NEUROPATHOLOGY FOR MEDICAL STUDENTS
Presented by William I. Rosenblum, MD
CHAPTER 3: CEREBROVASCULAR DISEASE
PRETEST: The answers are found in the text of the chapter or click on link at end of questions
Atherosclerosis and hypertension are the underlying conditions responsible for most cerebrovascular diseases. The major categories of cerebrovascular disease are caused either by rupture of a blood vessel, or by anoxia in its broadest sense. We will begin with the latter.
Anoxia of cerebral tissue can be produced by lung disease with generalized anoxia; by poisons like carbon monoxide, which binds to hemoglobin; by poisons like cyanide, which prevents oxygen in the blood stream from being utilized by the brain cell; by hypotension or cardiac arrest, which diminishes the amount of blood reaching the brain; by anemia, and by narrowing or blockade of a cerebral blood vessel or of the major vessels in the neck supplying the brain. The blockade of arteries is produced by emboli or thrombi. Thrombi form over atherosclerotic plaques. Emboli and thrombi produce infarcts, the prototype of the anoxic lesion.
If death occurs within a few hours, no gross or morphologic changes may be observed at the site of infarction. After about l2 hours, neuronal cytoplasm may begin to turn eosinophilic and nuclei are pyknotic (image below) by 24 hours, definite softening and some discoloration may be noted in the gross tissue. Accompanying breakdown of the blood-brain barrier permits edema fluid (plasma) to enter the tissue.
In a jaundiced individual, bilirubin may brightly discolor the edema around recent cerebral infarction (image below).
During the first 48 hours, neutrophils may enter the infarcted tissue, to be rapidly replaced by macrophages, which greatly increase in numbers until the infarcted tissue is about two weeks of age.
In the image above, virtually every cell is a macrophage, though only two are labeled with arrows. They then remain in large numbers for a variable period after which they decrease greatly (cyst formation). During the second week or so (all of these dates are approximate and overlapping), astrocytes begin to respond to the infarction in all of the ways described in the chapter concerning the cytopathology of the glia. Reactive astrocytosis reaches a peak somewhat later than does the increase in macrophages, and large numbers of reactive astrocytes may remain "forever" at the site of infarcts (rarified zone or cyst).
The image above displays reactive astrocytosis in a zone of rarefaction (H & E stain).
The image below show reactive astrocytosis in the form of dense PTAH stained astrocytic processes at the margin of a cyst.
The image below shows cysts representing old infarcts.
The cyst is the end stage of infarction. The larger the infarct the larger the cyst. Obviously the purpose of astrocyosis is something other than filling in the cyst and the term glial scar is a misnomer for astrocytosis.
A cyst is the endstage of any mass destruction of the brain irrespective of cause. The reactive astrocytes serve as suppliers of diverse cytokines whose role in the injured brain is yet to be understood. In addition astrocytes produce a substance[s] that induces formation of proteins in capillary walls. These proteins determine various aspects of the blood brain barrier[s]. Newly formed vessels at sites of damage do not have these new properties and one role of astrocytosis is apparently to transform these leaky new vessels into more normal brain capillaries with their blood brain barriers. This occurs as the astrocytic end feet appose themselves to the capillary. But it is not the physical barrier of the endfoot which makes the barrier but rather the diverse structural and chemical changes in the capillary endothelium whose expression is triggered by substances released from the endfeet.
PLEASE REMEMBER THE GENERAL RULES FOR DATING INFARCTS.
THESE RULES ABOUT MACROPHAGES AND ASTROCYTES ALSO APPLY TO OTHER DESTRUCTIVE LESIONS OF THE BRAIN OR CORD (E.G., TRAUMATIC LESIONS).
A few red cells pass into most infarcts and can be seen under the microscope. But a hemorrhagic appearance may not be present on gross inspection unless large amounts of red cells have passed through the damaged vessels. Only infarcts with grossly demonstrable hemorrhage are called "hemorrhagic." This is more likely to occur in infarcts produced by emboli rather than those produced by thrombi. However, red blood cells may be seen even in infarction produced by thrombi. It should be stressed that no matter how large the "hemorrhagic" component of a "hemorrhagic" infarct, the red cells do not form large aggregates or clots within the tissue, but instead remain dispersed and finely mixed with the intervening necrotic tissue. This is most important because it is so different from the character of a true intracerebral hemorrhage with which the infarct must not be confused. The picture below shows a large empty space which is artifact produced when a true hemorrhage--a big blood clot which displaced surrounding brain--fell out of the brain slice. the other brown lesions are hemorrhagic infarcts--intact but necrotic brain into which large amounts of blood has leaked.
