring focal ischemia is an intrinsic characteristicof some vessels of a size class, or the heterogeneity ofthe injury, is not yet certain.

Heterogeneity of the Microvascular Responses

Responses of the microvasculature in the territory atrisk (striatum) to focal ischemia are very rapid,nearly as rapid as those of the neurons they supply.Within the core regions, in the first minutes afterischemia onset, microvessel obstruction, endothelialcell receptor presentation, matrix degradation, anddetachment of astrocyte end-feet occur initially in


1842


a heterogeneous manner in pockets (‘mini-cores’).The initial microvessel responses are neither simul-taneous nor homogeneous. This suggests that in theearly moments after arterial occlusion, the entire

‘core’ is studded with pockets of ‘penumbra’, andthat these ‘mini-cores’ and their ‘mini-penumbras’evolve dynamically and heterogeneously dependingon local differences of microvessel perfusion.

Figure 4 Expression of microvessel-related gene products at 2 hours after MCA occlusion in the striatum. (A) Expression of theintegrin b1 subunit mRNA by microvessels in the striatum of three nonhuman primate subjects (a–c) after MCA:O (Tagaya et al,2001). Around mini-cores of absent b1 subunit mRNA, significant upregulation of the b1 subunit gene product was observed. Thecores and boundary zones of b1 subunit upregulation corresponded to the regions where cells incorporated dUTP (i.e., dUTP+ )(Tagaya et al, 1997). (B) Regions of VEGF expression (black), an subunit (gray), and expression of both products (hatched) byactivated microvessels in the striata of nonhuman primates that were subject to various periods of MCA occlusion (each imagerepresents a separate animal) (Abumiya et al, 1999). The mini-cores of integrin or VEGF expression by microvessels must also benoted. MCA, middle cerebral artery; VEGF, vascular endothelial growth factor.


1843


Imaging the Ischemic Penumbra andCore

Noninvasive brain imaging modalities have evolvedgreatly since the first descriptions of the core–penumbra relationships in large animal models.The use of PET and MR imaging techniques inhumans is now beginning to confirm and evenextend observations of the core and penumbradefined in animal models.

Positron Emission Tomography Imaging

Positron emission tomography techniques havepermitted classification of three regions within thedisturbed vascular territory: (1) the core of ischemiawhich usually undergoes necrosis with a flow rate< 12 mL/100 g per min, (2) a penumbra region (12 to22 mL/100 g per min) of still viable tissue, butwith uncertain chances for infarction or recoverydepending on reflow, and (3) a hypoperfused area( > 22 mL/100 g per min) not primarily damaged bydecreased blood supply that will not proceed toinfarction (see Figure 5). In humans, the brain tissuewith rCBF < 12 mL/100 g per min and rCMRO2

< 65 mmol/100 g per min (often measured severalhours after stroke) usually displays eventual evidenceof infarction on late computerized tomographic scans(Ackerman et al, 1981; Baron et al, 1981a; Lenzi et al,1982; Powers et al, 1985). Relatively preserved CMRO2

in regions of severely reduced CBF has been taken asan indicator of maintained neuronal function inregions with severely reduced CBF. This pattern,coined ‘misery perfusion’, has served as a PET-deriveddefinition of the penumbra (Baron et al, 1981b). It ischaracterized as the area of increased OEF ( > 80%from the normal value of B40%). Positron emissiontomography investigations imply that the extent of thepenumbra depends on the time of measurementrelative to the onset of ischemia. The volume is largeand rCBF is low if the penumbra is defined in the firsthours of ischemia; the penumbra volume is small, ifdefined later (Heiss et al, 1992). The heterogeneity inthe core of ischemia and in the penumbra over timecan be seen in the occurrence of OEF especially inanimal studies (Figure 5), but also seen repeatedly inhuman PET studies (e.g., Heiss et al, 1992).

As these measurements require arterial bloodsampling and complex logistics, a marker of neuro-nal integrity is preferable. The central benzodiaze-pine receptor ligand flumazenil (FMZ) binds to theGABA receptor that is abundant in the cerebralcortex. These receptors are sensitive to ischemicdamage and can identify early neuronal loss. Fluma-zenil, validated in the cat MCA occlusion model,predicted the size of final infarction in patients withacute ischemic stroke, and showed the efficacy ofthrombolytic treatment (Heiss, 2000). Another mar-ker of the penumbra, 18F misonidazol is trapped inviable hypoxic tissue (Takasawa et al, 2007). In early

stroke, increased 18F misonidazol uptake surroundsthe core, and there is a strong association betweenthe extent of 18F misonidazol-binding tissue that sur-vives and functional outcome (Markus et al, 2003).18F misonidazol uptake also assesses hypoxia in thewhite matter, where progression of ischemic damageis slower (Heiss et al, 2000; Spratt et al, 2007).

As PET methodology has improved, higher resolu-tion with shorter time sequences have detectedtemporal and spatial heterogeneity of the core andpenumbra (Kidwell et al, 2003), which were espe-cially evident in sequential studies in animal modelsof focal and reversible cerebral ischemia (Heiss et al,1994).

Magnetic Resonance Imaging

Although PET remains the imaging gold standard foridentification of tissue in the penumbra in humanstroke patients, MR studies using diffusion andperfusion imaging by MR have provided valuableinsights into the relationships between the core andthe penumbra. More than a decade ago, it washypothesized that the early diffusion-weighted ima-ging (DWI) lesion estimates the ischemic core inischemic stroke patients and adjacent criticallyhypoperfused tissue could be identified with per-fusion-weighted imaging (PWI) (Baird et al, 1997;Barber et al, 1998). Therefore, brain regions with aPWI lesion that did not show restricted diffusion(PWI/DWI mismatch) were hypothesized to repre-sent the penumbra. Subsequent studies haveextended this notion by showing that appropriatequantification of perfusion and diffusion imagingcan provide a reasonable estimate of the penumbra.

