171

Resuscitation, 15 (1987) 171-185
Elsevier Scientific Publishers Ireland Ltd.

LATE NEUROPATHOLOGICAL
STRANGULATION

CONSEQUENCES

RICHARD K. SIMPSON, Jr.‘, J. CLAY
CARAWAYb and DAVID S. BASKIN%’
“Department of Neurosurgery
Houston, TX 77030 (U.S.A.)

OF

GOODMANb, EMILIE

and bDepartment

of Pathology,

Baylor

ROUAHb, NANCY
College of Medicine,

(Received September 16th, 1986)
(Revision received December 22nd, 1986)
(Accepted January 2nd, 1987)

SUMMARY

A case of a young man who was a victim of strangulation is presented.
He arrived at the hospital in refractory status epilepticus, controlled only
with intravenous pentobarbital. The initial CT scan showed mild cortical
edema. Two days later, a CT scan showed diffuse cortical swelling and bilateral basal ganglia infarcts. Upon discontinuation of pentobarbital therapy,
his neurological examination revealed spontaneous ventilation and a gag reflex. A CT scan 4 weeks after the insult demonstrated hypodensities in both
cerebral hemispheres and hydrocephalus. EEG was isoelectric throughout
his hospitalization. He survived nearly 5 months and succumbed to pneumonia. Neuropathological examination demonstrated severe encephalomalacia, multiple cystic infarcts and generalized compensatory ventriculomegaly.
Microscopic examination was particularly remarkable for a pronounced gemistocytic astrocyte proliferation in the white matter. This case illustrates
the long-term neuropathological consequences of severe, global hypoxia/
ischemia and the paucity of intact brain required to maintain a persistent
vegetative state.
Key

words:

Persistent vegetative state - Neuropathology - Brain death

INTRODUCTION

Over 3500 people each year are victims of fatal hanging and
strangulation [l]. Survival from such insults or global ischemia/hypoxia
from a variety of other causes is becoming more frequent due to modem
*To whom correspondence should be sent.
o Elsevier Scientific Publishers Ireland Ltd.
0300-9572/87/$03.50
Printed and Published in Ireland

172

Fig. 1. Computed tomograms (CT) of the head. CT done emergently (a) reveals fattened gyri
and no visible sulci (straight arrow) or Sylvian fissures (curved arrow). CT done 2 days after
admission (b) reveals no visible parapineal &tens, small ventricles (straight arrow) and lucenties within the basal ganglia (curved arrow). CT done 1 month after admission (c) reveals diffuse cerebral lucencies (outlined arrow) and grossly enlarged ventricles (curved arrow).

resuscitative measures and critical care techniques [2--51. Neurological
recovery in survivors is variable with many patients being rendered
persistently comatose. Most succumb within days to weeks although some
may survive in this vegetative state for months to years [4,6].
Neuropathological studies are uncommon in these long-term survivors; the
present case illustrates the tissue changes in such a patient.
CASE REPORT

A 21-year-old white man developed generalized seizure activity following
an altercation 4 h earlier in which he was the victim of a vigorous ‘choke
hold’. When this strangling maneuver was applied he lost consciousness,
but the assailant, believing the patient to be intoxicated and dangerous,
continued the strangulation for several minutes. When released, the patient
was initially flaccid and unresponsive, but then had a generalized tonicclonic seizure. While in transit to the hospital, he had brief periods of generalized seizure activity interrupted by variable periods of quiescence. There

