0148-396X/79/0401-0007$02.00/0
NEUROSURGERY
Copyright © 1979 by the Congress of Neurological Surgeons

Vol. 4, No. 1, 1979
Printed in U.S.A.

Clinical and scientific communications

Cerebellar Infarction

Michael P. Feely, M.Ch.

Department of Neurosurgery, College of Medicine, University of South Alabama, Mobile, Alabama

Cerebellar infarction has been inadequately recognized by clinicians. A review of 75 cases showed that in 55 of them the
infarct acted as an expanding mass lesion and compressed the brain stem. Once this occurred, the mortality without
operation was very high. With surgical treatment, the mortality was reduced considerably. The computerized tomographic

scan is the diagnostic test of choice.

Key words: Brain stem compression, Cerebellar infarction, Computerized tomographic scan, Emergency surgery, Posterior

fossa mass

Infarction in a cerebellar hemisphere is considered to be a
rare occurrence. Three cases have been treated recently by the
staff of the Department of Neurosurgery at this university.
These, together with 20 case reports from the literature (3, 4, 9,
10, 12, 15) and a large series (13) in which the diagnosis was
made at postmortem examination, provide 75 cases for analysis.

Two distinct clinical syndromes have been described in the
literature. The first is cerebellar infarction with brain stem
compression (9). In this syndrome infarcted cerebellar tissue
acts as a rapidly expanding posterior fossa mass lesion, signs of
brain stem compression develop, and the outcome almost
invariably is fatal unless surgical intervention alters the course.
The second type of cerebellar infarction is described as being
in the territory of the posterior inferior cerebellar artery (PICA)
(3) and it is apparently more benign, resembling acute labyrin-
thitis. The symptoms may resolve spontaneously.

This distinction may be both artificial and dangerous. Pa-
tients presenting to the internist, family physician, or otolar-
yngologist with symptoms and signs of benign labyrinthitis
may progress in hours or days, developing signs of brain stem
compression. When this happens, unless the diagnosis is made
correctly and prompt action is taken, the mortality is exceed-
ingly high. Although the diagnosis is possible on clinical
grounds, it may be very difficult and apparently has been
overlooked frequently. The computerized tomographic (CT)
scan has proved invaluable in making the diagnosis.

My analysis and discussion center on 55 patients who devel-
oped signs of brain stem compression. A brief mention is
made of 20 additional cases of cerebellar infarction, including
Case 3, in which signs of brain stem compression did not occur.

CASE REPORTS
Case 1

An 18-year-old woman was admitted to the hospital with
drowsiness and slurred speech. She had been in good health
and had been taking a low dose contraceptive pill for 15
months before admission. No focal neurological findings were
detected. A lumbar puncture was performed and showed clear
fluid (cells, 0; protein, 40 mg/100 ml) with an opening pressure
of 160 mm of cerebrospinal fluid (CSF).

The next day, the patient remained drowsy but would answer

simple questions. Except for her eyes, which were deviated to
the right, the patient had full voluntary movement to com-
mand. The eyes did not move on cold caloric irrigation in the
ears. There was unsustained clonus at the right ankle.

Over the next 3 hours, the patient’s condition deteriorated.
She ceased to follow commands, and decerebrate posturing on
stimulation appeared on the right side. Four-vessel cerebral
angiograms showed no abnormality. The next morning decer-
ebrate posturing was present bilaterally on stimulation. Her
eyes were in midposition, and the pupils were small and
minimally reactive to light. A CT scan showed a large infarct
in the left cerebellar hemisphere (Fig. 1).

An emergency suboccipital craniectomy was performed. The
left cerebellar hemisphere was swollen and friable. The cere-
bellar tonsil was herniated to the level of the arch of C-1. A
subtotal resection of the cerebellar hemisphere was performed
on the left side and included the tonsil. Only a thin rim of
tissue around the outer margin of the hemisphere was left.
Histological examination confirmed the diagnosis of cerebellar
infarction.