Problems with the MRI definition of the penumbrainclude the fact that there is no threshold within aregion of oligemia (Kidwell et al, 2003; Kane et al,2007; Toth and Albers, 2009), and that the PWIabnormality often overestimates the final infarctionvolume, and hence the amount of tissue at risk(Parsons et al, 2001). In addition, the initial diffusionlesion does not necessarily define the core ofinfarcted tissue because some diffusion lesions canbe transiently or permanently reversed if blood flowis rapidly restored (Kidwell et al, 2000; Parsons et al,2002; Chalela et al, 2004).

Kidwell et al (2003) proposed a modified model ofischemia-compromised tissue in which the penum-bra includes the region of perfusion–diffusion mis-match, minus the region of benign oligemia plus aportion of the initial diffusion abnormality itself.Although several attempts have been made toidentify perfusion or apparent diffusion coefficient(ADC) thresholds so as to better differentiate theseregions, consensus on the best method has beenhampered by the lack of methodological standardiza-tion of image postprocessing and analysis. Theserestrict the pooling of data and cross-comparison ofresults across studies (Kane et al, 2007).


1844


Validation of MR signatures with regard to PETmeasurements might help in the interpretation ofthe respective findings and in the assessment ofthe accuracy of the various measures for predictingtissue outcome. Comparisons of PET and MR imaginghave been performed in small groups of patients toassess perfusion, and the delineation of tissue in thepenumbra (Figure 5).

To predict irreversible cortical damage, resultsfrom sequential early DWI and 11C-FMZ-PET out-comes were compared with infarct extension 24 to48 hours later on T2-weighted MRI in 12 acute strokepatients (Heiss et al, 2004). When the volumes oftissue beyond the defined thresholds of FMZ bindingand DWI intensity were compared, close correlations(1) between volumes with FMZ and DWI beyondthreshold and (2) between predicted and final infarctvolumes were obtained, but the volumes did notcompletely overlap. Whereas false-positive resultswere observed for 25.9% of the total volume of DWI

increase, they were negligible with FMZ-PET. Atime-to-peak delay of 4 seconds correlated with flowdecreases below 20 mL/100 g per min as identifiedby H2

15O-PET (Zaro-Weber et al, 2009).For demarcation of the volume of the penumbra,

the areas of PWI/DWI mismatch were compared withthose of increased OEF. Using PWI/DWI mismatch,there was considerable variability in penumbravolume, and all patients (13/13) showed areas ofmismatch. However, PET detected OEF in only 8 of13 patients. The areas of OEF elevation were alwayslocated within the areas of time-to-peak prolonga-tion, and were significantly smaller and covered only1% to 75% (median, 33%) of the time-to-peak areaon MRI. These data show a high sensitivity, and alow specificity of the chosen thresholds to identifythe penumbra as defined by PET. This suggests thatthe PWI/DWI volume ratios depicted by time to peakdo not reliably reflect the penumbra as definedby PET.

Figure 5 Sequential PET images of CBF, CMRO2, and OEF of permanent MCA occlusion in cats (left columns) compared with imagesof patients 12 hours after stroke (right columns). Two different cat subjects are shown (efficient and nonefficient perfusion) and threedifferent patients are shown (A: early OEF defect with corresponding infarction on MRI, B: effective reperfusion without infarction,and C: ineffective/delayed reperfusion with large final partially hemorrhagic infarction). In the cat, the progressive decrease of CMRO2

and the reduction of OEF predict infarction. In the patient, the area with preserved OEF is not infarcted (outside region on late MRI,upper part of figure A). If reperfusion occurs before OEF is reduced, tissue can be salvaged (left: cat, and left: patient in the lower partof the figure B). If reperfusion is achieved after this therapeutic window, tissue cannot be salvaged (right: cat, and right: patientin lower part of the figure C). CBF, cerebral blood flow; CMRO2, cerebral metabolic rate for oxygen; MCA, middle cerebral artery;MRI, magnetic resonance imaging; OEF, oxygen extraction fraction; PET, positron emission tomography.


1845


Advances in Perfusion-Weighted Imaging Techniques

Nonetheless, the spatial correspondence betweenearly DWI and PWI lesions, in conjunction withknowledge of whether reperfusion of the supplyartery(ies) occurs, allow a reasonable (but not ahighly accurate) prediction of the final infarctionvolume. More advanced PWI techniques, includingdeconvolution sequences that account for differencesin arterial input to vessels with normal flow, appearto produce more accurate estimates of the regions ofhypoperfusion that define the outer boundaries ofthe penumbra. Although PWI provides a number ofparameters including mean transit time and cerebralblood volume, recently there has been focus on themaximum of the tissue residue function (Tmax)obtained by deconvolution (Zaro-Weber et al, 2010).Although the physiologic interpretation of Tmax iscomplex, it reflects the degree of lag between thearterial input and the tissue response, as well asdispersion and mean transit time (Calamante et al,2010). This parameter appears to correlate with CBFvalues determined by Xenon computerized tomo-graphic scans and has performed well in twoprospective clinical stroke trials (Albers et al, 2006;Davis et al, 2008).