174

was no history of head trauma. His past medical history was unremarkable
except for reports of alcohol and drug abuse.
Examination revealed a blood pressure of 150170, pulse lOO/min and respirations 12lmin. Initial neurological examination showed his pupils to be 3
mm in diameter, equal, and reactive to light. He had weak comeal and gag
reflexes. Sensory testing revealed withdrawal of all extremities to noxious
stimulation. Examination of motor function showed diffusely increased
tone. He localized bilaterally to deep pain. Tendon reflexes were normal in
the upper extremities but were hyperactive in both lower extremities. No
Babinski’s responses were seen and ankle clonus was present bilaterally.
The patient’s general physical examination was unremarkable. Routine laboratory tests were normal. A computed tomogram (CT) scan showed mild cerebral edema (Fig. la).
Soon after admission, he developed sustained generalized seizure activity
which did not respond to maximal doses of intravenous diazepam,
phenytoin, phenobarbital or paraldehyde. He was intubated, ventilated, and
treated with intravenous pentobarbital. Continuous EEG monitoring
initially showed intermittent bursts of generalized spike activity for 2-3 s
each minute. Over the next 24 h, the EEG showed no electrical activity at
maximum sensitivity and the pentobarbital therapy was discontinued. A
second CT scan two days after admission showed diffuse cortical swelling
and bilateral basal ganglia lucencies (Fig. lb).
Approximately 1 week later the patient’s blood barbiturate level was
zero, however, his EEG continued to show no activity. Neurological examination revealed weak respiratory efforts and a gag reflex. Nearly 1 month
after admission a third CT scan revealed evidence of diffuse cerebral infarction and compensatory hydrocephalus (Fig. lc). EEGs, recorded approximately weekly, were unchanged. No further intervention or resuscitative
efforts were desired by the family. His neurological examination remained
unchanged. The patient was slowly weaned from the ventilator, however, he
suffered from recurrent pneumonia caused by antibiotic resistant organisms.
He expired approximately five months after admission.
Postmortem examination revealed a grossly shrunken brain (weight 1060
g) with palpable subcortical cystic areas (Fig. 2a). Considerable cortical atrophy and necrosis of the cerebellar tonsils was evident. All major vessels
were intact and normal. Coronal sections revealed diffuse laminar necrosis
(Fig. 2b). Multiple cystic infarcts in the watershed distribution of the
anterior, middle and posterior cerebral arteries were present bilaterally
(Figs. 2c-2g).
The basal ganglia including claustrum, as well as the
thalamus and hippocampus revealed extensive cavitation with loss of
cellular elements (Figs. 2d-2e). Ventriculomegaly was present with the
temporal horns of the lateral ventricles being particularly dilated (Fig. 2f).
Multiple large infarcts were seen in the cerebelhu white matter with
degeneration of the deep nuclei (Fig. 2g). The cerebellar cortex was soft and
necrotic. Extensive necrosis was evident throughout the entire brain&em,

175

most notably in the substantia nigra, pontine tegmentum and medullary
pyramids. Large fluid filled cysts were seen in the tegmentum of the pans.
Tan discoloration and cavitation was present in areas containing cranial
nerve nuclei and fiber pathways in the medulla and spinal cord (Fig. 2h).
Microscopic examination revealed extensive laminar necrosis of the
cerebral cortex with extreme rarefaction of gray matter, with only blood
vessels and macrophages remaining (Fig. 3a). Rarefied cortex was bordered
by intense gemistocytic astrocytosis (Fig. 3b). Basal ganglia, midbrain, and
medulla showed a similsr pattern (Fig. 3~). Examination of the cerebellar
cortex showed an absence of Purkinje cells and depletion of the granular cell
layer (Fig. 3d). The spinal cord and medulla had extreme loss in both
ascending
and descending
pathways,
particularly
involving
the
spinocerebellar and corticospinal tracts (Fig. 3e).
DISCUSSION

