Ischemic cerebellar infarct in a 5-year-old
boy: Sequela to minor back trauma

•

LORRAINE ROSE BYRD, DO
H. LAUREN VOGEL, DO

Minor trauma, a kick to the center of the back during roughhousing with an
8-year-old sibling, produced an ischemic cerebellar infarct in a 5-year-old boy. The boy was
brought to medical attention because of vomiting and headache. He also had ataxia. Results
of laboratory studies, structural examination,
and cerebrospinal fluid analysis were noncontributory. An irregular, low-density region
in the left cerebellar hemisphere of unknown
cause was seen on computed tomography
scans. Magnetic resonance imaging (MRI)
revealed an area of signal alteration hyperintense on the T2-weighted image and mildly
hyperintense on the proton-density image in
the leftward cerebellar hemisphere. The distribution was most compatible with that of
an acute superior cerebellar arterial infarction. No other abnormalities were detected.
The patient was treated with oral prednisolone,
recovered, and had no further episodes or
sequelae. The authors recommend that posttraumatic stroke be considered early in the
differential diagnosis of a child with the triad
of symptoms seen in the patient described.
They suggest that glucocorticoid therapy on
a short-term basis may have a place in more
rapid resolution of symptoms.
(Key words: Trauma, ischemic cerebellar
infarct, childhood cerebellar stroke)
Childhood cerebellar stroke may follow relatively minor nonhead trauma. Clinical presentation may include ataxia, vomiting, and headache
in a previously well child. We describe a 5-year-old
From Grandview Hospital, Dayton, Ohio, where at the time
this article was written, Dr Byrd was a resident in pediatrics,
and where, Dr Vogel was director of the pediatric residency
program and director of medical education. Dr Byrd is now in
private practice and Dr Vogel is Director of Medical Education,
Commuruty Health Center of Branin County, Coldwater, Mich.
Correspondence to Lorraine R. Byrd, DO, 201 S Wilder Ave,
Suite 700, Rockwood, TN 37854-2283.

Case report • Byrd and Vogel

boy with these symptoms who had an ischemic cerebellar infarct after minor trauma to his back.
Report of case
A 5-year-old boy with a wide-based gait was seen because
he had episodes of vomiting for approximately 12 hours.
He was lethargic and complained of headache. He made
intermittent inappropriate cognitive responses to questioning. His mother said that on the previous evening,
the child suddenly started vomiting, had a severe headache,
and had difficulty walking. Two hours before the onset
of symptoms, the patient and his 8-year-old sister were
scuffling, during which the patient was kicked in the
center of his back. He did not lose consciousness and
had no apparent head injury. The child had previously
been healthy, with no other history of trauma or recent
illness. He was taking no medications.
Initial vital signs were within normal limits. No evidence of trauma was found on physical examination.
Structural evaluation ofthe cervical, thoracic, and lumbar spine revealed no tenderness or motion restriction.
Tendon reflexes were symmetric and normal. Cranial
nerves II through XII were intact. Abnormal physical
findings included an impressive broad-based gait with deviation to the left. The child also had deviation to the left
when standing with feet together. Finger-to-nose maneuvers on the left side were abnormal.
Results of routine laboratory studies were all normal.
Computed tomography (CT) scans (Figures 1 and 2)
revealed an irregular, low-density region in the left cerebellar hemisphere of unknown cause. Differential possibilities included low-grade neoplasm, atypical cyst, or
early inflammatory process such as an abscess. An infectious etiology could not be ruled out, and a lumbar puncture was performed. Findings of cerebrospinal fluid
analysis were unremarkable. Magnetic resonance imaging (MRI) revealed an area of signal alteration hyperintense on the T2-weighted image (Figure 3, left) and
mildly hyperintense on the proton-density image in the
leftward cerebellar hemisphere (Figures 4, left and 5).
The distribution was most compatible with that of an
acute superior cerebellar arterial infarction. No other
abnormalities were detected.
In an attempt to prevent edema and its sequelae,
oral prenisolone therapy, 2 mg/kg per day in divided
doses, was started. The patient improved dramatically
and was discharged on the third day with ataxia virtu-

JAOA • Vol 96 • No 4 • April 1996 • 245

Figure 1. Initial noncontrast axial computed tomography scan
shows an oval hypodense area in the high left cerebellum effacing left portion of quadrigeminal plate cistern.

