Archives of Disease in Childhood, 1987, 62, 1025-1029

Specific enlargement of the fourth ventricle after
ventriculo-peritoneal shunt for post-haemorrhagic
hydrocephalus
A E O'HARE, J K BROWN, AND R A MINNS
Department of Paediatric Neurology, Royal Hospital for Sick Children, Edinburgh

SUMMARY Disproportionate enlargement or isolation of the fourth ventricle are rare complications of ventriculo-peritoneal shunt for post-haemorrhagic hydrocephalus. Obvious features
such as ataxia may indicate raised intracranial pressure in the posterior fossa. The child may,
however, present with less specific signs of vomiting, anorexia, and lethargy and these symptoms
may be misinterpreted as secondary to dilation of the lateral ventricular system due to
malfunction of the ventriculo-peritoneal shunt. Two children with disproportionate enlargement
of the fourth ventricle and two children with isolation of the fourth ventricle are described to
illustrate the wide variations in clinical presentation. These rare complications can be diagnosed
by real time ultrasound examination of the brain or computed tomography of the head. They
should be considered in the differential diagnosis of a child who deteriorates after lateral
ventriculo-peritoneal shunting for post-haemorrhagic hydrocephalus, and dealt with specifically
to avoid the risk of upward herniation of the enlarged fourth ventricle.

Intracranial haemorrhage is a common problem in
the premature neonate and occurs in up to 70% of
infants who are born at less than 35 weeks' gestation,
weigh less than 1500 g, and require assisted ventilation.' Post-haemorrhagic hydrocephalus may
result from inflammatory ependymitis; basilar
arachnoiditis; obstruction of the foramen of Munro,
aqueduct of Sylvius, or fourth ventricular outflow
foramina;2 or occlusion and scarring of surface
absorptive mechanisms.3 Many infants require
ventricular cerebrospinal fluid shunting to control
the dilatation of the ventricles4 and there is substantial morbidity associated with progressive posthaemorrhagic hydrocephalus.5 Intraventricular
haemorrhage may also occur in the infant born at
term; 35% subsequently require ventricular shunt-

Case reports
CASE 1

This infant was delivered at 30 weeks by caesarean
section because of severe maternal pre-eclampsia.
He did not breathe spontaneously at birth and
required immediate intermittent positive pressure
ventilation. His need for respiratory support
increased and at 28 hours his condition deteriorated
sharply; ultrasonographic examination showed a
grade four intraventricular haemorrhage.
He needed artificial ventilation for some weeks
and at the age of 2 months he was referred for
further management of post-haemorrhagic hydrocephalus. Three days later he had a ventriculoperitoneal shunt inserted. At first his postoperative
ing.6
course was uncomplicated, but three weeks later he
We described three premature infants and one developed paroxysmal episodes of cyanosis with
infant born at term who required lateral ventricular respiratory irregularities; ultrasound examination of
shunting for control of post-haemorrhagic hydro- the ventricular system showed no increase in
cephalus and who subsequently developed the rare enlargement of the lateral and third ventricles but a
complication of specific enlargement of the fourth disproportionate increase in the size of the fourth
ventricle. They illustrate a breadth of presentation ventricle (fig 1).
and management and are discussed in the context of
Intracranial pressure was measured from the right
lateral ventricle and was 1*07 kPa. A small amount
published experience.
1025

1026 O'Hare, Brown, and Minns

(:N\

K1-)

Fig. 1 a & b Case 1: disproportionate enlargement offourth ventricle on real time ultrasonography.

of ventricular fluid was withdrawn. Thereafter his
respiratory abnormalities resolved, and follow up
ultrasonographic examination showed resolution of
the enlargement of the fourth ventricle. Subsequently the infant made good progress, and at the
age of 18 months showed normal development apart
from slight delay in motor development with a
degree of ataxia.
CASE 2

This infant was delivered at term by spontaneous
vertex delivery. There was no asphyxia at birth, she
was breast fed, and given vitamin K supplements.
She was admitted at the age of 3 weeks with an acute
history of irritability, vomiting, and opisthotonus
and was shown to have bilateral haemorrhages into
the ventricular system. Two weeks later a Rickham
reservoir was inserted into the right lateral ventricle,
but daily removal of cerebrospinal fluid failed to
control her intracranial pressure and she had a
theco-peritoneal shunt inserted.
At the age of 41/2 months computed tomography
of the head showed marked ventricular enlargement
and her theco-peritoneal shunt was replaced with a
ventriculo-peritoneal shunt. After operation she
developed ventriculitis, and Escherichia coli was
isolated from the lumbar and ventricular cerebrospinal fluid. She required seven shunt revisions over
the next four years.
When she was 41/2 years old, she was admitted

with acute pain down the left side of her neck
associated with torticollis and vomiting. It was
impossible to measure intracranial pressure because
of the occluded Rickham reservoir. Computed
tomography showed enlargement of the right lateral
ventricle with obvious enlargement of the fourth
ventricle. (fig 2a). Revision of the right ventriculoperitoneal shunt was performed with subsequent
decrease in the size of the fourth ventricle and
resolution of her symptoms. (fig 2b).

rb;)
Fig. 2a Case 2: disproportionate enlargement offourth
ventrirle.
Fig. 2b Case 2: disproportionate enlargement offourth
ventricle less marked after revision of lateral
ventriculoperitoneal shunt.