Thresholding techniques that restrict the volumeof PWI lesions to brain regions where there aresubstantial delays in contrast arrival or transit timeshave also been used. This reduces the volume ofbenign oligemia captured by the PWI map. Predic-tion of the final infarction volume and salvage of thepenumbra seems to be more accurate using Tmaxmaps with higher thresholds. In the DEFUSE (Diffu-sion and Perfusion Imaging Evaluation for Under-standing Stroke Evolution) study, for instance, thecorrelation between infarct growth and the volumeof penumbra salvaged was significantly better forPWI lesions defined by Tmax > 6 seconds comparedwith > 2 seconds (Olivot et al, 2009). In addition, inpatients who did not experience early reperfusion,the > 4-second threshold was a more accuratepredictor of final infarction volume than Tmax> 2 seconds. Similarly, in EPITHET (EchoplanarImaging Thrombolytic Evaluation Trial), both thesize and the severity of PWI lesions correlated withclinical outcomes (Parsons et al, 2010). Recent datahave suggested a B90% sensitivity and specificityfor Tmax > 5.5 seconds to identify the PET-definedpenumbra rCBF threshold of 20 mL/100 g per min(Zaro-Weber et al, 2010). These results suggest thatquantification of the severity of PWI lesions canimprove accuracy for identification of tissue that isin the penumbra.

Advances in Diffusion-Weighted Imaging Techniques

Progress has been made in quantification of DWIusing ADC measurements. Animal studies andhuman trials are congruent, both showing correla-tions between CBF and the severity of ADC decline

(Kohno et al, 1995; Hoehn-Berlage et al, 1995;Dijkhuizen et al, 1997; Thijs et al, 2002; Lin et al,2003). With increasing time from symptom onset,ADC values become progressively reduced in acritically low CBF environment. Recent data haveshown that only a small volume of the acute DWIlesion is reversible (Chemmanam et al, 2010).Diffusion-weighted imaging lesions are most likelyto reverse if they have modest reductions in ADC andthere is early reperfusion (Olivot et al, 2009).However, observations in both animal models andin human stroke patients show that reperfusion isassociated with a rapid increase in ADC values, andthat this increase is not necessarily reflective oftissue salvage. Therefore, an isolated ADC map canbe an unreliable predictor of eventual infarct volume(Ringer et al, 2001).

The fact that early DWI lesions are not uniformlyincorporated into the final infarction has an impor-tant implication: it is likely that some part of theischemic penumbra is contained within the acuteDWI lesion. Additional research is required to moreaccurately determine which regions of the acute DWIlesions are most likely to be salvageable, i.e., part ofthe penumbra.

The Core and Penumbra are Dynamic

Experimental data from vascular and molecularmodeling studies, as well as recent imaging workindicate that the penumbra is dynamic. There appearto be substantial fluctuation in volumes, locations,and structures of DWI and PWI lesions during theearly hours after stroke onset (Ma et al, 2009).Coregistration of acute DWI lesions with PETimaging has shown that DWI lesions are hetero-geneous and include regions of variable metabolicdisruption and flow-metabolism coupling (Guadagnoet al, 2006). These findings are compatible with theobservations discussed above regarding the relation-ship among rCBF, ADC, and DWI. Therefore, pre-dicting the fate of ischemic brain tissue based on dataobtained from any single imaging modality per-formed at only one early time point may not havehigh specificity and sensitivity for prediction of thecore and penumbra.

Visualization of the spatial relationships betweenPWI and DWI lesions at different times after strokeonset can be achieved by coregistration of sequentialMRI scans in stroke patients. Three-dimensionalanalyses have shown that a central volume of restric-ted diffusion (‘core’) surrounded by a hypoperfusedpenumbra is not present in the majority of strokepatients imaged 3 to 6 hours after symptom onset(before administration of thrombolytic therapy).

Multiple discrete regions of restricted diffusion orperfusion were documented in one-third of thepatients enrolled in the DEFUSE study (Figure 6).Surprisingly, a substantial proportion of the acuteDWI lesions (54%) did not have a superimposed


1846


perfusion lesion (Olivot et al, 2009). These regionsof restricted diffusion without superimposed PWIlesions are assumed to represent regions whereperfusion deficits have resolved or shifted to adja-cent brain regions. The term RADAR (ReversibleAcute Diffusion lesion Already Reperfused) has beenused to describe these regions (Olivot et al, 2009).The RADAR regions appear to have an increased rateof DWI reversal. These findings emphasize theheterogeneity and complexity of the relationshipsbetween acute PWI and DWI lesions.

The Penumbra Revisited

The original notion that occlusion of a brain-supplying artery produces a central core of tissueinjury destined for infarction that is surrounded by apenumbra of metabolically metastable tissue withthe potential for full recovery still has great merit.This model is strongly supported by the pathologicfindings of stroke that show confluent regions oftissue infarction in animal models and patients.Recent experimental and imaging work offer therefinement that in the early minutes and hours afterischemia onset, the core contains pockets of injury

Figure 6 Heterogeneity of DWI and PWI lesion structure. (A) Examples of the three-dimensional structure of DWI and PWI lesions inacute stroke patients. Left, a single PWI lesion; center, a DWI lesion with multiple individual lesion components; right, a PWI lesionwith multiple components. (B) Three-dimensional imaging of DWI and PWI lesions in acute stroke patients reveals intertwinedregions of mismatch, DWI/PWI overlap, and early reperfusion. Regions that contain DWI lesions without superimposed PWI lesions(early reperfusion) are shown in blue; areas where DWI lesions have superimposed PWI lesions are shown in purple. The red areasare PWI lesions without superimposed DWI lesions (regions of mismatch). Reprinted from Stroke with permission from Olivot et al(2009). DWI, diffusion-weighted imaging; PWI, perfusion-weighted imaging.