Emboli are often multiple. Thus, when an infarct is caused by an embolus, it is often one of several infarcts having a similar age. The multiplicity of the infarcts plus their hemorrhagic character helps one arrive at the conclusion that the infarcts were embolic.
Sources of the emboli may vary from the heart or aorta, to the carotid or vertebral arteries in the neck or just entering the skull. Atrial fibrillation is a common cause of hemorrhagic infarcts since embolic atrial breaks off from thrombi that often form in the atrium or atrial appendages. This is a major reason for anticoagulating persons with persistent fibrillation.
TRANSIENT ISCHEMIC ATTACKS (TIA)
Emboli may fragment or lyse within occluded vessels before they have produced permanent damage to brain or brain blood vessels. Such episodes can produce periods of symptoms lasting only a few hours. Symptoms are often similar from one attack to the next because emboli may lodge repeatedly at the same site. Recent serial imaging studies indicate that a permanent lesion often, but not always, is present in or develops in an area producing transient symptom[s]. In such cases the reason for the disappearance of the symtpoms--the transient nature of the attack--is unclear. TIA are important because such patients are at much greater risk than others for the development during the next year of an infarct with permanent symptoms. Indeed such an infarct may appear within weeks of the TIA. thus TIA require immediate workup to identify a possibly curable cause. This cause may be atrial fibrillation [emboli from heart] or disease in the neck arteries supplying the brain.
IMPORTANCE OF PATHOLOGY IN NECK ARTERIES: Pathology in neck vessels will play a role in the development of embolic phenomena affecting the brain. Atherosclerotic plaques in these vessels may provoke local thrombosis especially if the plaque has eroded. Aggregated platelets are an important component of the local thrombus and these as well as other portions of the plaque may embolize from the thrombus to lodge downstream and produce occlusions that are transient [TIA's] or permanent [infarction] or both. In addition complete occlusion of a neck vessel may result in an immediate cerebral infarction particularly if arteries in the Circle of Willis which could bring in blood from the contralateral side are atretic or have already been occluded by atherosclerosis. In addition an occluded neck artery may set the stage for a future infarct. Indeed, large numbers of older persons may be walking around with one vertebral or carotid artery occluded. In such cases, the occluded vessel (or vessels) become an "Achilles heel" or point of weakness in so far as the effects of future occlusions, severe narrowings, or drops in blood pressure are concerned. Under such circumstances, the presence of one or more already occluded vessels may have exhausted the capacity of the collaterals, and the next occlusion, pressure drop, or little bit of additional narrowing in a still open vessel may be sufficient to reduce blood flow below the limits demanded by a normally functioning brain. In such cases, an infarct may occur in the distribution of the vessel with the old occlusion, since this is the vascular territory with the least collateral reserve. However complete occlusion frequently does not lead to infarction because the other vessels take over the flow of the occluded vessel, and the large anastomoses of the Circle of Willis permits all regions of the brain to be adequately perfused .
The importance of atherosclerosis and its complications [emboli to brain] in neck arteries has led to carotid endarterectomy as a treatment for persons with TIA. The TIA are a warning sign of impending permanent infarction caused by later shower of emboli which lodge in brain arteries or arterioles and do not lyse in time to prevent permanent brain damage. If an artery is at least 70% occluded and the artery is on the side of the lesion producing the TIA, then the patient is an acceptable candidate for endarterectomy provided the surgical team has a track record with less than 6% combined morbidity and mortality from the procedure. In such cases the reduction of future infarcts makes the procedure statistically worth while. Medical treatment of persons with TIA or persons who have had an infarct following TIA is also worthwhile. This prevents future infarction in a significant proportion of treated patients. Anti-platelet drugs are the drugs of choice.