Postoperatively, the patient made an uneventful recovery,
and she has resumed her previous occupation as a housewife
without any disability.

Case 2

This previously healthy 63-year-old woman developed re-
current episodes of dizziness for 1 week before admission to
the hospital. She was admitted complaining of severe head-
aches and dizziness of 24 hours duration. On physical exami-
nation, the patient was drowsy but responsive to command.
Her pupils were small and reacted feebly to light. Her eyes
were in fixed conjugate deviation to the left. There was a
moderate symmetrical hyperreflexia.

Lumbar puncture revealed clear spinal fluid (protein, 36
mg/100 ml; red blood cells, 0/cu mm; monocytes, 6/cu mm)
with an opening pressure of 410 mm of CSF.

The next day a ventriculogram was performed. This dem-
onstrated that the 4th ventricle was displaced forward and to
the left, suggesting a mass in the right cerebellar hemisphere.

An emergency suboccipital craniectomy was performed. The
right cerebellar hemisphere was enlarged, soft, and partially
liquified. The cerebellar tonsil had herniated down to the level

8

Fic. 1. Case 1. CT scan showing a massive left cerebellar infarction.
Comparison between the right and left cerebellar hemispheres shows
a large, low density area in the left cerebellar hemisphere.

of the arch of C-2. The infarcted tissue was resected. Histolog-
ical examination confirmed the diagnosis of cerebellar infarc-
tion.

The patient made an uneventful recovery.

Case 3

This 66-year-old woman had previously been in good health.
She stood up to cross a room and immediately experienced a
severe attack of vertigo and fell to the floor. She was admitted
to the hospital, where she complained of constant vertigo and
inability to stand.

Physical examination revealed acute distress with frequent
vomiting. There was rotary nystagmus on right lateral gaze.
The patient was unable to stand unaided. There was no dys-
metria of the arms, but heel to shin testing revealed bilateral
dysmetria of the legs.

A diagnosis of acute labyrinthitis was made. The patient was
treated with bedrest, and her symptoms cleared sufficiently so
that she could be discharged from the hospital 6 days after
admission. She was seen again in the outpatient clinic 3 weeks
later, when examination demonstrated a marked cerebellar
ataxia and inability to walk without support. The initial diag-
nosis of acute labyrinthitis was thought to be incorrect, and the
patient was readmitted to the hospital; a CT scan demonstrated
an infarct in the right cerebellar hemisphere. The patient made
a full recovery during the next 3 months.

ETIOLOGY

Based on the pathological investigations by Sypert and
Alvord (13), the majority of cases of cerebellar infarction seem
to be secondary to arteriosclerotic arterial occlusion. In their
study of 28 patients dying acutely after cerebellar infarction,
the vertebral or posterior inferior cerebellar artery was occluded

Ln iii)

Neurosurgery, Vol. 4, No. 1

in approximately 90% of cases. A superior cerebellar artery
occlusion occurred in approximately 10% of cases. Twenty-two
per cent (6 of 28) of these occlusions were embolic, as were 2
of 3 cases in another series (3). As would be expected in this
series, because the majority of patients were in the older age
group, a number of them had other manifestations of occlusive
vascular disease, particularly cerebral and cardiac. Other fac-
tors were: pre-existing hypertension, 16 cases; diabetes mellitus,
5 cases; polycythemia, 2 cases; and the contraceptive pill (2,
14), | case.

CLINICAL FINDINGS

The age and sex distributions of the clinical series (21 cases
of cerebellar infarction with brain stem compression) are de-
picted in Figure 2. All patients with cerebellar infarction
presented initially with the rapid onset of vertigo and ataxia.
Many also had headache and vomiting. One of three courses
then followed.

1. Rapid onset of brain stem compression occurred, mani-
fested by alteration of consciousness, long tract motor signs,
and oculomotor or gaze palsies that began within 24 hours of
the onset of the symptoms (Case 1).