Few reports of the neuropathological alterations following resuscitation
from global hypoxia&hernia with protracted survival are available [2,3,79]. Neuropathological studies of survivors from strangulation or hanging
either from attempted suicide or criminal assault are rare [l-3]. In general,
these reports describe patients surviving an acute global hypoxic or
ischemic episode from a variety of causes and expiring in less than one day
to a week. The constellation of neuropathological findings that occur in such
patients are those of ‘respirator brain classically described by Walker [lo].
These brains are soft or semiliquid, often stain poorly, possess
edematous white matter and swollen neurons and glial elements. The cortex
may show laminar necrosis of varying degree, particularly within the
hippocampus, and lytic lesions in the white matter. The thalamus is
generally spared, though perhaps swollen, but the basal ganglia often have
neuronsl loss. The cerebellum shows loss of Burkinje cells with granular
cells and molecular layers being relatively intact. The brainstem and spinal
cord are frequently normal but can show edema, hemorrhage or infarction
secondary to cerebellar tonsil herniation [lo].
Selective vulnerability of particular neuronal structures to ischemial
hypoxia was evident in the present case. The areas exhibiting the greatest
degree of tissue necrosis parallel areas possessing relatively high levels of
regional blood flow and relatively high rates of glucose metabolism, as
determined experimentally by Reivich et al. [ll] and Sokoloff et al. [12]. The
macmscopic changes in this patient were quite different from respirator
brain, and more closely resembled a combination of changes described in
vegetative or brain dead patients with prolonged survival [3,6,7,10].
Necropsy findings in 43 patients existing in a persistent vegetative state,
reported by Higashi et al. [6,13], revealed extensive gray and white matter
destruction but relative sparing of the brainstem. In a series of 25 brain
dead patients, Mohandas and Chou [7] described extensive pathology

176

177

178

Fig. 2. Post-mortem specimens obtained by coronal brain sectioning. Brain atrophy (narrow
gyri and widened s&i) and necrosis (loss of cerebellar tonsils) are seen on gross inspection (A).
Sections through the frontal cortex (B), the temporal lobe tips (C), the hippocampus (D), the
thalamus (E), the mesencephalon (F), the pons and cerebelhnn (G) and the medulla and spinal
cord (H) reveal extreme cellular loss and cavitation.

180

181

Fig. 3. Microscopic evaluation of the brain and spinal cord (H & E stain 9 e sections). Specimens include, (A) a cortical gyrus ( x 100) and (B) cortical white matter ( x 400) revealing laminar necrosis and gemistocytic astocytosis, respectively. Cerebellar folia (C) reveal Purkinje cell
loss (X 100). The brain stem (medullary olive) (D) and spinal cord (corticospinal tract) (E)
reveal gemistacytic astrocytosis and axonal loss, respectively ( x 400).

including severe brainstem autolysis. Dooling and Richardson [2], in
reporting a peculiar movement disorder in an 1l-year-old child surviving 13
weeks after strangulation, described extensive basal ganglia cavitation, with
sparing of the thalamus, brainstem
and cerebellum. Watershed
abnormalities and neuronal loss in the hippocampus was described but no
ventriculomegaly was seen. No abnormalities in glial elements nor myelin
loss was described. Parisi et al. [S], in a brief report, described reabsorbative
processes,
macrophage
infiltration,
active necrotic
decomposition,
recanilization of cerebral vessels and revascularixation of outer cortical
layers in a patient brain dead for 74 days. Because of softening of the
intracranial contents, the brainstem could not be adequately examined.
Ginsberg et al. [3], in three patients sustaining prolonged somatic survival
following global ischemia/hypoxia, reported that the external appearance of
the cerebrum and the cut surface of the thalamus and brainstem were
normal. The basal ganglia were necrotic in one case and laminar necrosis of
the cortex and hippocampus was present in another case.
A striking microscopic feature seen in the present case was the pronounced gemistocytic astrocytosis present throughout the cerebrum and