HYPOXIC ANOXIA OR GLOBAL ISCHEMIA
Unlike thromboemboli plugging individual arteries, hypoxic anoxia produced by respiratory arrest, poisons or transient total ischemia as seen during cardiac arrest, may produce infarcts in many areas of the brain at once. Since both cerebral hemispheres will be simultaneously deprived of oxygen by these insults, the infarcts will be bilateral, and of similar age. Sometimes the basal ganglia and/or thalamus are selectively infarcted. To these hallmarks of global anoxia are added the peculiar pattern of cortical involvement. Rather than the wedge-shaped infarct involving both cortex and white matter, which is characteristic of emboli or thrombi, the infarct which follows global anoxia spares (at least relatively) the outermost and innermost cortical layers. Thus, a linear portion of the mid-zone of the cortex is involved, and this pattern of necrosis has been called pseudolaminar.
You have now completed the portion of this chapter which concerns cerebral infarction. Before proceeding to the section concerned with hemorrhage, review the appropriate questions in the pretest and see if you know the answers.
Unlike cerebral infarcts which are caused by blockade of a vessel or by anoxia or ischemia, cerebral hemorrhages are caused by rupture of a vessel or vessels. Among the causes of vessel rupture, trauma and hypertension are probably first on the list. Hypertension may produce hemorrhage by increasing the pressure within a pre-existing anatomic defect or by causing damage to the walls of small arteries that make them susceptible to rupture.
The most common defects are "berry" aneurysms and vascular malformations. The latter may consist of masses of abnormal arteries and veins, or masses of smaller vessels resembling dilated capillaries.
SUBARACHNOID HEMORRHAGE FROM BERRY ANEURYSMS
The "berry" aneurysms are saccular dilations of the vessel, which appear at points of branching in the arterial tree, usually in the Circle of Willis or at its major branch points. These out-pouchings are thought to occur at places where the media and elastica of the vessel display a congenital defect. The aneurysms may vary in size from less than a millimeter to many centimeters in diameter. The larger the aneurysm the more likely to rupture. They increase in incidence during the first three decades of life, and are often multiple. Although hypertension may increase the incidence of rupture, these aneurysms often rupture in the absence of high blood pressure. The picture below shows the arteries at the base of the brain. The patient had three berry aneurysms [red arrows].
The picture below shows a microscopic section of a berry aneurysm at the neck of the aneurysm. The slide was stained with a trichrome stain which stains connective tissue blue and smooth muscle red. The normal was is in the left half of the figure. The aneurysm wall , devoid of smooth muscle, is in the right half of the figure. The intima of the aneurysm is greatly thickened by an atherosclerosis-like process. This often happens.
The figure below shows the same aneurysm neck in a section stained with the Verhoef van Giessen stain for elastic tissue. The wavy, black, internal elastic lamella is seen close to the lumen of the vessel wall on the left. It ends abruptly where the aneurysm begins. This change in structure is the key to proving that an aneurysm is truly a berry aneurysm and not simply saccular dilation at a site of atherosclerosis.
The site of aneurysm formation may be determined in part by hemodynamic factors in the circle of Willis--for example a rudimentary communicating artery which forces more blood to go elsewhere through the Circle.
Since the aneurysms form on middle-sized or small arteries within the subarachnoid space, the hemorrhage always begins as a subarachnoid hemorrhage. The hemorrhage often dissects into the brain, however, so that symptoms of an intracerebral lesion are common. The hemorrhage may even dissect through the brain and re-enter the cerebrospinal fluid via the ventricle into which it has ruptured.
Subarachnoid hemorrhage often leads to an accompanying infarct. The reason why subarachnoid hemorrhage causes cerebral infarction is not altogether clear, but may involve interruption of blood supply due to rupture of vessels, kinking of vessels by the hemorrhage, and spasm of the vessels because of the presence in the blood of some vasospastic material. Indeed progressive spasm, demonstrable on angiograms, can occur following subarachnoid hemorrhage and/or its surgical treatment. The presence of severe generalized spasm is a bad prognostic sign.
Ischemia is present at the margin of hemorrhages and is a result of vasospasm. This can produce infarction and/or edema. The edema, together with the increased intracranial mass produced by the hemorrhage itself, causes an increased intracranial pressure which may be lethal.
Subarachnoid hemorrhage due to rupture aneurysm is frequently lethal due to the factors just mentioned and in patients who recover there is a high incidence or rebleeding with , again, a high mortality and morbidity. Treatment is clipping of the neck of the aneurysm or occluding the aneurysm with a small metal coil. However, due to spasm in patients who have already bled, morbidity may be high even after successful clipping or occlusion.