mini-cores

normal+

mini-penumbras

normal

core

penumbra

Figure 7 A new hypothetical construct of the penumbra. Thearchitecture and reversibility depend on time and location ofrCBF reduction in the territory at risk after occlusion of a brain-supplying artery. The outside boundary represents the territory atrisk of cerebral tissue that functions normally until rCBF isreduced for an extended period. Mini-cores coalesce with theduration of rCBF reduction devouring micro-penumbras. Theperiod of evolution from normal function to the final state of theterritory at risk may depend on a number of factors, includingtissue location, depth of reduction of rCBF, inflammatory state atbaseline and/or degree of inflammatory response, and otherfactors. The location of the mini-cores and mini-penumbrasappear heterogeneously distributed. However, they in fact reflectthe microvascular supply of tissue and cell vulnerabilities. rCBF,regional cerebral blood flow.


1847


which we characterize as ‘mini-cores’, surroundedby ‘mini-penumbras’. We hypothesize that unlesssalvaged, such embedded ‘mini-penumbras’ areconsumed by the expanding ‘mini-cores’ of metabolicand functional failure to generate a lesion that isultimately homogenous and can grow into the sur-rounding injured tissue (Figure 7). Furthermore, it ishypothesized that the ‘mini-cores’ contain tissues inwhich neurovascular units have been irreversiblydamaged because of flow cessation and its conse-quences, whereas the ‘mini-penumbras’ have a propor-tion of neurovascular units that are viable. The separatenetworks of microvessels, neurons, and glia associatethe ‘mini-cores’ with the ‘mini-penumbras’. In thisway, if unimpeded, these mini-cores can grow intotheir respective mini-penumbras to encompass a largerregion of injury.

This refined view has several implications:

1. Multiple ‘mini-penumbras’ are apparent at themolecular, cellular, and microvessel levels, suchthat higher imaging resolution will be necessary tofollow their fate in patients,

2. these cell and microvessel events are related tolocal microvascular flow,

3. serial imaging studies are required to show theevolution of this dynamic core and penumbra,

4. inflammatory cell–endothelial interactions, peri-infarct depolarization, flow changes, and otherprocesses contribute to the engulfment of ‘mini-penumbras’ into the evolving ischemic core,

5. these observations underscore the need forextremely early interventions, and

6. depending on the vascular territory involved andthe timing and severity of the ischemic event,injury evolution may occur at different speedswithin different ‘mini-cores’/‘mini-penumbras’,such that certain interventions might have benefitlater.

Future studies will need to determine whether thepresence or absence of ‘mini-cores’ and ‘mini-penumbras’ might affect the efficacy of differenttypes of treatment, the timing of treatment, or impactultimate prognosis.

Acknowledgement

The authors thank G Berg for manuscript preparation.

Disclosure/Conflict of interest

The authors declare no conflict of interest.

References

Abumiya T, Lucero J, Heo JH, Tagaya M, Koziol JA,Copeland BR, del Zoppo GJ (1999) Activated micro-vessels express vascular endothelial growth factor and

integrin alpha(v)beta3 during focal cerebral ischemia.J Cereb Blood Flow Metab 19:1038–50

Ackerman RH, Correia JA, Alpert NM, Baron JC, GouliamosA, Grotta JC, Brownell GL, Taveras JM (1981) Positronimaging in ischemic stroke disease using compoundslabeled with oxygen 15. Initial results of clinicophysio-logic correlations. Arch Neurol 38:537–43

Albers GW, Thijs VN, Wechsler L, Kemp S, Schlaug G,Skalabrin E, Bammer R, Kakuda W, Lansberg MG,Shuaib A, Coplin W, Hamilton S, Moseley M, MarksMP (2006) Magnetic resonance imaging profiles predictclinical response to early reperfusion: the Diffusion andPerfusion Imaging Evaluation for Understanding StrokeEvolution (DEFUSE) study. Ann Neurol 60:508–17

Astrup J, Siesjo BK, Symon L (1981) Thresholds in cerebralischemia—the ischemic penumbra. Stroke 12:723–5

Astrup J, Symon L, Branston NM, Lassen NA (1977)Cortical evoked potential and extracellular K+ and H+

at critical levels of brain ischemia. Stroke 8:51–7Baird AE, Benfield A, Schlaug G, Siewert B, Lovblad KO,

Edelman RR, Warach S (1997) Enlargement of humancerebral ischemic lesion volumes measured by diffu-sion-weighted magnetic resonance imaging. Ann Neurol41:581–9

Barber PA, Darby DG, Desmond PM, Yang Q, Gerraty RP,Jolley D, Donnan GA, Tress BM, Davis SM (1998) Predic-tion of stroke outcome with echoplanar perfusion- anddiffusion-weighted MRI. Neurology 51:418–26

Baron JC (1999) Mapping the ischaemic penumbra withPET: implications for acute stroke treatment. Cerebro-vasc Dis 9:193–201

Baron JC, Bousser MG, Comar D, Soussaline F, Castaigne P(1981a) Noninvasive tomographic study of cerebralblood flow and oxygen metabolism in vivo. Potentials,limitations, and clinical applications in cerebralischemic disorders. Eur Neurol 20:273–84

Baron JC, Bousser MG, Rey A, Guillard A, Comar D,Castaigne P (1981b) Reversal of focal ‘misery-perfusionsyndrome’ by extra-intracranial arterial bypass in hemo-dynamic cerebral ischemia. A case study with 15Opositron emission tomography. Stroke 12:454–9

