Anda di halaman 1dari 6

J Neurosurg 81:20--23, 1994

Extracerebral fluid collections in infancy: role of


magnetic resonance imaging in differentiation between
subdural effusion and subarachnoid space enlargement
NOBUHIKO Ama,

M.D.

Department of Neurosurgery, Tokyo Metropolitan Ohkubo Hospital, Tokyo, Japan


The pathological process of extracerebral fluid collections in infancy includes subdural effusion and enlargement of the subarachnoid spaces. Both conditions have traditionally been investigated as a single clinical
entity, because of difficulty in differentiating between them. The prognosis of subdural effusion is not as benign
as that of enlargement of subarachnoid spaces, requiring differential diagnosis between these disorders. The
present study was conducted to elucidate whether this differentiation could be made on magnetic resonance
(MR) images.
The series consisted of 16 infants aged 10 months or younger, including eight with verified subdural effusion
and eight in whom a diagnosis of enlargement of the subarachnoid spaces was achieved by neuroimaging
studies other than MR imaging. In all eight patients with subdural effusion, the intensity of the fluid was greater
than that of cerebrospinal fluid (CSF) in at least one of the sequences using Tcweighted, proton-density, and
T 2-weighted MR images. The flow-void sign, indicating vessels in the fluid spaces, was not seen in any of
these eight patients. On the other hand, in all eight patients with enlargement of the subarachnoid spaces, the
fluid was isointense in relation to CSF, and vascular flow-void areas were seen in at least one of the MR
imaging sequences. Based on these observations, it is concluded that differentiation between subdural effusion
and enlargement of the subarachnoid spaces can be established by focusing on two aspects of MR imaging
findings: 1) the intensity of the fluid, which is either iso- or hyperintense relative to CSF, and 2) the presence
or absence of vascular flow-void areas in the fluid spaces.

Y'

KEY WoRDS extracerebral fluid collection subdural effusion


subarachnoid space magnetic resonance imaging infant

fluid collections in infancy include two distinct pathological conditions, one


involving the subdural spaces and the other the
subarachnoid spaces, manifesting as subdural effusion
and enlargement of the subarachnoid spaces, respectively. Most reports on benign subdural effusion in infancy have not definitely excluded patients with enlargement of the subarachnoid spaces. 61217 Conversely,
reports on enlargement of the subarachnoid spaces or
external hydrocephalus seem to include patients with
subdural effusion. 111 Moreover, in some series, these
two conditions were investigated as a single clinical
entity with the diagnosis of extracerebral or extra-axial
fluid collections in infancy.7816 Subdural effusion in
infancy is a pathological condition with chronic subdural hematoma, which may increase in volume due
to exudation and repeated hemorrhage, leading to a less
favorable prognosis? Furthermore, when surgical inXTRACEREBRAL

20

tervention is indicated, different procedures are required for these two conditions: lumboperitoneal or
ventriculoperitoneal shunting is preferred for enlargement of the subarachnoid spaces, 3 while subdural effusion is usually treated by subdural-peritoneal shunting. 4
For these reasons, it is necessary to differentiate
between subdural effusion and enlargement of the
subarachnoid spaces. Previous neuroimaging studies,
including high-resolution computerized tomography
(CT) scanning, cannot reliably differentiate between
these conditions. In view of this, the advent of magnetic resonance (MR) imaging seems promising for definitively differentiating between subdural effusion and
enlargement of the subarachnoid spaces; however, few
studies have focused on the differentiation of these
conditions by MR images. 9 The aim of the present
study is to elucidate whether MR images, as a single
neuroimaging method, can differentiate between subJ. Neurosurg. I Volume 81 I July, 1994

Extracerebral fluid collections in infancy


TABLE 1
Magnetic resonance (MR) imaging features in eight
patients with subdural effusion*

dural effusion and enlargement of the subarachnoid


spaces.