2. The onset of brain stem compression was more delayed,
developing between 24 and 96 hours after the onset of symp-
toms. The manifestations were essentially the same as those of
the first variety; once brain stem compression occurred, dete-
rioration could be rapid (Case 2).

3. Cerebellar infarction mimicked a benign labyrinthine
disorder, with the sudden onset of vertigo, ataxia, vomiting,
and nystagmus. Brain stem compression did not develop, and
recovery was the rule (Case 3).

The “motor” abnormalities that developed in the 21 clinical
cases with signs of brain stem compression are outlined in
Table 1.

Detailed lists of different physical findings have been com-
piled (9, 13). However, the physician must consider cerebellar
infarction in the differential diagnosis of any patient presenting
with ataxia and dizziness and a course suggestive of a rapidly
expanding posterior fossa mass lesion; the physician will find
this to be 4 more useful and practical approach than trying to
remember a detailed analysis of occasional findings.

No.
Patients

8

7

Age (Years) 10 20 30 40 50 60 70 80

Fic. 2. Cerebellar infarction with brain stem compression. Age and
sex distributions of 21 clinical cases.

January 1979

TaBLe |
“Motor” Abnormalities among 21 Clinical Cases of Cerebellar
Infarction with Brain Stem Compression

“Motor” Abnormality No. Cases
Gaze palsy or 6th nerve palsy 14
Extensor plantar response, bilateral 1
Cerebellar ataxia 6
Decerebrate posturing 5
Facial palsy 2
1

Extensor plantar response, unilateral

INVESTIGATIONS

A CT scan should be performed immediately when cerebel-
lar infarction is suspected. It not only localizes the lesion and
suggests the pathological process, but it also shows any asso-
ciated hydrocephalus. Lumbar puncture showed an elevated
CSF pressure in approximately half of the cases reviewed. The
fluid was usually clear, although there were up to 1000 red
blood cells/cu mm on microscopic examination. Lumbar punc-
ture adds little to the diagnosis and it may be dangerous in the
presence of an expanding posterior fossa mass lesion. This is
supported by a report that mentions one patient, fully alert at
the time of the lumbar puncture, who died within an hour
afterwards (13). Posterior fossa angiography, pneumoenceph-
alography, and ventriculography have all proved to be helpful
in diagnosis, when used.

NEUROPATHOLOGY

The diagnosis of infarction was proven either by surgical
biopsy or by autopsy in every case in this series except Case 3.
in which it was made by CT scan. I could not determine the
exact size and extent of the infarction from the surgical reports.
Apparently the necrotic areas were large, and descriptions such
as “most of the cerebellar hemisphere” and “the entire infero-
lateral aspect” were commonly used to indicate the extent of
infarction.

Autopsy reports on those patients dying acutely of brain
stem compression indicate that the infarction was at least one-
third the volume of the cerebellar hemisphere in all cases and
frequently was from one-half to two-thirds the volume of the
hemisphere (3, 9, 13). Herniation of the cerebellar tonsil below
the foramen magnum was described, either at operation or at
autopsy, in 41 of the 55 cases with signs of brain stem compres-
sion. A few of these cases had bilateral cerebellar infarction.
No information is available on cases in the remainder of this
group.

Pathological information concerning the patients who did
not develop signs of brain stem compression is less detailed
(13). Apparently the volume of the infarct was considerably
smaller in those cases. Interestingly, “moderate hydrocephalus”
was found in more than half (13 of 24) of the patients in the
autopsy series who had prolonged survival after cerebellar
infarction (13).

TREATMENT AND OUTCOME

A depressed level of consciousness varying from stupor to
coma developed in 55 patients in this series. These patients also
had long tract motor signs and cranial nerve palsies. The
treatment and outcome in these 55 cases are reported in
Table 2. The operation performed in all cases was suboccipital
craniectomy and resection of the necrotic cerebellar tissue. A
sufficient volume of tissue was resected to provide an adequate
decompression of the posterior fossa.

