вход по аккаунту


Brain death II. Neuropathological correlation with the radioisotopic bolus technique for evaluation of critical deficit of cerebral blood flow

код для вставкиСкачать
Brain Death: 11. Neuropathological
Correlation with the Radioisotopic Bolus
Technique for Evaluation of Critical
Deficit of Cerebral Blood Flow
J o h n Pearson, MD, Julius K o r e i n , MD, James H. Harris, MD, PhD,
Melvin W i c h t e r , MD, and Philip Braunstein, MD
Diffuse necrosis and autolysis were f o u n d in the brains f r o m 6 comatose, respirator-supported patients i n whom the
bolus technique demonstrated no cerebral blood flow d u r i n g a period exceeding 20 hours p r i o r to cardiac death.
W h e n blood flow was insufficient to p r o d u c e a bolus, t h e r e was no evidence of active tissue response to necrosis. In 6
similar patients, w h e n cerebral blood flow was detected b y the bolus technique, less extensive necrosis was observed
postmortem a n d there was active tissue response.
Pearson J, Korein J, Harris JH, et al: Brain death: 11. Neuropathological correlation with the radioisotopic bolus
technique for evaluation of critical deficit of cerebral blood flow. Ann Neurol 2:206-210, 1977
In this paper critical deficit of cerebral blood flow
(CBF), as d e t e r m i n e d b y the bolus t e c h n i q u e [6], is
correlated with neuropathological findings i n the
brains of comatose, apneic patients who suffered cardiac death 20 or more hours after bolus studies. The
2 0 - h o u r survival p e r i o d was selected i n order that
those brains i n which blood flow persisted would have
time to d e v e l o p inflammatory a n d vascular responses.
All patients satisfied the criteria for entry into the NINDS
Collaborative Study on Cerebral Survival [ 11; that is, they
had been in unresponsive coma without spontaneous
breathing, requiring a respirator for at least 15 minutes.
The patients remained aprieic if removed from the respirator for up to 6 minutes with 100% oxygen delivered to
them passively, and they had no cephalic reflexes. All had
periodic complete medical and neurological examinations
and initial toxicological tests, repeated electroencephalographic studies, and continuous electrocardiographic
monitoring. Patients were treated appropriately and followed for three months or until irreversible cardiac
standstill occurred, either spontaneously or following cessation of support after the EEG had been isoelectric for 24
hours in the absence of toxins.
Thirteen patienrs surviving 20 hours or more after bolus
studies form the basis of this report. Six of the 13 had no
bolus, 1 had an initial bolus followed by an "intermediate"
trace, and 6 had a bolus present. Mean intervals from bolus
studies to death were 21/2 days for the patients with no bolus
and 5y2 days for those with a bolus present.
The bolus technique used for estimating the presence or
absence of high CBF has been described previously [2,7]. It
consists of rapid intravenous injection of 2 mCi of
technetium 99m pertechnetate in a volume of 2 ml, followed by monitoring of this radioactive bolus as it passes
collimated detectors sensitive only to the contents of supratentorial cranial vessels. A cerebral bolus causes marked
excursion of a p e n recorder only if flow higher than 24% of
normal is present [71. If CBF is below 20% of normal, the
trace shows n o bolus [6]. Occasionally, extracerebral upper
cranial blood flow (for instance, through the skin) produces
a small but readily distinguishable deviation of the baseline
trace. An intermediate trace can also be caused by aberrant
intracranial passage of blood considered incompatible with
cerebral viability [61. A control detector is placed over a
large extracranial artery (femoral), where alarge excursion is
recorded in the presence of adequate injection technique
and persistent systemic circulation.
In these studies, clectrocercbral silence (isoelectric EEG)
was defined as absence of electrical potencials indisputably
cerebral in origin which exceeded 2 pv under standardized
recording conditions [6].
Brains were fixed for one to two weeks in 20% formalin
acidified by addition of 1% acetic acid. Following gross
description and photography, a standard set of ten blocks
were studied, supplemented as indicated by examination of
specific lesions and of spinal cord when available. The bulk
of the microscopical data was derived from sections stained
From the Departments of Pathology and Neurology and the Division of Nuclear Medicine, New York Universiry Medical Center
and Bellevue Hospital Center, New York, NY.
Address reprinc requests ro Dr Pearson, Department of Pathology,
New York University Medical Center, 550 Firsr Ave, New York,
N Y 10016.
Patients Studied
Accepted for publication Apr 26, 1977.
with hemotoxylin and eosin and with Luxol fast blue/
periodic acid-Schiff for myelin.