Clinical Material and Methods


Between January, 1991, and December, 1992, 16 infants under the age of 12 months were referred to the
Department of Neurosurgery, Tokyo Metropolitan Fuchu Hospital, for evaluation of extracerebral fluid collections by MR imaging. At the time of presentation,
each of these infants was noted to have an occipitofrontal head circumference above the 98th percentile
for their age 14 as well as an extracerebral collection
more than 5 mm thick with a density indistinguishable
from that of cerebrospinal fluid (CSF) on CT scanning.
Based on the results of several neuroimaging studies,
including CT with intravenous contrast infusion, 13 CT
cisternography, and ultrasonography, 15 and details of
the clinical history (particular attention being paid to
a history suggestive of infantile acute subdural hematoma5), a differential diagnosis between subdural effusion and enlargement of the subarachnoid spaces was
established in all16 patients. Infants with extracerebral
fluid collections that were not defined as being located
in the subdural or subarachnoid spaces by the abovedescribed neuroimaging methods were excluded from
this study.
Eight patients were diagnosed as having subdural effusion; the diagnostic methods used included contrastenhanced CT in five, ultrasonography in five, and CT
cisternography in two. Besides these neuroimaging
studies, an episode indicating infantile acute subdural
hematoma5 was highly suggestive of subdural effusion
in two patients. Subdural effusion was later verified by
CT subdurography 4 and surgery in all eight patients;
two also had enlargement of the subarachnoid spaces.
These eight patients included four boys and four girls,
ranging in age from 2 to 10 months (mean 4.6 months).
The remaining eight patients were diagnosed as
having enlargement of the subarachnoid spaces; the diagnostic neuroimaging methods differentiating their
condition from subdural effusion included contrastenhanced CT in six and ultrasonography in five. These
eight patients comprised five boys and three girls, ranging in age from 3 to 9 months (mean 6.25 months).
Extracerebral fluid collections in these 16 patients
were evaluated by T 1-weighted, proton-density, and T2weighted MR images. Particular attention was given
to the intensity of the fluid compared to that of CSF,
and to the presence or absence of vascular flow-void
areas in the fluid spaces.

Results
The intensity of the subdural effusion was greater
than that of CSF in six (75%) of eight patients on Tc
weighted MR images, in all six patients so studied on
proton-density images, and in seven (87.5%) of eight
patients on T2-weighted images (Table 1). Consequently, in all eight patients, at least one sequence of
the MR images demonstrated an intensity greater than
that of CSF (Fig. 1). The flow-void sign, indicating
J. Neurosurg. I Volume 81 I July, 1994

MR Imaging Findings
Age
Case
(mos),
No.
Sex
1

2
3
4

5
6
7
8

T1Weighted
Image

ProtonDensity
Image

T,Weighted
Image

Etiology

3, F hyper, V(-) hyper, V(-) iso, V(-)


head trauma
4,M iso, V(-)
ND
hyper, V(-) meningitis
6, F iso, V(-)
ND
hyper, V(-) acute subdural
hematoma
10, F hyper, V(-) hyper, V(-) hyper, V(-) acute subdural
hematoma
2, F hyper, V(-) hyper, V(-) hyper, V(-) head trauma
3,M hyper, V(-) hyper, V(-) hyper, V(-) unknown
3,M hyper, V(-) hyper, V(-) hyper, V(-) child abuse
6,M hyper, V(-) hyper, V(-) hyper, V(-) meningitis

* Imaging appearance relative to cerebrospinal fluid: hyper = hyperintense, iso = isointense. V(-) = absence of vascular flow-void area;
NO = study not done.

FIG. 1. Left: Computerized tomography scan showing extracerebral fluid collections indistinguishable from cerebrospinal fluid (CSF). Right: Magnetic resonance T1-weighted image demonstrating bilateral subdural effusions, which are
clearly separated from CSF in the subarachnoid spaces. Note
the coexisting enlargement of the subarachnoid spaces on the
right cerebral convexity (asterisk).

vessels in the fluid spaces apart from those in the sulci


or those abutting to the cortical surface, was not delineated in any of the eight patients with subdural effusion.
In all eight patients with enlargement of the subarachnoid spaces, the intensity of the fluid spaces was
consistent with that of CSF. The presence of vascular
flow-void areas apart from the cortical surface was confirmed in two (25%) of eight patients on Tcweighted
images, in all seven patients so studied on proton density, and in seven (87.5%) of the eight patients on T2weighted MR images (Table 2). Consequently, in all
eight patients, at least one sequence of the MR images
demonstrated the flow-void sign (Fig. 2). The results
of MR imaging study are listed in Tables 1 and 2.
21

N. Aoki
TABLE 2
Magnetic resonance (MR) imaging features in eight
patients with enlargement of the subarachnoid spaces*
MR Imaging Findings
Case
No.