We have mentioned that hypertension may increase the incidence with which berry aneurysms rupture, and may contribute to rupture of vascular malformations. Hypertension appears to be more directly involved, however, in the major cause of intracerebral hemorrhage, rupture of a small arteriole within the brain.
Hypertension causes fibrinoid necrosis of these penetrating arterioles. The massive intracerebral hemorrhage which is a complication of hypertension, arises from rupture of a necrotic arteriole or from rupture of a minute "miliary" aneurysm formed at the site of necrosis. These aneurysms were first described by CHARCOT and BOUCHARD. The frequency of fibrinoid necrosis and miliary aneurysm formation in vessels within basal ganglia and thalamus accounts for the frequency of intracerebral hemorrhage in those locations. Fibrinoid is identified by its structureless or sometimes granular red appearance on H&E stain and by the fact that , unlike hyalinized smooth muscle which is also eosinophilic, the fibrinoid areas stain with stains for fibrin such as PTAH or Putz stain or with certain trichrome stains. The fibrinoid change in these vessels was called lipohyalinosis by Miller-Fisher in a very influential series of articles. However that term is confusing because hyalinized arteries are arteries whose media has undergone a pathologic change which is not fibrinoid necrosis and which by itself does not lead to rupture. Indeed hyalinized arterioles are common in hypertension. The term lipohyalinosis stresses the presence of fat in the degenerate arteriolar wall but again this change is not the hallmark of the arterioles that are in danger of rupturing or forming miliary aneurysms. The fibrinoid change is the critical change in these diseased arteriolar segments and looks and stains just like the fibrinoid seen in renal and other arterioles in malignant hypertension. The important point to remember is that, for unknown reasons, the brain arterioles can undergo fibrinoid necrosis even in so-called benign hypertension--that is in patents with only modest blood pressure elevation. For that reason it is important to treat even benign hypertension. The series of pictures below illustrates the pathologic processes that can lead to rupture.
The pictue below shows the wall of an arteriole stained with H&E. The amorphous pink [eosinophilic] material in the wall is fibrinoid.
The section below was stained with azocarmine. An arteriole in the subarachnoid space has an amorphous red material occupying a good portion of its wall. This is fibrinoid. Collagen or hyalinized collage would have stained blue.Fibrinoid is frequently segmental in distribution so that the entire circumference may not be involved and other areas along the length of the vessel may also be spared.
The slide below was also stained with azocarmine. The arteriole wall is replaced by red fibrinoid and displays aneurysmal dilation.
Sometimes a miliary aneurysm thrombosis rather than ruptures. It then appears as a fibrous ball which may be separated from the parent vessel due to the plane at which the section has been cut. If the section is close to the parent arteriole there will be elastic tissue at the margin of the ball. This elastic tissue stains black with the VVG stain in the pictures below.;
The pathologist got lucky when the section below was taken. Here a miliary aneurysm that has neen converted to a fibrous ball or globe is shown in longitudinal section still connected to the parent arteriole by a thin neck. The wisps of elastica in the upper part of the fibrous globe prove that this is not simply an organized clot arising from a vessel rupture but is, rather, an aneurysmal extension of the nearby parent arteriole. The blood that once entered the aneurysm through its neck has clotted and organized.
The intracerebral hemorrhage produced by rupture of a miliary aneurysm or of a necrotic vessel first appears as a large space-occupying mass (image below).
Thus, if the clot were to be dislodged as it sometimes is at the autopsy table, a large cavity is left behind (image below). In this picture there are other hemorrhagic lesions. These are hemorrhagic infarcts. Note that the brain, albeit infarcted, is still present in these areas into which there has been leakage of large amounts of red blood cells.
Necrotic tissue is present at the periphery of the clot, but not within it. The necrosis at the periphery is histologically identical to that seen in infarcts, and is produced by interruption of the blood supply due to broken blood vessels, and compression of tissue. When the hemorrhage itself resolves, it does so via the macrophage, which carries away the blood pigment as hemosiderin.
As resolution occurs, the mass of clot becomes smaller and smaller, and the edges of the displaced tissue around the clot begin to come closer together. Finally, a linear slit will remain as the only sign of what was a large oval hemorrhage (image below).
