X-Message-Number: 16178
From: 
Date: Thu, 3 May 2001 00:21:27 EDT
Subject: INTRODUCTION TO ISCHEMIA, Part III

GLOBAL ISCHEMIA: A COMPREHENSIVE INTRODUCTION, PART III

By Mike Darwin, CEO Kryos, Inc.

Part of the last installment was not transmitted to Cryonet. Therefore the 
entire section on excitotoxicity is being repeated at this time.

Excitotoxins

    A rapidly growing body of evidence indicates that excitatory 
neurotransmitters, which are released during ischemia, play an important role 
in the etiology of neuronal ischemic injury [95-97]. Those areas of the brain 
which show the most "selective vulnerability" to ischemia, such as the 
neocortex and hippocampus, are richly endowed with excitatory AMPA 
(alpha-amino-hydroxy-5-methyl-4-isoxazole proprionic acid) and NMDA 
(N-methyl-d-aspartate) glutamate receptors [98].

    Initially there was much optimism that blockade of the NMDA receptor 
would provide protection against delayed neuronal death following global 
cerebral ischemia [99], [100], [101]. The use of NMDA receptor blocking drugs 
has shown significant promise in ameliorating focal cerebral ischemic injury. 
A number of studies have demonstrated a marked reduction in the severity of 
ischemic injury in focal areas (particularly the poorly perfused "penumbra" 
surrounding the no-flow area) as a result of treatment with 
glutamate-blocking drugs such a dextrorophan [102] or the experimental 
anticonvulsant MK-801 [101].  In vitro studies with cultured neurons have 
demonstrated that excitatory neurotransmitters cause neuronal injury and 
death even in the absence of hypoxic or ischemic injury [95]. In vivo studies 
have also confirmed a massive release of glutamate and aspartate during both 
regional and global cerebral ischemia [100].

    In regional or focal cerebral ischemic injury, the NMDA receptor remains 
activated for a long period due to the prolonged interval of poor perfusion 
in the area at the edges of the infarct (the "penumbra").  However, in 
complete or global ischemia there is good resumption of blood flow following 
restoration of circulation with prompt uptake of glutamate and aspartate and 
rapid inactivation of the NMDA receptors [103]. Another factor limiting the 
role of the NMDA receptor in mediating injury in global cerebral ischemia may 
be the rapid and pronounced drop in pH which occurs in global (as opposed to 
focal) ischemia, since low pH is known to inactivate the NMDA receptor.  
These reasons are probably why NMDA receptor inhibitors have not proved 
effective in preventing global cerebral ischemic injury [104]. Recently, 
attention has turned to non-NMDA receptor antagonists such as inhibitors of 
the kainate and AMPA receptors [105], [106].

    The mechanisms by which excitotoxins cause cell injury is not yet fully 
understood. It is known that they facilitate calcium entry into neurons 
[107-109].  However, the excitatory amino acids released during cerebral 
ischemia are neurotoxic, even in cell culture where the medium is calcium 
free [96]. In the case of kainate and AMPA receptor activation, the likely 
mode of injury is sensitization of the CA1 pyramidal cells during ischemia 
such that when normal signaling is restored at the end of the ischemic 
insult, and normal intensity input from the Schaffer collaterals is resumed, 
lethal cell injury results, perhaps from abnormal calcium regulation in the 
CA1 cells or other metabolic derangements not yet understood.

Neutrophil Activation

    Since the late 1960s, polymorphonuclear leukocytes (PMNLs) and 
monocytes/macrophages have been implicated as significant causes of pathology 
in cerebral ischemia.  During the last decade there has been a veritable 
explosion of research documenting the role of PMNLs in reperfusion injury.  
Most of the initial work done in this area focused on PMNL-mediated 
reperfusion injury to the myocardium, establishing that PMNL activation and 
subsequent plugging and degranulation (resulting in release of oxidizing 
compounds) is responsible for the no-reflow phenomenon following myocardial 
ischemia [110]. In particular, the work of Engler demonstrated that PMNL 
activation is responsible for plugging at least 27% of myocardial capillaries 
and is further responsible for the development of edema and arrhythmias upon 
reperfusion [111].

    To what extent leukocyte plugging occurs in the brain following global 
cerebral ischemia remains controversial [112-114].  Anderson, et al. have 
examined the question of how rapidly leukocyte plugging occurs following 
cerebral ischemia using a bilateral carotid artery plus hypotension model in 
the dog. They noted no leukocyte plugging after 3 hours of reperfusion 
following a 40-minute ischemic episode [115].