9
10
11
12
13
14
15
16

Age
(mos),
Sex

T,-

T,-

Weighted
Image

ProtonDensity
Image

Weighted
Image

6,F
8,M

iso, V(+)
iso, V(-)

iso, V(+)
iso, V(+)

iso, V(+)
iso, V(+)

9,F

iso,
iso,
iso,
iso,
iso,
iso,

ND
iso,
iso,
iso,
iso,
iso,

iso,
iso,
iso,
iso,
iso,
iso,

4,M
6,M
6,M

3,F
8,M

V(+)
V(-)
V(-)
V(-)
V(-)
V(-)

V(+)
V(+)
V(+)
V(+)
V(+)

V(-)
V(+)
V(+)
V(+)
V(+)
V(+)

Etiology

unknown
Solos
syndrome
unknown
unknown
unknown
unknown
unknown
unknown

FIG. 2. Computerized tomography scan (left) and protondensity magnetic resonance (MR) image (right) showing an
abnormal accumulation of fluid, which is identical to cerebrospinal fluid on both neuroimaging studies. Note the vascular flow-void areas (arrows) apart from the cortical surface
on the MR image.

* Imaging appearance: iso = isointense relative to cerebrospinal fluid.


V(-) = absence of vascular flow-void area; V( +) = presence of vascular flow-void area; ND = study not done.

Discussion
The diagnostic approach and treatment indications
for infants with an abnormally enlarged head associated
with an extracerebral fluid collection indistinguishable
from CSF on CT are still uncertain. One reason for this
controversy is associated with the traditional trend in
which subdural effusion and enlargement of the subarachnoid spaces are managed similarly without distinguishing between them. From the author's experience, infants with subdural effusion have often had an
unfavorable prognosis due to an increase in the volume
of the effusion and repeated hemorrhage with subsequent development of chronic subdural hematoma. 2
Previous reports on benign subdural effusion in infancy
seem to include infants with enlargement of the subarachnoid spaces. 612 17 It is necessary to evaluate the
prognosis of the patient with extracerebral fluid collections after differentiating between subdural effusion
and enlargement of the subarachnoid spaces.
Previous neuroimaging studies, including cerebral
angiography, CT with intravenous infusion of contrast
medium, 13 cr subdurography, 4 and ultrasonography, 15
are known to be useful in differentiating between subdural effusion and enlargement of the subarachnoid
spaces; however, cerebral angiography and CT subdurography are not practical because of their invasiveness. Moreover, in my experience, contrast-enhanced
CT and ultrasonography do not always definitely differentiate between the two entities. In fact, in the
present series, the diagnosis of enlargement of the subarachnoid spaces in eight patients was achieved by
contrast-enhanced CT in six (75%) and by ultrasonography in five (62.5%). Magnetic resonance imaging,
used as a single neuroimaging technique, is expected
to offer a definitive differential diagnosis between subdural effusion and enlargement of the subarachnoid
spaces. To my knowledge, however, few studies using
MR imaging have focused on the differentiation of
these two conditions in infancy. 9

22

In the present study, the MR imaging features were


evaluated with particular attention to two aspects: 1)
the intensity of the fluid compared to that of CSF, and
2) the flow-void sign, indicating vessels within the fluid
spaces. It is known that subdural effusion is at times
accompanied by mild enlargement of the subarachnoid
spaces 10 (Fig. 1); in this series, such patients were assigned to the subdural effusion group. In all eight patients with subdural effusion, hyperintensity relative to
CSF was revealed in at least one of the MR imaging
sequences using Tcweighted, proton-density, and T 2weighted images, and failed to delineate vascular flowvoid areas. On the other hand, in all eight patients with
enlargement of the subarachnoid spaces, the lesion exhibited isointensity relative to CSF, and demonstrated
the presence of vascular flow-void areas apart from the
cortical surface in at least one of the MR imaging sequences. These vascular flow-void areas on MR imaging are consistent with the high-density areas on CT
after infusion of contrast material in patients with enlargement of the subarachnoid spaces in infancy. 13
It may be useful to evaluate protein content in the
fluid collections and resultant signal intensity changes
on T 1 - and T 2-weighted images because high protein
content is reportedly associated with frequent occlusion of subdural-peritoneal shunting. Moreover, although no contrast-enhanced MR images were obtained in the present series, this technique might help
to demonstrate not only cortical vessels but also outer
and/or inner membranes of subdural effusion, making
a more definitive differentiation between the two conditions.
Theoretically, extracerebral fluid collections would
include the accumulation of normal CSF in the subdural spaces and the accumulation of high-protein fluid
in the subarachnoid spaces; in this series, however, no
such features were delineated on MR images. This may
be attributed to the small number of the patients in the
present study. The differentiation between subdural effusion and enlargement of the subarachnoid spaces was
established in these patients by focusing on two aspects
of MR images: 1) the intensity of the fluid, which is
1. Neurosurg. / Volume 81 /July, 1994