Brains from patients having no bolus, indicating a
critical deficit of CBF, had congested, dusky leptomeninges, were diffusely soft, and showed marked
herniations sometimes associated with tissue disintegration and displacement into the spinal subarachnoid
space. The brainstem and cerebellum were involved
in the diffuse process. Active responses in the form of
macrophage migration and endothelial proliferation
were characteristically absent. Astrocytic reaction was
not a feature. The grayish, discolored deep tissue
often appeared to contain less blood than normal.
Microscopically the changes could be most succinctly described as autolysis. (Autolysis refers to tissue breakdown that is dependent on intrinsic cellular
enzymes and independent of factors derived from
flowing blood.) Some blood vessel walls showed blurring of endothelial detaiI with indistinct cell margins,
nuclear pallor, and diffuse, slight basophilia. Glial cytoplasm was either wispy and eosinophilic or not
visible, and nuclei showed changes ranging from extreme pallor to marked pyknosis with irregularity of
their outline. Neurons showed similar nuclear variations. Their cytoplasm was indistinctly vacuolated or
homogeneous and smudgily basophilic, eosinophilic,
or extremely pale. Very few neurons had normal characteristics. Frequently all cells in a section were pale
and appeared ghostlike. The changes were distinct
from those seen in delayed autopsies of brains cooled
during the postmortem period and in which blood
flow had persisted until close to the moment of death.
The changes observed can be reproduced by keeping
brain moist and maintaining it at 37°C for 24 hours
from the time of death. In some cases pathological
lesions such as trauma o r hemorrhage preceded the
cessation of blood flow, and the components of such
lesions also showed autolysis.
T h e central nervous system of patients in whom
adequacy of CBF was demonstrated by the presence
of a bolus had less severe pathological changes, and
these tended to be focal rather than diffuse. None
showed as diffuse superficial congestion or as severe
softening as was seen in the patients with no bolus. At
most, there was moderate cerebral softening and congestion (2 cases). Microscopically, even these patients
showed evidence of active macrophage and astrocytic
responses. Herniation of brain tissue was usually mild
or absent. Minimal brainstem changes were seen in 1
of these patients, but in all others the brainstem was
grossly normal. Most patients had focal lesions in
which neurons tended t o be eosinophilic or near normal in appearance. Glial cells showed reactive
changes. Macrophage response was invariably pres-
ent, and endothelial proliferation was common.
Minor autolytic changes were seen in small regions in
some cases, but they were never diffuse and in 2
instances were in zones of coagulative necrosis. Vessels never displayed autolytic changes.
Congestion of deep blood vessels was not prominent in either group. When congestion was present in
the patients with a bolus, the vessels contained red
cells that were not usually lysed. In the group showing
no bolus, vessel contents were usually lysed and autolysis included vessel walls in 5 cases. Absence of vasczLlar filling with fresh blood, lack of endothelial swelling,
and absence of macrophage response i n the presence of
autolysis are taken as pathological confirmation of insignificant blood f l o w i n the patients with no bolus (Figure).
A patient in whom a bolus was present at first but
who later had an “intermediate” tracing [6] experienced a massive primary brainstem hemorrhage. The
pathological changes were intermediate in severity
between the bolus and no bolus groups. There was no
herniation. Slight tissue reaction and fresh blood clot
within the diffusely softened, necrotic brain indicated
persistence of some blood flow after the initial catastrophe.
The overall results of the study are presented in the
Table. The pathological changes indicated were subjectively graded on a scale of 0 to 5 +, and means of the
scores are indicated. (Individual case histories and
pathological findings have been omitted for the sake
of brevity but are available on request.)
The bolus technique permits a diagnosis of cerebral
death within 1 hour of its application to a patient with
clinical signs of brain death. The term cerebral death is
used to indicate death of all or the great majority of
cells in supratentorial structures. It is in those structures that the bolus technique measures blood flow.
Brain death refers to similar necrosis of both supratentorial and infratentorial structures.
All patients with rwo “no bolus” traces have died.
Conceivably, a single study could detect transient suppression of CBF; it is extremely unlikely that two
studies would coincide with such transient suppressions. In none of the patients who had repeat studies
has there been any evidence of return of an initially
absent bolus. Irreversible cellular degeneration is
present after 20 minutes of ischemia [SI. Two negative
bolus studies 1 hour apart indicate a condition incompatible with cerebral survival. In practice, absence of
bolus activity invariably correlated with diffuse necrosis and autolysis of the entire brain. Thus, the use of
this simple technique for bedside estimation of CBF
in the prediction of brain death appears validated.