    However, it is clear from a growing body of work that neutrophils are a 
major mediator of ischemic injury in a variety of organ systems, and that 
their acute activation is responsible for many of the effects of ischemia 
observed in the brain and other body tissues, including the loss of capillary 
integrity and the degradation of ultrastructure upon reperfusion [116]. 

    When PMNLs are activated they generate large amounts of hydrogen 
peroxide.  A large fraction of the hydrogen peroxide, aided by myeloperoxide 
(also released by activated PMNLs), reacts with the halides Cl-, Br-, or I- 
to produce their corresponding hypohalous acids (HOX) [117].  Because the 
concentration of Cl- is more than a thousand times greater than the other 
halides, the hydrogen peroxide-myeloperoxidase system probably generates Cl- 
most often in the form of HOCl.  HOCl is more commonly known as household 
bleach, and is capable of damaging a wide range of organic molecules 
including most of those that make up the structure of the cells and 
proteinaceous extracellular matrix [118], [119]. As Klebanoff has pointed 
out, the amounts of HOCl generated by the neutrophil are awesome: 106 
neutrophils can generate 2 x 107 mole of HOCl - enough to destroy 150 million 
E. Coli in a matter of milliseconds [117].

    However, the direct destructive effects of HOCl are probably limited in  
vivo by a variety of mechanisms [120].  Most probably the hypohalous acids 
act to inflict the majority of injury by interacting with collagenase, 
elastase, gelatinase, and other proteinases.  As is shown in the diagram 
below, it is now believed that the oxidants released from the neutrophil 
create a halo of oxidized alpha-1-proteinase inhibitor that allows released 
elastase (and probably others of the 20 or so known neutrophil-secreted 
proteolytic enzymes) to begin degrading the extracellular matrix, thus 
destroying capillary integrity and interfering with tissue metabolism and 
anabolism [121].

    In complete circulatory arrest, it is clear that neutrophil activation 
with accompanying release of HOCl and activation of elastase is a key factor 
in initiating the systemic cascade of inflammation/immune response which 
terminates in delayed multisystem organ failure [122]. The extent to which 
this pathway is a factor in acute global cerebral ischemic injury in cardiac 
arrest is not yet clear.

Hypoperfusion Following Reperfusion

    An apparently significant contributor to reperfusion injury is 
hypoperfusion after restoration of spontaneous circulation.  The work of 
Hossman, et al., [123] and Sterz, et al., [124] has demonstrated the critical 
importance of providing adequate circulatory support following global 
cerebral ischemia.  Loss of autonomic regulation, depressed myocardial 
function secondary to ischemic insult of the myocardium, and autonomic 
dysfunction all serve to depress MAP and cerebral perfusion following 
restoration of circulation.  Both Hossman's and Sterz's work has demonstrated 
significant improvements in neurological outcome if circulation is supported 
both extracorporeally and/or with pressors during reperfusion.

Sedimentation

A heretofore unappreciated source of reperfusion injury is the sedimentation 
of the formed elements of the blood in both large and small blood vessels 
under the influence of gravity. Sedimented red blood cells and leukocytes, 
particularly in the presence of acidosis, become sticky and adhere to each 
other. Thus, after 15 minutes of global ischemia the dependent portion of all 
areas of the vasculature become filled with a hyperviscous sludge of blood 
cells which is refractory to resuspension when perfusion is reinitiated. 

Research at Critical Care Research (CCR) in Rancho Cucamonga, CA has shown 
that this hyperviscous cell sludge is forced into the small caliber dependent 
vessels when the circulatory system is repressurized at the start of 
reperfusion. This is one of the reasons why achieving blood washout is so 
difficult in human cryopatients despite perfusion with in excess of 100 
liters of perfusate.

When flow is first restarted during an attempt to achieve blood washout in 
cryopatients after a prolonged ischemic episode the plasma column layering 
the top of the larger caliber vessels is the path of least resistance and is 
quickly replaced with perfusate, while the sludge of formed blood elements 
remains in place. As pressures and flows are increased, packed cells are 
forced into capillaries. These capillaries are already decreased in diameter 
(which typically less than that of red blood cell; 7.7 microns) from 
endothelial and parenchymal cell swelling and do not allow flow of the packed 
cell sludge. This results in uneven, patchy perfusion and prolonged or 
permanent sequestration of tissues supplied by these plugged vessels from 
restoration of flow. This interferes with adequate cryoprotectant agent (CPA) 
distribution and can lead to severe osmotic damage if vessels are eventually 
opened to flow from dehydration of adjacent tissue by CPA.