Residual, hemosiderin-laden macrophages may impart an orange color to the wall of the slit or cyst. In other words, if the original hemorrhage is compatible with survival of the patient, the actual tissue damage and residual symptoms may be considerably less than those produced by an infarct of comparable size, since the infarcted mass is all dead brain, while the original hemorrhage destroys brain only along the path cleaved by the hemorrhage and at the periphery of the mass of blood.
In hypertensive hemorrhages the hemorrhage is generally in the basal ganglia or pons. The arteries to these areas are short branches from more major vessels and so the pressure within them is relatively high. Thus the location of the microvascular changes in these portions of the vascular tree suggests that blood pressure level has something to do with the fibrinoid degeneration. In contrast, there are hemorrhages in the peripheral portions of the various "lobes" of the cerebrum--e.g. frontal lobe or occipital lobe. These hemorrhages are called lobar to distinguish them from more centrally located hypertensive hemorrhages. Their cause is usually deposition of beta A4 amyloid--the same amyloid as is deposited in Alzheimers disease. They tend to occur in individuals over 60 years old. Fibrinoid has also been reported in adjacent vessels in such cases. Such reports deny the presence of hypertension in these people. If true, then this would represent the only entity other than hypertension in which arterioles of the brain have undergone fibrinoid necrosis.
TEST YOUR KNOWLEDGE ABOUT HEMORRHAGE BY REVIEWING THE PERTINENT QUESTIONS IN THE PRETEST AT BEGINNING OF CHAPTER.
Unfortunately, death often results from intracerebral hemorrhage during its acute or subacute stages. This death is caused by the increased intracranial pressure produced either by the hemorrhagic mass itself or by the associated cerebral edema. The final common pathway to death (increased intracranial pressure) is then identical to that often seen in infarction, brain trauma and tumor. Since increased intracranial pressure produced by edema is often the ultimate cause of death in cerebral vascular disease, and since edema is, in itself, a basically vascular phenomenon, it seems appropriate to discuss it here in greater detail.
TYPES OF EDEMA--There are 2 major types of edema--cytotoxic or intracellular on the one hand and vasogenic on the other. Since edematous tissue must show a net increase in water to meet the definition of edema, cell swelling alone cannot cause edema unless the volume of the swollen cells occurs without equalizing diminution of the extracellular space, as might occur if the swollen cells encroached upon and squeezed that space. Animal studies suggest that cellular swelling of that magnitude can occur and this would be true cytotoxic edema. For unknown reasons cytotoxic edema occurs preferentially in grey matter. Therefore one might think that the neurons swell. But, in fact, while neuronal dendritic processes do swell, the neuronal cell bodies do not. In fact, in animal models of ischemia the neurons shrink. It is the astrocytes that swell. The swelling begins within 30 seconds of hypoxia due to disabling of ionic pumps by the energy shortage produced by hypoxia.The net effect on brain water content and brain volume [i.e. brain edema] depends upon the relative contributions of the swollen and shrunken cells as well as the volume of the extracellular space. It is unclear whether cytoxic edema by itself can produce increases of intracranial pressure sufficient to produce produce herniation of the brain during infarction, . In any case within hours the venules and capillaries become leaky and both protein and water leak into the extracellular space. This is vasogenic edema. It is responsible for the morbidity and mortality of infarction, hemorrhage, infections, etc. because the increased intracranial pressure compromises the brains function as explained in the section below concerning herniations.
Present therapy of edema consists primarily of infusion of hypertonic solutions into the blood stream. These solutions contain large molecules which cannot easily leave the blood stream. The rationale for this therapy is the idea that such fluids will retard the leakage of vasogenic edema fluid from the vessels and cause edema fluid to move from the tissues back into the vascular compartment. However, oncotically active substances like mannitol will leave the leaking vessels, therefore the actual removal of brain water in patients treated with these solutions, is from the areas with intact vessels. Because the skull is relatively sealed this removal of water will reduce overall intracranial pressure. In addition. it has been found in experimental studies, that such fluids may reduce intracranial pressure by another means: hemodilution, which decreases hematocrit, blood viscosity and shear. Decreased shear reduces the release of dilating local messengers [e.g nitric oxide] from the endothelium This reduction in local dilators causes vasoconstriction and thus decreases intravascular volume within the skull. The decreased intravascular volume decreases intracranial pressure.