CEREBELLAR INFARCTION 9

TABLE 2
Treatment and Outcome in 55 Cases of Cerebellar Infarction with
Brain Stem Compression

Outcome
Treatment No. Cases Full recov- _ Partial re-
Death
ery covery
No operation 37 6 oi 3
Operation 18 8 2 3

In the group of patients who developed signs of brain stem
compression, 5 of the 18 treated surgically and 31 of the 37 not
treated surgically died. The neurological deficits in the 5 pa-
tients who made an incomplete recovery after operation were
usually of mild or moderate severity. However, | patient was
described as being “disoriented” | year after operation. Of the
20 patients not developing signs of brain stem compression, 19
made a satisfactory recovery without operative treatment. One
patient died 7 days after the onset of symptoms of an acute
myocardial infarction (3).

DISCUSSION

Infarction of the cerebellum is an entity that has received
little attention in the literature and is overlooked in some
standard works (1, 17). The true incidence is difficult to ascer-
tain. In one series, 4 cases were seen in a 1000-bed general
hospital over a 10-year period (9). However, Sypert and Alvord
indicate that 52 cases were found at autopsy during the same
period in three general hospitals associated with a university
medical center (13). In another investigation specifically study-
ing brain stem vascular phenomena no case of cerebellar
infarction was found among 1687 autopsies (5). It is likely that
the incidence of clinical cases will increase as CT scanning is
used more frequently in the assessment of patients with stupor
and coma.

Cerebellar infarction with brain stem compression can be
differentiated from acute hypertensive cerebellar hemorrhage,
which may have similar symptoms and signs. Pre-existing
hypertension makes the latter more probable, and patients with
hemorrhage show a more rapid clinical deterioration in many
instances (6, 11). Patients with a subdural or epidural hema-
toma in the posterior fossa are likely to have a history of
trauma (16). A rapidly expanding tumor or an abscess in the
cerebellar hemisphere may also present in a manner similar to
that of infarction. Brain stem infarction can be differentiated
from cerebellar infarction by the greater variety of cranial
nerve palsies (Table 1), particularly those other than ocular
and facial palsies, or by the immediate appearance of the
complete deficit (7). The CT scan is quite helpful in differen-
tiating these conditions.

However, if the clinical situation is sufficiently critical and
the diagnosis of cerebellar infarction is suspected, one should
not delay for radiological confirmation but should proceed
with emergency posterior fossa decompression. This should be
done even when the patient is in extremis because dramatic
improvement has occurred in several cases, including Case 1.
The situation here seems remarkably similar to that described
by Fisher et al. in relation to cerebellar hemorrhage, where a
“critical state (decerebrate posture, cessation of respiration,
fixed pupils) which in other conditions is nearly always fatal,
can be relieved” by surgical decompression (6). “It is our
observation that the hopeless state of cerebellar hemorrhage
may be more apparent than real” (6). This seems to apply
equally to cerebellar infarction.

The etiology of cerebellar infarction does not seem remark-

10 FEELY

ably different from that of infarction in other regions of the
brain. The age distribution is also similar to that for other cases
of cerebral infarction.

The evidence points to the fact that a cerebellar infarct may
act as a progressively enlarging mass lesion in the posterior
fossa; the clinical course is indicative of progressive brain stem
compression. Twenty patients showed radiological evidence of
either hydrocephalus, displacement of the 4th ventricle, or
aqueductal occlusion. These findings suggest a posterior fossa
mass lesion. Herniation of a cerebellar tonsil below the level of
the foramen magnum was found either at operation or at
autopsy in 41 cases.

Fluid accumulation begins soon after ischemic injury to the
brain. As necrosis of cerebral tissue proceeds, water uptake and
recognizable brain edema increase greatly (8). Progressive ex-
pansion of the infarct in the narrow confines of the posterior
fossa compresses the brain stem; for this reason the onset of
symptoms and signs of brain stem compression may be delayed
for several days after the onset of the infarction.