Extracerebral fluid collections in infancy


iso- or hyperintense relative to CSF, and 2) the presence
or absence of vascular flow-void areas in the fluid.

11.

References
1. Andersson H, Elfverson J, Svendsen P: External hydrocephalus in infants. Childs Brain 11:398-402, 1984
2. Aoki N: Chronic subdural hematoma in infancy. Clinical
analysis of 30 cases in the CT era. J Neurosurg 73:
201-205, 1990
3. Aoki N: Lumboperitoneal shunt: clinical applications, complications, and comparison with ventriculoperitoneal shunt.
Neurosurgery 22:988-1004, 1990
4. Aoki N, Masuzawa H: Bilateral chronic subdural hematomas without communication between the hematoma cavities: treatment with unilateral subdural-peritoneal shunt.
Neurosurgery 22:911-913, 1988
5. Aoki N, Masuzawa H: Infantile acute subdural hematoma.
Clinical analysis of 26 cases. J Neurosurg 61:273-280,
1984
6. Briner S, Bodensteiner J: Benign subdural collections of
infancy. Pediatrics 67:802-804, 1980
7. Carolan PL, McLaurin RL, Towbin RB, et al: Benign extraaxial collection of infancy. Pediatr Neurosci 12:140-144,
1986
8. Date H, Ito C, Yamaura A, et al: [Extracerebral fluid collections in infancy.] CurrTrends Hyd 2:20-27, 1992 (Jpn)
9. Fukushima K, Takagi T, Nagai H, et al: [MRI of subdural
fluid collections in infancy.] Shoni No Noshinkei 13:
381-389, 1988 (Jpn)
10. Masuzawa H, Aoki N, Sato J: [Subdural effusion (hema-

J. Neurosurg. I Volume 81 I July, 1994

12.
13.
14.
15.

16.
17.

toma) accompanied by enlargement of subarachnoid


spaces.] Neurol Med Chir 22:751-756, 1982 (Jpn)
Ment LR, Duncan CC, Geehr R: Benign enlargement of the
subarachnoid spaces in the infant. J Neurosurg 54:
504-508, 1981
Mori K, Handa H, Itoh M, et a!: Benign subdural effusion
in infancy. J Comput Assist Tomogr 4:466-471, 1980
Morota N, Sakamoto K: [Infantile expanding subarachnoid
collection and CT findings.] Prog Comput Tomogr 10:
153-160, 1988 (Jpn)
Nellhaus G: Head circumference from birth to eighteen
years. Practical composite international and interracial
graphs. Pediatrics 41:106-114, 1968
Nishimoto H, Aihara T, Tsubokawa T: Infantile subdural
fluid collection and enlargement of subarachnoid spaces in
infancy. Diagnosis by ultrasound. Childs Nerv Syst 6:
281-282, 1990 (Abstract)
Nogueira GJ, Zaglul HF: Hypodense extracerebral images
on computed tomography in children. "External hydrocephalus": a misnomer? Childs Nerv Syst 7:336-341, 1991
Robertson WC Jr, Chun RW, Orrison WW, et a!: Benign
subdural collections of infancy. J Pediatr 94:382-385,
1979

Manuscript received April 16, 1993.


Accepted in final form November 3, 1993.
Address reprint requests to: Nobuhiko Aoki, M.D., Department of Neurosurgery, Tokyo Metropolitan Ohkubo Hospital,
44-1 Kabukicho 2-chome, Shinjuku-ku, Tokyo 160, Japan.

23