Steroids have also been used to treat edema. Their mechanism of action is unknown and they have not been shown to be beneficial in treating the edema of infarction..
CONSEQUENCES OF EDEMA:
As suggested above, the most common cause of edema that produces brain swelling with a clinically important elevation of intracranial pressure is either endothelial injury in capillaries and venules or more massive damage to blood vessels. This edema is VASOGENIC edema. For reasons that are not definitely established, the edema fluid, which is initially comparable to blood plasma, passes predominantly into white matter, rather than grey. This is true even when the lesion is in the grey, so that the edema fluid enters the brain at that point. The fluid may travel considerable distances in the white matter.
The image above illustrates some inportant points about vasogenic edema. Fortuitously the edema in the picture has been stained green by bilrubin in the plasma which leaked out of the vessels. Formalin changed it to biliverdin which is green. Note the color appears to spare [almost] the arcuate zones [arrows] of white matter which underlies the cortex. We are not sure why the arcuate zone is relatively spared by vasogenic edema. In any case, edema can not only cause brain swelling but, by interfering with white matter nutrition, it can cause degeneration of the affected white matter. If the patient survives, the affected areas of white matter become rarified or cystic and the borders of the original lesion, for example an infarct or a contusion, are extended by the adjacent zone of damage produced by the edema. These areas of extension are characterized by the fact that they occupy the deeper white matter coniguous with the original lesion and spare the arcute zone--also known as the "U" fibers.Vasogenic edema is the only basic pathophysiologic process known to distribute itself in the deep white matter with relative sparing of "U" fibers.So when we see a cystic or semicystic lesion whose boundaries are demarcated as just described we know that edema added to the effects of the original insult.
The image below shows an abscess near the grey-white junction. The lesion is surrounded by greatly expanded white matter, the expansion of which was caused by edema fluid entering where blood brain barrier broke down near the abscess, and spreading great distances between myelinated white matter axons.
The image above shows a great increase in white matter mass especially on the right. This is another level through the brain of the case illustrated in the previous image.But here we are beyond the site of the abscess.The edema fluidhas spread through white matter remote from the source which is the leaky vessels within or around the abscess.The increased mass within the edematous hemisphere causes a bulging and flattening of the cerebral surface, unless a defect is present in the skull. In the latter case, the surface beneath the defect simply protrudes through the defect, while the remainder flattens against the inner aspect of the intact skull.
Not only can an edematous brain herniate through a hole in the skull, but shifts of cerebral tissue can also occur within the skull. When one hemisphere is swollen, it may be displaced toward the opposite side of the skull and a portion of the cingulate gyrus may be forced under the falx. Similarly, the temporal lobe may be forced downward and a portion of its medial aspect may be forced under the tentorium. This portion of the temporal lobe is called the uncus, and the term "uncal herniation" is applied to this lesion. Prior to the stage of herniation, the uncus may be pushed against the sharp edge of the tentorium causing the tentorium to make an "uncal groove" along the medial surface of the temporal lobe. A swollen hemisphere may also force the brainstem toward the opposite side of the skull producing a notch in the contralateral peduncle where it is compressed against the edge of the tentorium on the side opposite the swollen hemisphere.. This is known as Kernohan's notch. These compressions are not merely morphologic alterations, but are accompanied by malfunction of the compressed tissue. In addition, cranial nerves or vessels may be compressed by the swelling or displaced brain. For example, compression of posterior cerebral arteries may produce infarcts in the distribution of these vessels. Thus, one cerebral infarct can, through the consequences of an accompanying edema, produce a second infarct in a distant part of the brain. Increased intracranial pressure can also cause the brain stem to herniate downward in an attempt to relieve pressure through the foramen magnum. When this occurs the vessels to the stem are stretched and tear. This produces secondary brain stem hemorrhage. The hemorrhage and/or related damage to vital cardiorespiratory centers in the brain stem results in death and this is frequently the ultimate cause of death in patients with infarct, hemorrhages, traumatic injuries, etc. which originally affected only the cerebrum.
THE FIGURE ILLUSTRATES A SECONDARY BRAIN STEM HEMORRHAGE.
TEST YOUR KNOWLEDGE ABOUT CEREBRAL EDEMA BY REVIEWING PERTINENT QUESTIONS IN THE PRETEST.