ally absent. The patient had no headache or emesis.
Prednisolone therapy was continued for an additional 5
days.
Follow-up examination 4 days after discharge from
the hospital revealed total resolution of abnormal gait,
headache, and emesis. Nine days after discharge, follow-up MRI studies showed resolving cerebellar infarct
(Figure 3, right ) and a resolving small left thalamic
infarct (Figure 4, right) that had not been noticed on the
original studies. Computed tomography 7 months later
showed evidence of the old infarcted regions without
new signs of infarction or intracerebral changes. Ten
months after the initial problem, the patient had no neurologic sequelae, and he has had no recurrence of ataxia or severe headache.
D iscu ssion
When ataxia is accompanied by headache and vomiting, suspicion of a posterior fossa etiology must be
raised, with cerebellar stroke or posterior fossa
tumor as distinct diagnostic possibilities. The presence of nystagmus, which was not seen in our
patient, is also consistent with a pathologic process
of the posterior fossa. Childhood stroke is infrequent and etiologies are diverse (Table). Traumainduced stroke in children is especially uncommon,
and cerebellar stroke is rare.
A review of the literature since 1972 found only
19 cases of stroke related to trauma in children,I-9
and only 7 of those involved the vertebrobasilar
system, which includes the cerebellar arteries .
Strokes are more commonly caused by occlusion in
the anterior carotid system. Sometimes, the cause
of the stroke becomes a diagnostic enigma, especially in previously well children. Baumann and
associates 10 reviewed data on 215 consecutively
imaged children referred to them because of neurologic disease. Of those 215 children, 28 showed

246· JAOA· Vol 96 • No 4· April 1996

an arterial infarction pattern. Of these , a third
had no associated illness or adverse event. Thirty-four percent of the cases of ischemic stroke cited
in the Mayo Clinic Rochester study on childhood
stroke l l had no identified etiology. Others4,6,12 likewise report childhood stroke of unknown cause in
their data. Mitral valve prolapse,13 arteriosclerosis,14
and migraines 15 are proposed by some authors as
the pathologic process involved in these strokes of
unknown origin. We propose that minor trauma
may be implicated as a causal factor as well and that
it may be more prevalent in its association with
childhood stroke than is currently reported in the
literature.
The temporal relationship between the event and
the onset of symptoms can be varied and prolonged.
Delays of 2 to 8 months have been reported.l·2 Others7 cite development of minor symptoms after 48
hours and more severe symptoms after 72 hours.
Symptoms are more likely to develop in 2 to 24
hours, as was observed in our patient. Because of
these delays, the traumatic event may not even be
remembered or may be dismissed as trivial and
not associated with symptoms. In the case reported here, a history of only a very minor trauma was
elicited. The patient and his 8-year-old sister were
playing, and she accidentally kicked him in the
back. Debehnke and Singer7 reported cerebellar
and pontine infarcts after a minor bicycle accident
without head trauma or loss of consciousness. Dooley and SmithS found a history of only having been
jarred as a passenger in an automobile collision 8
months before onset of symptoms in a previously
healthy 6-year-old stroke victim. Ko and Berbrayer3 describe cerebellar infarct in a 9-year-old boy 10
days after a day of tobogganing without specific
head or neck trauma. These cases suggest that the

Figure 2. Initial co ntrast axial computed tom ography scan
shows no enhancement of lesion.

Case report· Byrd and Vogel

Figure 3. Left: Initial axial T2·w eighted magnetic resonance
image (MRl) demonstrating focal increased s ignal involving
white and gray matter of left superior cerebellar hemisphere.
Right: Axial T2·weighted MRI 9 days later demonstrates less

hyperintense edema in the left superior cerebellum when compared with earlier study (left). Lack of progression and, in fact,
resolution of edema and size of lesion over time is consistent
with infarct.

Figure 4. Left: Initial coronal Tl- weighted magnetic resonance
image demonstrating a well· demarcated hypo intense lesion in
the left superior cerebellar hemisphere. Right: Coronal T2-weight·

ed MRI9 days later. Hyp erintense lesion involving white and
gray matter in the selective distribution of the left superior cere·
bellar artery.

traumatic etiology for childhood stroke may be
underestimated and that children who have even
minor head trauma may sustain cerebellar injury.
The mechanism of injury in our patient was
thought to be possible vertebral artery dissection
resulting from a whiplash kind of impact or trauma to the back. We also postulated that a sudden
force could have potentially slammed the superior
cerebellar artery up against the tentorium that
divides it from the posterior cerebral artery. Others 1 have suggested that stretching of the vertebral arteries with platelet aggregation and thrombosis secondary to traction, hyperextension, or

rotational injuries to the neck might be the mechanism of injury. Fraser and Zimbler2 discovered
anterior subluxation of vertebra C-l on vertebra
C-2, a common congenital anomaly, in their patient.
They proposed that this anomaly caused mechanical deformation and occlusion of the vertebral
arteries leading to stroke in their 6-year-old patient.
Diagnostic evaluation of children with ataxia,
vomiting, and headache begins, as always, with
taking a thorough history and completing a physical examination. Subtle historical clues must be
sought, such as history of antecedent viral illness
or minor trauma. Etiologies that may require imme-