Limits To Reversibility: Histological Ultrastructural Change?

Understanding the time course and nature of the biochemical, ultrastructural 
and gross changes which occur as a result of systemic normothermic ischemia, 
with special attention to the brain (the most vulnerable organ) is essential 
to determining the realistic limits to resuscitation given known and 
foreseeable technology.

Ischemic changes in cell architecture begin almost as rapidly as ischemic 
changes in biochemistry.  Within seconds of the onset of cerebral ischemia, 
brain interstitial space almost completely disappears.  Loss of interstitial 
space is a consequence of cell swelling secondary to sodium influx and 
failure of membrane ionic regulation.  There have been several studies of the 
ultrastructural alterations associated with prolonged global cerebral 
ischemia.  Notable is the work of Kalimo et al., in the cat, [125] as well as 
Karlsson and Schultz, [126] and Van Nimwegen, et al., [127] in the rat.  
These investigators describe the following changes in common in these 
animals' brain ultrastructure after varying periods of global cerebral 
ischemia (GCI):

1) Changes At 10 Minutes

After 10 minutes of GCI, a significant number of cells (but not all) show 
clumping of nuclear chromatin and a modest increase in electron lucency 
(probably due to dilution of the cytosol by extracellular fluid).  After 30 
minutes, further changes include increased cytoplasmic swelling (particularly 
in the astrocytes), swelling and shape change of the mitochondria, and some 
loss of mitochondrial matrix density.  Microtubules disappear and there is 
detachment of the ribosomes from the cisternae of the endoplasmic reticulum.  
There is also disassociation of the polyribosomes, and single ribosomes lose 
their compact structure with associated failure of protein synthesis.  Of 
note is the stability of the lysosomes over this time course [127].

2) Changes At 60 Minutes

After 60 minutes of GCI, the above changes have become more pronounced with 
more conspicuous swelling of the ER cisternae.  The mitochondria begin to 
show slight inner matrix swelling and occasional flocculent densities 
(probably due to accumulated calcium).

3) Changes At 120 Minutes

After 120 minutes of GCI, the changes discussed above are more pronounced and 
a larger number of mitochondria exhibit the presence of flocculent densities 
evidencing calcium overload, which is currently considered irreversible.  
Published electron micrographs reveal intact lysosomes and seem to confirm 
other studies which indicate that lysosomal rupture and subsequent 
catastrophic autolysis is not a feature of early (1-4 hours) ischemic injury 
[128].

From a feasibility of reversal standpoint (i.e., likely contemporary periods 
of normothermic cardiac arrest after which successful resuscitation can be 
achieved) it is important to point out that throughout even a 
120-minute-period of normothermic cerebral ischemia, the appearance of the 
plasma membrane layers, including synapses and myelin sheaths, is only 
altered modestly.  Indeed, the first ultrastructural changes associated with 
what is currently considered lethal cell injury are to the mitochondria and 
ribosomes, and these do not usually appear until after 30 minutes of GCI. The 
extent to which these changes are irreversible remains unknown. However, by 
120 minutes of noromothermic circulatory arrest it is likely that micro and 
macrovascular clotting may be underway effectively preventing restoration of 
circulation to the organism without very technologically demanding and 
dramatic interventions such as low flow CPB with asanguineous solutions 
containing thrombolytics, sub-micron PFC-emulsion-based oxygen delivery 
capability, and complex mixtures of drugs and metabolites to restore 
homeostasis upon thrombolysis. Also required would be reversal of endothelial 
and parenchymal edema, and restoration of normothermic metabolism.

For the foreseeable future, in the absence of pre-arrest intervention, a 
period of ~30 minutes of normothermic cardiac arrest would seem the outer 
limit achievable with current technology [86].

At least one study of post-mortem ultrastructural degradation has been 
conducted on a small number of human subjects [129]. The histological and 
ultrastructural changes observed in humans with 25 to 85 minutes of GCI, 
without extensive pre-mortem brain trauma or pre-mortem cerebral no-reflow of 
prolonged duration, closely parallel those observed in animal models of GCI. 
The changes typically seen are astrocytic edema, clumping of nuclear 
chromatin, disassociation of the polyribosomes, detachment of the ribosomes 
from the ER cisternae, and swelling of the mitochondria with the presence of 
flocculent densities.  Stability of the lysosomes and conservation of the 
structure of the neuropil over this time-course are well documented.

End of Part III

[References Available Upon Request]

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=16178