Beckmann RP, Lovett M, Welch WJ (1992) Examining thefunction and regulation of hsp 70 in cells subjected tometabolic stress. J Cell Biol 117:1137–50

Beckmann RP, Mizzen LE, Welch WJ (1990) Interaction ofHsp 70 with newly synthesized proteins: implicationsfor protein folding and assembly. Science 248:850–4

Bergstedt K, Hu BR, Wieloch T (1993) Postischaemicchanges in protein synthesis in the rat brain: effects ofhypothermia. Exp Brain Res 95:91–9

Branston NM, Strong AJ, Symon L (1977) Extracellularpotassium activity, evoked potential and tissue bloodflow. Relationships during progressive ischaemia inbaboon cerebral cortex. J Neurol Sci 32:305–21

Branston NM, Symon L, Crockard HA, Pasztor E (1974)Relationship between the cortical evoked potential andlocal cortical blood flow following acute middlecerebral artery occlusion in the baboon. Exp Neurol45:195–208

Calamante F, Christensen S, Desmond PM, Ostergaard L,Davis SM, Connelly A (2010) The physiological signi-ficance of the time-to-maximum (Tmax) parameter inperfusion MRI. Stroke 41:1169–74

Chalela JA, Kang DW, Luby M, Ezzeddine M, Latour LL,Todd JW, Dunn B, Warach S (2004) Early magneticresonance imaging findings in patients receiving tissueplasminogen activator predict outcome: insights into


1848


the pathophysiology of acute stroke in the thrombolysisera. Ann Neurol 55:105–12

Chemmanam T, Campbell BC, Christensen S, Nagakane Y,Desmond PM, Bladin CF, Parsons MW, Levi CR, Barber PA,Donnan GA, Davis SM (2010) Ischemic diffusion lesionreversal is uncommon and rarely alters perfusion-diffusionmismatch. Neurology 75:1040–7

Craig EA, Gambill BD, Nelson RJ (1993) Heat shockproteins: molecular chaperones of protein biogenesis.Microbiol Rev 57:402–14

Davis SM, Donnan GA, Parsons MW, Levi C, Butcher KS,Peeters A, Barber PA, Bladin C, De Silva DA, Byrnes G,Chalk JB, Fink JN, Kimber TE, Schultz D, Hand PJ,Frayne J, Hankey G, Muir K, Gerraty R, Tress BM,Desmond PM (2008) Effects of alteplase beyond 3 hoursafter stroke in the Echoplanar Imaging ThrombolyticEvaluation Trial (EPITHET): a placebo-controlledrandomised trial. Lancet Neurol 7:299–309

Dawson DA, Hallenbeck JM (1996) Acute focal ischemia-induced alterations in MAP2 immunostaining: descrip-tion of temporal changes and utilization as a marker forvolumetric assessment of acute brain injury. J CerebBlood Flow Metab 16:170–4

DeGracia DJ, Hu BR (2007) Irreversible translation arrest inthe reperfused brain. J Cereb Blood Flow Metab 27:875–93

DeGracia DJ, Jamison JT, Szymanski JJ, Lewis MK (2008)Translation arrest and ribonomics in post-ischemicbrain: layers and layers of players. J Neurochem106:2288–301

Dienel GA, Pulsinelli WA, Duffy TE (1980) Regionalprotein synthesis in rat brain following acute hemi-spheric ischemia. J Neurochem 35:1216–26

Dijkhuizen RM, Berkelbach van der Sprenkel JW,Tulleken KA, Nicolay K (1997) Regional assessmentof tissue oxygenation and the temporal evolution ofhemodynamic parameters and water diffusion duringacute focal ischemia in rat brain. Brain Res 750:161–70

Guadagno JV, Warburton EA, Jones PS, Day DJ, AigbirhioFI, Fryer TD, Harding S, Price CJ, Green HA, Barret O,Gillard JH, Baron JC (2006) How affected is oxygenmetabolism in DWI lesions? A combined acute strokePET-MR study. Neurology 67:824–9

Harris RJ, Symon L (1984) Extracellular pH, potassium, andcalcium activities in progressive ischaemia of rat cortex.J Cereb Blood Flow Metab 4:178–86

Hata R, Maeda K, Hermann D, Mies G, Hossmann KA(2000a) Dynamics of regional brain metabolism andgene expression after middle cerebral artery occlusionin mice. J Cereb Blood Flow Metab 20:306–15

Hata R, Maeda K, Hermann D, Mies G, Hossmann KA(2000b) Evolution of brain infarction after transient focalcerebral ischemia in mice. J Cereb Blood Flow Metab20:937–46

Hata R, Mies G, Wiessner C, Hossmann KA (1998)Differential expression of c-fos and hsp72 mRNA infocal cerebral ischemia of mice. Neuroreport 9:27–32

Heiss WD (1992) Experimental evidence of ischemicthresholds and functional recovery. Stroke 23:1668–72

Heiss WD (2000) Ischemic penumbra: evidence fromfunctional imaging in man. J Cereb Blood Flow Metab20:1276–93

Heiss WD, Graf R, Lottgen J, Ohta K, Fujita T, Wagner R,Grond M, Weinhard K (1997) Repeat positron emissiontomographic studies in transient middle cerebral arteryocclusion in cats: residual perfusion and efficacy of

postischemic reperfusion. J Cereb Blood Flow Metab17:388–400

Heiss WD, Graf R, Wienhard K, Lottgen J, Saito R, Fujita T,Rosner G, Wagner R (1994) Dynamic penumbra demon-strated by sequential multitracer PET after middlecerebral artery occlusion in cats. J Cereb Blood FlowMetab 14:892–902