("1 0 0 0

rQ o ort •

Bvrn fi nd VOQ:el

JAOA • Vol 96 • No 4 · April 1996 · 247

Table
Major Etiologies of Childhood Stroke 14
• Arterial thrombosis and embolism

o Arteriosclerosis
o Vasculitis- systemic lupus erythematosus
o Fibromuscular hyperplasia
o Sickle cell diseases
o Cardiac disease

Figure 5. Initial sagittal Tl-weighted m agnetic resonance image
shows focal wedge-shaped hypointense lesion in the left superior
cerebellar hemisphere.

diate medical or surgical intervention should be
considered first, including infection, stroke, hydrocephalus, posterior fossa neoplasm, and cervical
vertebral fracture or instability
Findings consistent with acute cerebellar disease should have early evaluation by CT or MRI.
Gentry and colleagues 16 have shown MRI to be
highly sensitive in detecting both hemorrhagic and
nonhemorrhagic lesions. They compared MRI and
CT examination to determine their relative sensitivities. They report that CT has not been particularly useful for classifying and staging injury and
that it has low sensitivity for nonhemorrhagic
lesions. They state that MRI offered distinct advantages over other imaging methods for classifying
traumatic cerebral lesions. The ability of MRI to
image in multiple planes was cited as particularly beneficial for localization of the lesion. Computed tomography has a diagnostic accuracy of
approximately 60% in the first week after the
episode, increasing to 100% in the second weekP
Computed tomography findings in our patient
were inconclusive, and MRI was needed to make
the final diagnosis. In the case of the bicycle accident reported by Debehnke and Singer,? the CT
scan showed no abnormality while MRI revealed
cerebellar and pontine infarcts. The findings cited
by Gentry and associated 16 and the implications
drawn from these two cases suggest that MRI be
the initial diagnostic test when an ischemic event
is suspected and that CT is more appropriate for
follow-up evaluation 2 or more weeks later. In our
institution, CT is more readily available and therefore was performed first.
Other diagnostic tests may need to be done, as

248 • JAOA • Vol 96 • No 4 • April 1996

_ Cyanotic congenital heart disease
_ Mitral valve prolapse
_ Myocardial infarction
o Infectious disorders
_ Meningitis
_ Bacterial infections
_ Mycoplasma spp infections
o Neurofibromatosis
o Moyamoya syndrome
o Homocystinuria
o Vertebrobasilar occlusion
o Trauma
• Venous thrombosis

o Want dehydration
o Polycythemia associated with cyanotic
heart disease
o Pregnancy
o Oral contraceptives
o Nephrotic syndrome
o Purulent venous thrombosis

_
_
_
_

Mastoiditis
Otitis media
Scalp infections
Facial infections
o Sagittal sinus thrombosis with acute
lymphoblastic leukemia
•

Intracranial hemorrhage

o Arteriovenous malformation
o Galen's vein malformation
o Aneurysm
o Intracerebral hemorrhage

_ Hypertension
_ Trauma

• Vascular disease of the spinal cord

o Thrombotic infarction of the spinal cord from
sickle cell disease

dictated by historical and clinical findings .
The major goal of therapy in stroke includes
treatment to preserve brain function by maintaining circulation and oxygenation and prevention of subsequent neurologic deficits through the
use of positioning, hyperventilation, and hyperosmolar agents. Steroid therapy is effective in treating cerebral edema that occurs after a stroke, and
aspirin therapy has been shown to prevent recurrent ischemic events.1 8 The value of steroids in
preventing neurologic deficit has not been proved,
and their use in the pediatric stroke patient has
not been established. However, we elected to admin-