Specific enlargement of the fourth ventricle after ventriculo-peritoneal shunt 1027
CASE 3

This baby boy was delivered by emergency caesarian
section at 34 weeks after a concealed placental
haemorrhage. He was asphyxiated at birth and
received full intensive care to minimise secondary brain swelling. During his early course he was
hypoglycaemic and had tonic convulsions. On the
fifth day he sustained a bowel perforation caused by
necrotising enterocolitis. On day 13 he was noted to
have post-haemorrhagic hydrocephalus. Ventricular
taps and removal of cerebrospinal fluid had no effect
and initially he had a theco-peritoneal shunt inserted;
this blocked after 38 hours and he then had a lateral
ventriculo-peritoneal shunt performed.
Follow up ultrasonography showed good control
of his hydrocephalus. At the age of 8 months he was
admitted with ventriculitis due to Staphylococcus
albus, for which he was treated with intravenous and
intrathecal gentamicin and flucloxacillin and shunt
revision. At this time he had cerebral palsy with a
diplegia and a right hemiplegia and microcephaly.
At the age of 3 years 4 months he was readmitted
with a one week history of anorexia and drowsiness.
There was no change in his neurological condition
but intracranial pressure monitoring carried out
through his left Rickham reservoir showed intermittent high pressure readings of 2-67 kPa. There
was a history of a minor head injury one month
before his admission. Computed tomography

Fig. 3a Case 3: isolated enlargement offourth ventricle.

showed obvious enlargement of the fourth ventricle
with small lateral ventricles and a small third
ventricle (fig 3a). He underwent posterior fossa
craniectomy with exploration of the large fourth
ventricle and insertion of a cysto-peritoneal shunt.
Postoperatively computed tomography showed
considerable reduction in the size of the fourth
ventricle (fig 3b) and his symptoms resolved.
CASE 4

This baby girl was delivered by spontaneous vertex
delivery at 28 weeks; no asphyxia or respiratory
distress was noted at birth. Her occipitofrontal
circumference increased abnormally suggesting the
development of hydrocephalus, and an ultrasound
examination at 2 months of age showed gross
enlargement of both lateral ventricles and a left
porencephalic cyst. At the age of 2 months a
ventriculo-peritoneal shunt was inserted. A month
later isolated enlargement of the fourth ventricle
was detected on ultrasound examination. At the age
of 6 months she had a left hemiparesis.
When she was 9 months old she presented with
paroxysmal episodes of right sided weakness
associated with fever, and computed tomography
showed that the lateral ventricles were not grossly
enlarged and the third ventricle was small and
displaced. There was a large cystic area in the
posterior fossa comprising the grossly distended

Fig. 3b Case 3: resolution ofenlargement offourth
ventricle after cysto-peritoneal shunt diversion.

1028

O'Hare, Brown, and Minns

t cl I

j

Fig. 4a Case 4: isolated enlargement offourth ventricle.
Fig. 4b Case 4: reduction in size offourth ventricle after
cysto-peritoneal shunt diversion.

fourth ventricle, and there appeared to be no
obvious communication between the fourth ventricle and the supratentorial ventricular system. Her
right hemiparesis resolved completely on conservative management.
Four months later she was admitted with an acute
attack of anorexia and vomiting. Clinical examination
revealed neck retraction, hypertension, and bradycardia. Intracranial pressure measured through a
right lateral ventricle Rickham reservoir was
1-33-1*60 kPa. Computed tomography showed
persisting enlargement of an isolated fourth
ventricle (fig 4a) and a cysto-peritoneal shunt was
inserted. Follow up computed tomography (fig 4b)
showed an improvement in the size of the ventricle
and the raised intracranial pressure in the posterior
fossa had resolved.
Discussion

Specific enlargement of the fourth ventricle can take
the form of a disproportionate enlargement or a
complete isolated encystment, and although the
presentation of the two conditions may be similar in
children who have had shunts inserted for posthaemorrhagic hydrocephalus the management is
quite different.
In disproportionate enlargement of the fourth
ventricle the aqueductal canal is patent and this can
be shown using air contrast studies or Conray
ventriculography, whereas the foramina of Luschka
and Megendi are occluded.7 The abnormality arises
when this type of occlusion occurs at the same time
as abnormal weakness of the brain stem parenchyma
around the fourth ventricle. Collision of cerebrospinal fluid against the obstruction produces a