183

brain&em. Although gliosis is a regular feature of focal cerebral infarction
[14], glial reactivity is said not to occur as a chronic process in brains sub
jetted to global hypoxialischemia. In a classic review by Kjeldsberg [15], the
lack of a true glial reaction in severely hypoxiclischemic brains is
emphasized. Parisi et al. [8], noted only focal accumulations of lipid laden
macrophages, plasma cells and lymphocytes in subpial tissue. White matter
exhibited only a few polymorphonuclear leukocytes, rare macrophages and
karyorrhectic debris. Similar findings were observed by Ginsberg et al. [3]
where no gemistocytic astrocyte reaction could be ascertained in three
cases. Garcia [16], however, noted that reactive astrocytosis can occur. The
degree of astrocytosis, is suggested by a recent series [9], to be directly
dependent upon the interval between insult and brain death ranging from 1
h to 13 days in 190 autopsies.
CT examination of this patient’s brain showed a radiologic progression of
hypoxiclischemic changes. The initial study performed 4 h after the insult
revealed only a mild degree of cortical swelling whereas the study 48 h
later showed bilateral basal ganglia lucencies and considerable cortical swelling. The correlation of these early CT findings in such injuries to pathological specimens has only recently been verified by Tippin et al. [17]. Low
densities in the basal ganglia and watershed cortex and white matter as
well as diffuse swelling of the hemispheres with loss of gray-white matter
differentiation has been discussed [Ml. Based on these results, the early diagnosis, by radiographic methods, of diffuse hypoxia/ischemia was proposed. The third CT study of the present case illustrated severe, diffuse
encephalomalacia and ventriculomegaly which correlated well with the final
autopsy results. Although CT abnormalities have been demonstrated in patients in a vegetative state, radiographic studies are often normal even
3 weeks following an injury [18]. In fact, lack of late CT manifestations of
diffuse hypoxiclischemic damage, despite angiograpbic evidence of arrested
cerebral circulation, has been reported in six patients by Radberg and
Soderhmdh [19]. The results of the present case, however, indicate that
early CT abnormalities may not only be useful in the rapid diagnosis of
global ischemialhypoxia, but also for predicting final clinical outcome [17].
Although debated, several investigations indicate that EEG activity does
not always correlate with outcome, nor does it differentiate between brain
dead patients or those persisting in a vegetative state [20,21]. Several cases
of patients declared clinically brain dead have demonstrated low voltage
EEG activity [7]. Conversely, Higashi et al. [16] described isoelectric EEGs
in three patients persisting in a vegetative state. Gur patient demonstrated
isoelectric EEGs for nearly 5 months yet by current criteria was not brain
dead.
There have been several attempts to correlate brain death or vegetative
existence to both neuropathological changes and the neurological exam [4,7].
These studies have met with limited success because the evidence from such
reports describe a continuum of changes and not a rigid classification

184

system. This patient certainly met the neuropathological criteria for brain
death, as described by Black [22] and met the CT and EEG based criteria as
well. However, he did not meet clinical conditions for brain death because
he had persistent brainstem reflexes and respiratory efforts (231. He existed
in a persistent vegetative state for nearly 5 months with the resultant
emotional and economical consequences. Long-term mortality rates for
patients in this condition range 73-98%
in recent surveys [4,13]. As
illustrated in the present case, little functioning nervous tissue is required
to sustain this type of existence. Although considerable clinical variability
in neurological function is evident, the present case clearly demonstrates the
artificial nature of rigid boundaries between brain death and vegetative
patients. The ethical and moral issues regarding the care of such patients
are numerous and complex [24,25].
REFERENCES