Heiss WD, Hayakawa T, Waltz AG (1976) Cortical neuronalfunction during ischemia. Effects of occlusion of onemiddle cerebral artery on single-unit activity in cats.Arch Neurol 33:813–20

Heiss WD, Huber M, Fink GR, Herholz K, Pietryzyk V,Wagner R, Weinhard K (1992) Progressive derangementof peri-infarct viable tissue in ischemic stroke. J CerebBlood Flow Metab 12:193–203

Heiss WD, Kracht L, Grond M, Rudolf J, Bauer B, Wienhard K,Pawlik G (2000) Early [(11)C]Flumazenil/H(2)O positronemission tomography predicts irreversible ischemiccortical damage in stroke patients receiving acutethrombolytic therapy. Stroke 31:366–9

Heiss WD, Rosner G (1983) Functional recovery of corticalneurons as related to degree and duration of ischemia.Ann Neurol 14:294–301

Heiss WD, Sobesky J, Smekal U, Kracht LW, Lehnhardt FG,Thiel A, Jacobs AH, Lackner K (2004) Probability ofcortical infarction predicted by flumazenil binding anddiffusion-weighted imaging signal intensity: a compara-tive positron emission tomography/magnetic resonanceimaging study in early ischemic stroke. Stroke 35:1892–8

Hoehn-Berlage M, Norris DG, Kohno K, Mies G, Leibfritz D,Hossmann KA (1995) Evolution of regional changes inapparent diffusion coefficient during focal ischemiaof rat brain: the relationship of quantitative diffusionNMR imaging to reduction in cerebral blood flow andmetabolic disturbances. J Cereb Blood Flow Metab15:1002–11

Hohfeld J, Cyr DM, Patterson C (2001) From the cradle tothe grave: molecular chaperones that may choosebetween folding and degradation. EMBO Rep 2:885–90

Hossman K-A, Mies G (2007) Multimodal mapping of theischemic penumbra in animal models. In: The IschemicPenumbra: Pathophysiology, Imaging and Therapy(Donnan GA, Baron J-C, Davis SM, Sharp FR, eds),New York: Informa Healthcare, 77–92

Hossmann KA (1993) Disturbances of cerebral proteinsynthesis and ischemic cell death. Prog Brain Res96:161–77

Hossmann KA (1994) Viability thresholds and the penum-bra of focal ischemia. Ann Neurol 36:557–65

Hossmann KA (2006) Pathophysiology and therapy ofexperimental stroke. Cell Mol Neurobiol 26:1057–83

Hu BR, Wieloch T (1993) Stress-induced inhibition ofprotein synthesis initiation: modulation of initiationfactor 2 and guanine nucleotide exchange factoractivities following transient cerebral ischemia in therat. J Neurosci 13:1830–8

Hughes JL, Beech JS, Jones PS, Wang D, Menon DK, Baron JC(2010) Mapping selective neuronal loss and microglialactivation in the salvaged neocortical penumbra in therat. Neuroimage 49:19–31

Jones TH, Morawetz RB, Crowell RM, Marcoux FW,Fitzgibbon SJ, DeGirolami U, Ojemann RG (1981) Thresh-olds of focal cerebral ischemia in awake monkeys.J Neurosurg 54:773–82

Kane I, Carpenter T, Chappell F, Rivers C, Armitage P,Sandercock P, Wardlaw J (2007) Comparison of 10


1849


different magnetic resonance perfusion imaging proces-sing methods in acute ischemic stroke: effect on lesionsize, proportion of patients with diffusion/perfusionmismatch, clinical scores, and radiologic outcomes.Stroke 38:3158–64

Kaufman RJ (2004) Regulation of mRNA translation byprotein folding in the endoplasmic reticulum. TrendsBiochem Sci 29:152–8

Kaufman RJ, Scheuner D, Schroder M, Shen X, Lee K, LiuCY, Arnold SM (2002) The unfolded protein response innutrient sensing and differentiation. Nat Rev Mol CellBiol 3:411–21

Kidwell CS, Alger JR, Saver JL (2003) Beyond mismatch:evolving paradigms in imaging the ischemic penumbrawith multimodal magnetic resonance imaging. Stroke34:2729–35

Kidwell CS, Saver JL, Mattiello J, Starkman S, Vinuela F,Duckwiler G, Gobin YP, Jahan R, Vespa P, Kalafut M,Alger JR (2000) Thrombolytic reversal of acute humancerebral ischemic injury shown by diffusion/perfusionmagnetic resonance imaging. Ann Neurol 47:462–9

Kiessling M, Dienel GA, Jacewicz M, Pulsinelli WA(1986) Protein synthesis in postischemic rat brain: atwo-dimensional electrophoretic analysis. J Cereb BloodFlow Metab 6:642–9

Kinouchi H, Sharp FR, Chan PH, Koistinaho J, Sagar SM,Yoshimoto T (1994) Induction of c-fos, junB, c-jun, andhsp 70 mRNA in cortex, thalamus, basal ganglia andhippocampus following middle cerebral artery occul-sion. J Cereb Blood Flow Metab 14:808–17

Kinouchi H, Sharp FR, Hill MP, Koistinaho J, Sagar SM,Chan PH (1993a) Induction of 70-kDa heat shock proteinand hsp70 mRNA following transient focal cerebralischemia in the rat. J Cereb Blood Flow Metab 13:105–15