'water hammer' effect and results in enlargement
of the fourth ventricle. This disproportionate
enlargement can be detected by real time ultrasonography of the posterior fossa8 or computed tomography of the head.
The presenting symptoms of both the cases with
this condition suggested raised pressure in the
posterior fossa with bradycradia and respiratory
irregularities prominant in case 1, and neck pain and
torticollis in case 2. Although the intracranial pressure
was not raised in case 1 when measured through a
Rickman reservoir in his right lateral ventricle, his
symptoms resolved when cerebrospinal fluid was
removed from the lateral ventricle. In case 2 the
disproportionate enlargement of the fourth ventricle
improved and she became asymptomatic after
revision of her lateral ventriculoperitoneal shunt.
Presumably these forms of management are
effective because the enlarged fourth ventricle
remains in communication with the rest of the
ventricular system.
The fourth ventricle may, however, become
completely isolated from the rest of the ventricular
system when both the aqueduct of Sylvius and the
exit foramina of Lushka and Magendi are occluded.9
This acquired condition must be distinguished from
other cystic malformations of the posterior fossa.1'
It is never a primary disease in that it occurs only
after shunting of the lateral ventricular system for
communicating hydrocephalus, aqueduct stenosis,
or obstruction to the outlets of the fourth ventricle'1
and only in the presence of the foreign body of the
shunt.
The initial reports of the condition were in
patients who had communicating hydrocephalus
after infections of the central nervous system.'2 13
Even in the children who develop the condition after
shunting for post-haemorrhagic hydrocephalus the
likelihood of isolation of the fourth ventricle
developing is greater if they have had episodes of
ventriculitis and many children have had multiple
shunt revisions.14
Isolation of the fourth ventricle may be associated
with clinical signs of an expanding lesion in the
posterior fossa and ataxia, neck retraction, torticollis,
nystagmus, diplopia, and dysarthria should alert a
clinician to the condition in children who have
previously had a shunt inserted for post-haemorrhagic
hydrocephalus. Not only might the symptoms be of
a non-localising nature, for example, anorexia and
lethargy,9 but a subtle deterioration may be difficult
to detect in these children, who often have several
handicaps, and therefore a high index of suspicion is
required.
Measurement of the intracranial pressure through
a reservoir in the lateral ventricle does not always

Specific enlargement of the fourth ventricle after ventriculo-peritoneal shunt 1029
help: it was raised in one of our cases and normal in
another. If the intracranial pressure is measured
directly from the fourth ventricle, however, it is
raised in symptomatic cases. 15 As the fourth ventricle
is completely isolated from the rest of the ventricular system, a mistaken diagnosis of malfunction
of the lateral ventriculoperitoneal shunt leading to
subsequent revision of the shunt may lead to upward
herniation of the enlarged fourth ventricle, whereas
independent shunting of the cystic ventricle results
in control of the condition.9
References
Bowerman RA, Donn SM, Silvers TM, Jaffe MH. Review.
Natural history of periventricular-intraventricular haemorrhage
and its complications: sonographic observations. American
Journal of Neuroradiology 1984;143:1041-52.
2 Levene NI, Williams JL, Fawer CL. Ultrasound of the infants
brain. Spastics International Medical Publications, Oxford:
Blackwell Scientific Publications Ltd, 1985.
3 Ellington JL, Zachary RB. Block of arachnoid villus by
subarachnoid haemorrhage. J Neurosurg 1969;30:651-7.
4 Hill AE, Morgan MEI. Post haemorrhagic hydrocephalus in the
newborn term infants. Arch Dis Child 1985;60:706-9.
5 Cook RWI. Early prognosis of low birthweight infants treated
for progressive post haemorrhagic hydrocephalus. Arch Dis
Child 1983;58:410-4.
6 Scher MS, Wright FS, Lockman LA, Thomson TR. Intra-

ventricular haemorrhage in the full term neonate. Child
Neurology 1982;39:769-72.
7 Matsumoto M, Kushida Y, Shibata I, Seiki Y, Terao H.
Disproportionately large communicating fourth ventricle.
Report of two cases. No Shinkei Geka 1983;11:1185-90.
Grant EG, Schellinger D, Richardson JD. Real time ultrasonography of the posterior fossa. J Ultrasound Med 1983;2:73-87.
9 Scotti G, Musgrave MA, Fitz CR, Harwood-Nash DC. The
isolated fourth ventricle in children: CT and clinical review of
sixteen cases. American Journal of Neuroradiology 1980;135:
1233-8.
Rayband C. Cystic malformation of the posterior fossa.
J Neuroradiol 1982;9:103-33.
Hawkins JC, Hoffman HJ, Humphries RP. Isolated fourth
ventricle as a complication of ventricular shunting. Report of
three cases. J Neurosurg 1978;49:910-13.
12 Defeo DR, Foltz EL, Hamilton AE. Double compartment
hydrocephalus in a patient with cysticercosis meningitis. Surg
Neurol 1975;4:247-51.
13 Kalsbeck JE, Desousa AL, Kleiman MB, Goodman JM,
Franken EA. Compartmentalisation of the cerebral ventricles as
a sequelae of neonatal meningitis. J Neurosurg 1980;52:547-52.
14 Mickle JP, Friedman WAE. Cysted fourth ventricle. Surg
Neurol 1981;16:150-3.
15 Foltz EL, Defeo DR. Double compartment hydrocephalus-a
new clinical entity. Neurosurgery 1980;7:551-9.

Correspondence to Dr A E O'Hare, Department of Paediatric
Neurology, Royal Hospital for Sick Children, Sciennes Road,
Edinburgh.
Received 22 April 1987