I find no report of attempts to treat the cerebral edema in
these cases by nonsurgical means such as steroids, diuretic
agents, and hyperventilation. Provided that the correct diag-
nosis has been made and that other expanding mass lesions in
the posterior fossa have not been overlooked, this form of
treatment may be appropriate, particularly in those cases in
which the signs do not suggest that brain stem compression is
severe. As indicated in Table 2, surgical decompression is
mandatory when deterioration is occurring or if brain stem
compression is already advanced. With prompt diagnosis and
early treatment, we should be able to reduce the mortality
considerably from the present levels.

ACKNOWLEDGMENT
I thank Dr. R. H. Mudd for permission to present Case 2.

Received for publication, April 15, 1978; accepted, July 22, 1978.

Presented at the Third Joint Meeting on Stroke and Cerebral
Circulation, New Orleans, Louisiana, February 17, 1978 (Feely, M. P.
and Dempsey, P. J. Acute cerebellar infarction—A surgical emergency.
Stroke, 9: 96, 1978 (abstr.).)

Reprint requests: Michael P. Feely, M.Ch., Department of Neuro-
surgery, College of Medicine, University of South Alabama, 2451
Fillingim Street, Mastin Building 712, Mobile, Alabama 36617.

REFERENCES

1. Baker, A. B. and Baker, L. H. Clinical Neurology. Hagerstown,
Maryland, Harper and Row, 1971. Updated with revisions: 1976.

2. Collaborative Group for the Study of Stroke in Young Women.
Oral contraception and increased risk of cerebral ischemia or
thrombosis. N. Engl. J. Med., 288: 871-878, 1973.

3. Duncan, G. W., Parker, S. W., and Fisher, C. M. Acute cerebellar
infarction in the PICA territory. Arch. Neurol., 32: 364-368, 1975.

4. Fairburn, D. and Oliver, L. C. Cerebellar softening: A surgical
emergency. Br. Med. J., 1: 1335-1336, 1956.

5. Fang, H. C. H. and Palmer, J. J. Vascular phenomena involving
brainstem structures: A clinical and pathologic correlation study.
Neurology (Minneap.), 6: 402-419, 1956.

6. Fisher, C. M., Picard, E. H., Polak, A., Dalal, P., and Ojemann, R.
G. Acute hypertensive cerebellar hemorrhage: Diagnosis and sur-
gical treatment. J. Nerv. Ment. Dis., 140: 38-57, 1965.

7. Gilroy, J. and Meyer, J. S. Medical Neurology. 2nd Ed. New York,
MacMillan and Co., 1975, pp. 509-589.

8. Klatzo, I. Pathophysiologic aspects of cerebral ischemia. In The
Basic Neurosciences, edited by Brady, R. O., pp. 313-322, New
York, Raven Press, 1975. (The Nervous System, Vol. 1)

9. Lehrich, J. R., Winkler, G. F., and Ojemann, R. G. Cerebellar
infarction with brain stem compression: Diagnosis and surgical
treatment. Arch. Neurol., 22: 490-498, 1970.

Neurosurgery, Vol. 4, No. 1

10. Lindgren, S. O. Infarctions simulating brain tumors in the posterior
fossa. J. Neurosurg., 13: 575-581, 1956.

11. McKissock, W., Richardson, A., and Walsh, L. Spontaneous cer-
ebellar haemorrhage: A study of 34 consecutive cases treated
surgically. Brain, 83: 1-9, 1960.

12. Plum, F. and Posner, J. B. Diagnosis of Stupor and Coma. 2nd Ed.
Philadelphia, F. A. Davis Co., 1972, p. 137.

13. Sypert, G. W. and Alvord, E. C. Cerebellar infarction: A clinico-
pathological study. Arch. Neurol., 32: 357-363, 1975.