Kinouchi H, Sharp FR, Koistinaho J, Hicks K, Kamii H,Chan PH (1993b) Induction of heat shock hsp70 mRNAand HSP70 kDa protein in neurons in the ‘penumbra’following focal cerebral ischemia in the rat. Brain Res619:334–8

Kleihues P, Hossmann KA (1971) Protein synthesis in thecat brain after prolonged cerebral ischemia. Brain Res35:409–18

Kohno K, Back T, Hoehn-Berlage M, Hossmann K (1995)Relationship between diffusion-weighted magnetic re-sonance images, cerebral blood flow and energy state inexperimental brain infarction. Magn Reson Imaging13:65–71

Lenzi GL, Frackowiak RSJ, Jones T (1982) Cerebral oxygenmetabolism and blood flow in human cerebral ischemicinfarction. J Cereb Blood Flow Metab 2:321–35

Lin W, Lee JM, Lee YZ, Vo KD, Pilgram T, Hsu CY (2003)Temporal relationship between apparent diffusioncoefficient and absolute measurements of cerebral bloodflow in acute stroke patients. Stroke 34:64–70

Lindquist S, Craig EA (1988) The heat-shock proteins.Annu Rev Genet 22:631–77

Lindquist S, Kim G (1996) Heat-shock protein 104 expres-sion is sufficient for thermotolerance in yeast. Proc NatlAcad Sci USA 93:5301–6

Ma H, Zavala JA, Teoh H, Churilov L, Gunawan M, Ly J,Wright P, Phan T, Arakawa S, Davis SM, Donnan GA(2009) Fragmentation of the classical magnetic reso-nance mismatch ‘penumbral’ pattern with time. Stroke40:3752–7

Mabuchi T, Lucero J, Feng A, Koziol JA, del Zoppo GJ(2005) Focal cerebral ischemia preferentially affects

neurons distant from their neighboring microvessels.J Cereb Blood Flow Metab 25:257–66

Markus R, Reutens DC, Kazui S, Read S, Wright P,Chambers BR, Sachinidis JI, Tochon-Danguy HJ,Donnan GA (2003) Topography and temporal evolutionof hypoxic viable tissue identified by 18F-fluoromiso-nidazole positron emission tomography in humans afterischemic stroke. Stroke 34:2646–52

Massa SM, Swanson RA, Sharp FR (1996) The stress generesponse in brain. Cerebrovasc Brain Metab Rev 8:95–158

Mies G, Iijima T, Hossmann KA (1993) Correlation betweenperi-infarct DC shifts and ischaemic neuronal damage inrat. Neuroreport 4:709–11

Mies G, Paschen W, Ebhardt G, Hossmann KA (1990)Relationship between of blood flow, glucose metabo-lism, protein synthesis, glucose and ATP content inexperimentally-induced glioma (RG1 2.2) of rat brain.J Neurooncol 9:17–28

Milner R, Hung S, Wang X, Berg GI, Spatz M, del Zoppo GJ(2008) Responses of endothelial cell and astrocytematrix-integrin receptors to ischemia mimic thoseobserved in the neurovascular unit. Stroke 39:191–7

Olivot JM, Mlynash M, Thijs VN, Purushotham A, Kemp S,Lansberg MG, Wechsler L, Bammer R, Marks MP, AlbersGW (2009) Relationships between cerebral perfusionand reversibility of acute diffusion lesions in DEFUSE:insights from RADAR. Stroke 40:1692–7

Parsons MW, Barber PA, Chalk J, Darby DG, Rose S,Desmond PM, Gerraty RP, Tress BM, Wright PM,Donnan GA, Davis SM (2002) Diffusion- and per-fusion-weighted MRI response to thrombolysis instroke. Ann Neurol 51:28–37

Parsons MW, Christensen S, McElduff P, Levi CR,Butcher KS, De Silva DA, Ebinger M, Barber PA,Bladin C, Donnan GA, Davis SM, Echoplanar ImagingThrombolytic Evaluation Trial (EPITHET) investigators(2010) Pretreatment diffusion- and perfusion-MR lesionvolumes have a crucial influence on clinical responseto stroke thrombolysis. J Cereb Blood Flow Metab30:1214–25

Parsons MW, Yang Q, Barber PA, Darby DG, Desmond PM,Gerraty RP, Tress BM, Davis SM (2001) Perfusionmagnetic resonance imaging maps in hyperacute stroke:relative cerebral blood flow most accurately identifiestissue destined to infarct. Stroke 32:1581–7

Paschen W (1996) Disturbances of calcium homeostasiswithin the endoplasmic reticulum may contribute to thedevelopment of ischemic-cell damage. Med Hypotheses47:283–8

Paschen W, Gissel C, Linden T, Althausen S, Doutheil J(1998) Activation of gadd153 expression through tran-sient cerebral ischemia: evidence that ischemia causesendoplasmic reticulum dysfunction. Brain Res MolBrain Res 60:115–22

Powers WJ, Grubb Jr RL, Darriet D, Raichle ME (1985)Cerebral blood flow and cerebral metabolic rate ofoxygen requirements for cerebral function and viabilityin humans. J Cereb Blood Flow Metab 5:600–8

Quistorff B, Chance B, Hunding A (1977) An experimentalmodel of the Krogh tissue cylinder: two dimensionalquantitation of the oxygen gradient. Adv Exp Med Biol94:127–36

Rajdev S, Hara K, Kokubo Y, Mestril R, Dillmann W,Weinstein PR, Sharp FR (2000) Mice overexpressing ratheat shock protein 70 are protected against cerebralinfarction. Ann Neurol 47:782–91