7
8
9
10
11
12

13

14
15
16
17

T.P. McHugh and M. Stout, Near-hanging injury, Ann. Emerg. Med., 12 (1983) 774-776.
E.C. Dooling and E.P. Richardson, Delayed encephalopathy after stranghng, Arch. Neurol., 33 (1976) 196-199.
M.D.
Ginsberg,
E.T.
HedIyWhyte
and
E.P.
Richardson,
Hypoxic-ischemic
leukoencephalopathy in man, Arch. Neural., 33 (1976) 5-14.
D.E. Levy, R.P. KniRJonea and F. Plum, The vegetative state and its prognosis following
nontraumatic coma, Ann. N.Y. Acad. Sci., 315 (1978) 293-306.
P.M. Yagnik and R.M. Burns, Delayed anoxic encephalopathy presenting with psychiatric
symptoms, Paychoaomatics, 25 (1984) 413-415.
K. Higaahi, Y. Sakata, M. Hatano, S. Abiko, K. Ihara, S. Katayama, Y. Wakuta, T.
Okamura, H. Ueda, M. Zenke and H. Aoki, Epidemiological studies on patients with a
persistent vegetative state, J. Neural. Ne uroeurg. Paychiatr., 40 (1977) 876-885.
A. Mohandaa and S.N. Chou, Brain death. A cIinicaI and pathological study, J. Neuromug., 35 (1971) 211-218.
J.R. Parisi, R.C. Kim, G.H. Collins and M.F. HiIfinger, Brain death with prolonged 80
matic survival, New Engl. J. Med., 306 (1982j 14-16.
R. Schroder, Later changes in brain death. Signs of partial recirmdation, Acta Neuropathol., 62 (1983) 15-23.
A.E. Walker, Neuropathological findings in cerebral death. In: Cerebral Death, Editor:
A.E. Walker, 2nd Edn., Urban and Schwarxenberg, Baltimore, 1981, pp. 97-112.
M. Reivich, J. Jehle, L. Sokoloff and S.S. Kety, Measurement of regional cerebral blood
flow with antipyrine-14C in awake cats, J. Appl. PhysioL, 27 (1969) 296-300.
L. Sokoloff, M. Reivich, C. Kennedy, M.H. Des Rosiers, C.S. Patlak, K.D. Pettigrew, 0.
Sakurada and M. Shinohara, The [14Cjdeoxyglucoae method for the measurem ent of local
cerebral glucose utilization: Theory, procedure, and normal values in the conscious and
anesthetized albino rat, J. Nemo&em., 28 (1977) 897-916.
K. Higashi, M. Hatauo, S. Akibo, K. Ihara, S. Katayama, Y. Wakuta, T. Okamura and
T.Yamashita, Five-year follow-up study of patients with persistent vegetative state, J.
NeuroL N eurosurg. Psychiatr., 44 (1981) 552-554.
W.F.McCormick, Vascular diseases. In: The CIinical Neuroaciences, Editor: R.N. Rosenberg, Vol. 3, ChurchiIl Livingston, New York, 1983, pp. 35-83.
C.R. Kjeldaberg, Respirator brain. In: Pathology of the Nervous System, Editor: J.
Minckler, Vol. 3, McGraw-Hill Book Co., New York, 1968, pp. 2952-2961.
J.H. Garcia, Iechemic injuries of the brain, Arch. Pathol. Lab. Med., 107 (1983) 157-161.
J. Tippin, H.P. Adams and W.R.K. Smoker, Early computed tomographic abnormalities
following profound cerebral hypoxia, Arch. Neurol., 41 (1984) 1098-1100.

185
18
19
20
21
22
23
24
25

B.O. Kjoe, M. BrantrZawadzki end R.G. Young, Early CT findings of global central
nervous system hypoperfueion, Am. J. Neuroradiol., 141(1983) 1227-1232.
C. Radberg and S. Soderhmdh, Computed tomography in cerebral death, Acta Radiol., 346
(1975) 119-129.
B. Jennett,. Brain death, Intens. Care Med., 8 (1982) l-3.
J.E. Lee&ma, J.R. Hughes and E.R. Diamond, Temporal correlates in brain death. EEG
and clinicaI relationships to the respirator brain, Arch. Neurol., 41 (1984) 147-152.
P.M. Black, Brain death, New Engl. J. Med., 299 (1978) 338-344; 393-400.
R.J. Joynt, A new look at death, J. Am. Med. Assoc., 252 (1984) 680-682.
P.M. Black, Clinical problems in the uee of brain-death standards. Arch. Int. Med., 143
(1983) 121-123.
J.P. Freer, Chronic vegetative states: Intrinsic value of biological process, J. Med. PhiIos.,
9 (1984) 395-407.