CaseJ"enort. • Rvrd a nd VOl7l>cl

ister oral prednisolone to our patient in an attempt
to prevent further sequelae from edema. Clinical
improvement was rapid and dramatic, and our
patient has had no neurologic sequelae or recurrent
attacks.
KO,3 Rothrock and coworkers,15 and Jain and colleagues 19 report pronounced improvement after
the administration of steroids to their pediatric
stroke victims as well. We propose that glucocorticosteroid administration on a short-term basis
may bring about rapid resolution of symptoms in
childhood stroke victims and that its efficacy in
this setting warrants further study. Aspirin therapy was not used in our patient. Patients with a more
advanced pathologic process may require surgical intervention.
Children seem to recover from strokes differently
than adults. It is thought that this may be due to
the development of axonal sprouting and the formation of new synapses or functional reorganization of normal structure in the young brain.17 Children can, however, still have devastating sequelae.
Comment
A physician must consider a vertebrobasilar or posterior fossa pathologic process when evaluating a child
who is seen with ataxia, vomiting, and headache. Thorough history taking and physical examination are
essential for discovering sometimes subtle clues as
to the etiology of the symptoms and for dictating
diagnostic studies that should be done. It is prudent to include CT or MRI in the initial diagnostic
workup in order to expeditiously diagnose potentially lethal conditions that may require immediate medical or surgical interventions. It is suggested
that MRI is more sensitive for evaluating patients
with suspected ischemia and, where possible, should
be a diagnostic consideration even if there is history of only minor nonhead trauma. Rapid diagnosis and treatment may prevent further damage and
subsequent morbidity. Glucocorticosteroid therapy
may have a place in more rapid resolution of symptoms and its use needs further study. Children seem
to recover better from strokes than do adults.
The Strokes in Children Study Group, a section of the Joint Committee for Stroke Facilities,
suggested in 1973 that strokes in children should be
reported to a registry to permit better understanding
of the natural history of these disorders. To our
knowledge, this practice was never established. It
would seem advisable that this project again be
suggested. Perhaps with widespread reporting, we
would have better insight into the prevalence and

etiology of stroke in children and spark interest in
exploring treatment possibilities, such as glucocorticosteroid administration early in the course of
the episode. Any progress in improving morbidity
for these children would make this an important
project.
References
1. Zimmerman AW, Kumar AJ, Gadoth N, et al: Traumatic vertebrobasilar occlusive disease in childhood. Neurology 1978;28:185188.
2. Fraser RAR, Zimbler SM: Hindbrain stroke in children caused
by extracranial vertebral artery trauma. Stroke 1975;6:153-159.
3. Ko GD, Berbrayer D: Childhood stroke after minor neck trauma: Case report. Arch Phys Med Rehabil 1990;71:923-926.
4. Wanifuchi H, Kagawa M, Tadeshita M, et al: Ischemic stroke
in infancy, childhood, and adolescence. Childs Neru Syst 1988;4:219230.
5. Thompson JR, Simmons CR, Hasso AN, et al: Inclusion of
the intradural vertebrobasilar artery. Neuroradiology 1978;14:219230.
6. Hilton..Jones D, Warlow CP: The causes of stroke in the young.
N eurology 1985;232:137-143.
7. Debehnke DJ, Singer JI: Vertebrobasilar occlusion following
minor trauma in an 8 year old boy. Am J E merg Med 1991;9:495 l.
8. Dooley JM, Smith KR: Occlusion of the basilar ar tery in a 6
year old boy. N eurology 1968; 18: 1034-1036.
9. Ouvrier RA, Hopkins IJ: Occlusive disease of the vertebral
basila r a rterial system in childhood . Deu Med Child N eurol
1970;12: 186-192.
10. Baumann RJ, Carr WA, Shuman RM: Pa tterns of cerebral
arterial injury in children with neurologic disabilities. J Child
Ne urol 1987;2:298-306.
11. Schoenberg BS, Mellinger JF, Schoenberg DG: Cerebrovascular disease in infan ts and children : A study of incidence, clinical features, and survival. Neurology 1978;28: 763-768.
12. Eeg-Olofss on 0 , Ringheim Y: Stroke in children : Clinical
characteristics and prognosis. Acta Paediatr Scand 1983;72 :391395.
13. Rice GP, Ebers GC, Bondar RL, et al: Mitral valve prolapse:
A cause of stroke in children? Deu Med Child Neurol 1981;23:352368.
14. Daniels SR, Bates S, Lukin RR, et al: Cerebrovascular arterioplasty (arteriosclerosis) and ischemic childhood stroke. Stroke
1982;13:360-365.
15. Rothrock JF, Walicke P , Swenson MR, et al : Migrainous
stroke. A rch Neurol 1988;45:63-67.
16. Gentry LR, Godersky J C, Thompson B: MR imaging of head
trauma: Review of the distribution and radiopathologic features
of traumatic lesions. AJR 1988; 150:663-672.
17. Golden GS : Stroke syndromes in childhood . N eurol Clin
1985;3:59-75.
18. Adam s RD , Victor M: Principles of N eurology, ed 5. New
York, NY, McGraw-Hill, Inc, 1993, pp 705-709.
19. J ain S, Maheshwari MC, Tandon PN, et al: Idiopa thic basilary artery occlusion in childhood. Case report. Stroke 1984;15:563565.

JAOA • Vol 96 • No 4 • April 1996 • 249