1850


Ringer TM, Neumann-Haefelin T, Sobel RA, Moseley ME,Yenari MA (2001) Reversal of early diffusion-weightedmagnetic resonance imaging abnormalities does notnecessarily reflect tissue salvage in experimental cere-bral ischemia. Stroke 32:2362–9

Rosenberg RD, Aird WC (1999) Vascular-bed–specifichemostasis and hypercoagulable states. N Engl J Med340:1555–64

Rosner G, Graf R, Kataoka K, Heiss WD (1986) Selectivefunctional vulnerability of cortical neurons followingtransient MCA-occlusion in the cat. Stroke 17:76–82

Ruetzler CA, Furuya K, Takeda H, Hallenbeck JM (2001)Brain vessels normally undergo cyclic activation andinactivation: evidence from tumor necrosis factor-alpha,heme oxygenase-1, and manganese superoxide dismu-tase immunostaining of vessels and perivascular braincells. J Cereb Blood Flow Metab 21:244–52

Schroder M, Kaufman RJ (2005) The mammalian unfoldedprotein response. Annu Rev Biochem 74:739–89

Senger DR, Ledbetter SR, Claffey KP, Papdopoulos-SergiouA, Peruzzi CA, Detmar M (1996) Stimulation ofendothelial cell migration by vascular permeabilityfactor/vascular endothelial growth factor through co-operative mechanisms involving the aVb3 integrin,osteopontin, and thrombin. Am J Pathol 149:293–305

Sharp FR, Lu A, Tang Y, Millhorn DE (2000) Multiplemolecular penumbras after focal cerebral ischemia.J Cereb Blood Flow Metab 20:1011–32

Sokoloff L (1999) Energetics of functional activation inneural tissues. Neurochem Res 24:321–9

Somjen GG (2001) Mechanisms of spreading depressionand hypoxic spreading depression-like depolarization.Physiol Rev 81:1065–96

Spratt NJ, Howells DW, Donnan GA (2007) Imaging thepenumbra: positron emission tomography fluoromiso-nidazole. In: The Ischemic Penumbra: Pathophysiology,Imaging and Therapy (Donnan GA, Baron J-C, Davis SM,Sharp FR, eds), New York: Informa Healthcare, 149–64

Strong AJ (2009) Spreading depolarisations: Tsunamis inthe injured brain. Adv Clin Neurosci Rehab 9:32–5

Symon L (1980) The relationship between CBF, evokedpotentials and the clinical features in cerebral ischaemia.Acta Neurol Scand Suppl 78:175–90

Symon L, Pasztor E, Branston NJM (1974) The distributionand density of reduced cerebral blood flow followingacute middle cerebral artery occlusion: an experimentalstudy by the technique of hydrogen clearance inbaboons. Stroke 5:355–64

Tagaya M, Haring H-P, Stuiver I, Wagner S, Abumiya T,Lucero J, Lee P, Copeland BR, Seiffert D, del Zoppo GJ

(2001) Rapid loss of microvascular integrin expressionduring focal brain ischemia reflects neuron injury.J Cereb Blood Flow Metab 21:835–46

Tagaya M, Liu K-F, Copeland B, Seiffert D, Engler R,Garcia JH, del Zoppo GJ (1997) DNA scission after focalbrain ischemia. Temporal differences in two species.Stroke 28:1245–54

Takasawa M, Beech JS, Fryer TD, Hong YT, Hughes JL,Igase K, Jones PS, Smith R, Aigbirhio FI, Menon DK,Clark JC, Baron JC (2007) Imaging of brain hypoxiain permanent and temporary middle cerebral arteryocclusion in the rat using 18F-fluoromisonidazole andpositron emission tomography: a pilot study. J CerebBlood Flow Metab 27:679–89

Thijs VN, Adami A, Neumann-Haefelin T, Moseley ME,Albers GW (2002) Clinical and radiological correlates ofreduced cerebral blood flow measured using magneticresonance imaging. Arch Neurol 59:233–8

Toth G, Albers GW (2009) Use of MRI to estimate thetherapeutic window in acute stroke: is perfusion-weighted imaging/diffusion-weighted imaging mismatchan EPITHET for salvageable ischemic brain tissue? Stroke40:333–5

Wagner S, Tagaya M, Koziol JA, Quaranta V, del Zoppo GJ(1997) Rapid disruption of an astrocyte interaction withthe extracellular matrix mediated by integrin alpha6be-ta4 during focal cerebral ischemia/reperfusion. Stroke28:858–65

Welch WJ (1993) Heat shock proteins functioning as mole-cular chaperones: their roles in normal and stressedcells. Philos Trans R Soc Lond B Biol Sci 339:327–33

Welch WJ, Brown CR (1996) Influence of molecular andchemical chaperones on protein folding. Cell StressChaperones 1:109–15

Yenari MA (2002) Heat shock proteins and neuroprotec-tion. Adv Exp Med Biol 513:281–99

Zaro-Weber O, Moeller-Hartmann W, Heiss WD, Sobesky J(2009) The performance of MRI-based cerebral bloodflow measurements in acute and subacute stroke com-pared with 15O-water positron emission tomography:identification of penumbral flow. Stroke 40:2413–21

Zaro-Weber O, Moeller-Hartmann W, Heiss WD, Sobesky J(2010) Maps of time to maximum and time to peak formismatch definition in clinical stroke studies vali-dated with positron emission tomography. Stroke 41:2817–21

Zhan X, Kim C, Sharp FR (2008) Very brief focal ischemiasimulating transient ischemic attacks (TIAs) can injurebrain and induce Hsp70 protein. Brain Res 1234:183–97


1851


jurnals

Documents