In experimental primates, irreversible changes inevitably leading to death of the animal may occur after
Pearson et al: Neuropathological, Bolus, and Blood Flow Correlates of Brain Death
10 minutes of total ischemia [ l l ] . Some electrophysiological signs of persistent neuronal function
have been reported after 1 hour of ischemia in animals
continuously anesthetized with barbiturates [5]. The
latter study is open to criticism in that the quality of
clinical survival was not studied and because the presence of barbiturates may markedly prolong the period
during which the brain can survive hypoxia and
ischemia [lo]. During total ischemia the energy reserves of the brain available for anaerobic metabolism
are depleted in approximately 5 minutes, after which
irreversible metabolic changes may occur [ 121. Electron microscopical changes in the form of swollen
neuronal mitochondria are evident after 15 minutes of
ischemia but do not necessarily reflect irreversible
damage 131.
While the bolus technique is designed only to detect cerebral circulation (posterior fossa studies being
prone to interference from flow in major extracranial
vessels), it has, in conjunction with clinical criteria,
predicted necrosis of the entire brain. A critical deficit
of CBF was always associated with severe cerebral
swelling and prominent herniation effects.
The objective of the current study was to test the
utility of the bolus technique and to correlate the results with neuropathological changes rather than t o
demonstrate the limitations of pathological methods.
Eosinophilia of neurons may occur as early as 6%
hours after necrosis [3], but 10 to 12 hours is a more
208 Annals of Neurology
Vol 2 No 3
(A) Bolzls absent. Vascular contents are lysed, the vessel wall
i s necrotic, and there is n o macrophage response to deud cells
in tissue. ( B ) Bolus present. Red cells i n vessels are intact, the
vessel U J U ~isI reactive, and there is a prominent macrophage
re.rponse t o tissue necrosis. (Both X 3 1 5 before lQ%
generally observed interval. This change cannot indicate the status of central nervous system blood flow
prior to systemic death, since eosinophilia can occur
when flow returns (a5 in some infarcts) or i n the continued absence of flow (as in coagulative necrosis). i n
the presence of blood flow, neutrophil and macrophage responses to necrosis begin after approximately
18 to 20 hours. Since we were searching for pathological evidence of the presence or absence of physiologically significant circulation, we have included in this
report only those patients surviving 20 hours after
bolus studies. The bolus studies were taken as the
initial timing event since it was their value in predicting diffuse brain death and autolysis that we wished to
test. The status of CBF was unknown prior to bolus
A patient whom we observed but excluded from the
current study illustrates the reason for insisting on the
20-hour interval. The patient experienced transient
cardiorespiratory arrest due to heroin and barbiturate
intoxication. Four and one-half hours after she met
the criteria for inclusion in the study, a head bolus was
September 1977
Puthological Changes in the Bruin in 6 Patients ziith und 6 without a Rudinactave B n h s a
N o Bolus
No Bolus
Neuronal eosinophilia
Astrocytic reaction
aValues are means of subjective scores on a scale of 0 to 5 + . Note that patients with no bolus had agreater degree of softening and herniation
hut less macrophage and astrocytic response than those who had a bolus. When congestion was present in the patients with no bolus, vessel
contenrs were usually lysed. Eosinophilia of neurons does not discriminate between the two groups of patients.
present. Five and one-half hours later she died and was
found to have eosinophilia of all neurons in the spinal
cord and brain. The period to death w a s inadequate to
permit development of reactive changes in endothelium or migration of leukocytes. Hence it was
not possible on histological grounds to judge whether
or not blood flow had returned after the initial cardiac
arrest and neuronal necrosis. In the early hours after
necrosis, autolytic or vascular changes are not sufficiently developed to permit histological verification
of clinical estimates of CBF.
The pathological changes seen in the present study
resemble those attributed to partial failure of brain
circulation in hypoxemia and do not show the predilection for brainstem and spinal cord which is said to
occur experimentally after transient episodes of total
ischemia or anoxia [lo, 121. Whatever the conditions
of blood flow or arterial oxygen tension leading to
death of the brain in the patients with no bolus, it is
evident from the lack of active macrophage response
after an interval of 20 hours or more that circulation
never recurred.
An attempt has been made to define respirator brain
(which we prefer to call brain death and autolysis) by
polling opinions from a large group of neuropathologists 191. While agreeing with many of
the favored criteria, we differ in some observations.
Reactive changes are not seen in astrocytes, nor is
congestion of deep vessels a prominent feature; indeed, empty and even autolyted blood vessels are
more likely to be found. Relative diencephalic preservation is not observed. We confirm the absence of
inflammatory, microglial, and vascular response [4].