14. Vessey, M. P. Oral contraceptives and thromboembolic disease.
Am. Heart J., 77: 153-157, 1969.

15. Wood, M. W. and Murphey, F. Obstructive hydrocephalus due to
infarction of a cerebellar hemisphere. J. Neurosurg., 30: 260-263,
1969.

16. Wright, R. L. Traumatic hematomas of the posterior cranial fossa
J. Neurosurg., 25: 402-409, 1966.

17. Youmans, J. R. Neurological Surgery: A Comprehensive Reference
Guide to the Diagnosis and Management of Neurosurgical Problems.
Philadelphia, W. B. Saunders, 1973.

COMMENTS

The author summarizes the relatively small number of re-
ported cases of cerebellar infarction, adds three new cases, and
again calls attention to the importance of early recognition and
surgical treatment of this problem when brain stem compres-
sion occurs. He also demonstrates the CT scan findings of
cerebellar infarction.

The real incidence of cerebellar infarction will probably only
be obtained at a center in which a large number of CT scans
are performed for cerebrovascular disease. It is not known how
many patients have a small area of infarction in the territory
of the PICA that resolves spontaneously.

Table 2 records 37 cases with brain stem compression that
were not operated upon. Because this includes a number of
cases from the autopsy series, the true incidence of full and
partial recovery in this group is not known.

It is possible that the CT scan could be negative early in the
course of the illness. I would completely agree with the author's
statement, “ ... if the clinical situation is sufficiently critical
and the diagnosis of cerebellar infarction is suspected, one
should not delay for radiological confirmation but should
proceed with emergency posterior fossa decompression.”

Robert G. Ojemann, M.D.
Boston, Massachusetts

In spite of rapid advances in the diagnosis and treatment of
neurological diseases, acute vascular mass lesions of the cere-
bellum often remain undiagnosed during life and are not
recognized as potentially treatable neurological emergencies in
patients with stroke syndromes. Acute massive cerebellar in-
farction has been thought to be rare compared with infarction
elsewhere in the brain. In addition, early attempts at recogniz-
ing a reliable clinical syndrome have been unsuccessful, as the
signs and symptoms are considered too protean. However, the
author clearly documents that this pathological state is not rare
and that early diagnosis and prompt surgical attention are
mandatory before the development of irreversible brain stem
damage. Furthermore, the author is first to demonstrate the
importance of computed tomography in the early diagnosis of
this potentially rapidly expanding posterior fossa mass lesion.
Our experience with early computed tomographic evaluation
of these cases is consistent with the author’s conclusion that
the CT scan is the diagnostic test of choice. A high index of
suspicion, an early diagnosis, and prompt surgical decompres-
sion will certainly result in a reduced mortality; many patients
will recover with little or no disability.

Dr. Feely does suggest that attempts to treat the cerebellar
edema in these cases by nonsurgical means such as steroids,

January 1979

diuretic agents, and hyperventilation may have a place in the
treatment of acute cerebellar infarction associated with massive
cerebellar swelling. If instituted early in the course, it is possible
that this form of management will be beneficial. However, if
brain stem compression occurs, it is exceedingly unlikely, with
current techniques, that a significant decompression of the
brain stem can be obtained by these nonsurgical methods.
There is no evidence to suggest that glucocorticoids will have
any significant effect on the swelling associated with cerebellar
infarction. Furthermore, the small confines of the posterior
fossa will make diuretic agents and hyperventilation as modes

CEREBELLAR INFARCTION 11

of therapy not particularly effective because they seem to work
primarily on normal cerebral structures. In the case of acute
massive cerebellar infarction, a majority of the posterior fossa
cerebral tissue has been infarcted and is, thus, unlikely to
respond to these modes of management. Therefore, it is most
probable that, in patients with evidence of brain stem compres-
sion, prompt surgical decompression will be required and that
any delay for other modes of management will probably cul-
minate in an unsatisfactory result for the patient.

George W. Sypert, M.D.
Gainesville, Florida