The term respirator brain has been avoided since we
believe that i t is not the presence of a respirator but
the absence of major intracranial blood flow which is
associated with brain death and autolysis. Terms such
as nonperfused brain or panencephaloiJchemia might be
preferable if they were not so inelegant. The term
bloodleu brain would be evocative of the situation but
inaccurate since blood may persist in preexisting
hemorrhages or in vessels congested at the time brain
circulation ceases.
We consider unjustified the term partial respirator
brain [13], which implies partially dead brain. Focal
necrosis of brain is commonplace in surviving patients
with a wide variety of diseases. Partial survival of the
brain improves the prognosis of comatose patients,
and we have observed 5 survivors among 30 patients
with bolus evidence of persistent blood flow. That 25
of the patients with a bolus died is not surprising.
Extensive brain damage may exist in the presence of
good blood flow through those tissues which remain
intact .
Braiw from patients with detectable cerebral flow
(“bolus present”) usually resembled those most commonly encountered at autopsy, being firm and having
leptomeninges that were not markedly congested or
cyanotic. Softening, hemorrhage, necrosis, and
edema, when present, were focal rather than diffuse.
Herniation was either absent or explicable in terms of
the focal lesions. Posterior fossa structures were usually intact. In 2 of the patients who had a bolus present, the gross changes partially resembled those of
cerebral death, but microscopically the brains were
distinct from “no bolus” specimens in showing foci of
established macrophage response and reactive astrocytosis. It is possible that blood flow ceased at some
time between bolus studies and death in these 2 patients and some of the others who, when tested, had a
trace showing a bolus.
The pathological findings confirm that the bolus
technique distinguishes distinct sets of comatose patients. When critical deficits of blood flow are detected repeatedly by bolus study, autolysis of the brain
confirms the absence of cerebral circulation. An absent bolus signifies cerebral death and thus indicates
irreversible coma.
Supported in part by National Institutes of Health Contract 7 1 2319 and by Neurology Research Fund Nu. 8-0168-708, New
York University Medical Center.
Pearson et al: Neuropathological, Bolus, and Blood Flow Correlates of Brain Death 209
1. An appraisal of the criteria of cerebral death-a summary
statement. (Final report: Collaborative Study of Cerebral Survival. HEW, NIH, NINDS, Contract 1-NS-1-23 16. Bethesda,
MD, Apr 1974.)JAMA 237:982-986, 1977
2. Braunstein P, KricheffI, Korein J, et al: Cerebral death: a rapid
and reliable diagnostic adjunct using radioisotopes. J Nucl Med
14:122-124, 1973
3. Brierley JB, Meldrum BS, Brown AW: The threshold and
neuropathology of cerebral "anoxic-ischemic" cell damage.
Arch Neurol 29:367-373, 1973
4. Grunnet ML, Paulson G: Pathological changes in irreversible
brain death. Dis Nerv Syst 32:690-694, 1971
5 . Hossmann K-A, Kleihues P: Reversibility of ischemic brain
damage. Arch Neurol 29:375-384, 1973
6. Korein J, Braunstein P, George A, et al: Brain death: I.
Angiographic correlation with the radioisotopic bolus technique for evaluation of critical deficit of cerebral blood flow.
Ann Neurol, pp 195-205 this issue
7. Korein J, Braunstein P, Kricheff I, et al: Radioisotopic bolus
210 Annals of Neurology Vol 2
No 3
September 1977
technique as a test to detect circulatory deficit associated with
cerebral death: 142 studies in 8 0 patients of an innocuous IV
procedure as an adjunct in the diagnosis of cerebral death.
Circulation 5 1:924-939, 1975
Matakos G, Ceros-Navarro J, Schneider H : Experimental
brain death: I. Morphology and fine structure of the brain. J
Neurol Neurosurg Psychiatry 36497-508, 1973
Moseley JI, Molinari GF, Walker AE: Respirator brain: report
of a survey and review of current concepts. Arch Pathol Lab
Med 100:61-73, 1976
Myers RE: Two classes of dysergic brain abnormality and their
conditions of occurrence. Arch Neurol 29394-399, 1973
Myers RE, Yamaguchi M: Effects of serum glucose concentration on brain response to circulatory arrest (abstract). J
Neuropathol Exp Neurol 35:301, 1976
Siesjo BK, Ljunggren B: Cerebral energy reserves after prolonged hypoxia and ischemia. Arch Neurol29:400-407,1973
Walker AE, Diamond EL, Moseley J: The neuropathological
findings in irreversible coma. J Neuropathol Exp Neurol
34:295-323, 1975
Без категории
Размер файла
529 Кб
flow, death, neuropathological, evaluation, critical, brain, cerebral, blood, bolus, correlation, radioisotope, deficit, techniques
Пожаловаться на содержимое документа