Pediatrics International (2002) 44, 680–682

Patient Report

White matter lesion due to dural sinus thrombosis in an infant with
subdural hematoma
KIMITAKA TAKITANI,1 ATSUSHI NISHINO,1 TAKUYA TANABE,1 HARUKI TANAKA,1
KOSHI HARADA,2 TAKASHI MIMAKI3 AND HIROSHI TAMAI1
1
Department of Pediatrics, Osaka Medical College, 2Department of Radiology, Sakai Municipal Hospital,
Osaka and 3Faculty of Education, Gifu University, Gifu, Japan
Key words

dural sinus thrombosis, infant, subdural hematoma, white matter lesion.

In infants with acute subdural hematoma, computed
tomography (CT) scans often reveal diffuse hemispheric low
density.1 The pathophysiology of this cerebral white matter
lesion is probably venous congestion caused by injury to
bridging veins. In the present case, a diffuse cerebral white
matter lesion shown by CT due to the thrombus of dural
sinus diagnosed by magnetic resonance imaging (MRI) in an
infant with acute subdural hematoma is described.

Case report
A 2-month-old girl was admitted for vomiting and decreasing
level of consciousness. She had no prior history of trauma or
infection. Clinical examination was remarkable for lethargy,
bulging and pulsatile fontanel, prolonged light reflex in the
right pupil, bilateral deviation of eyes to the right, and
increased deep tendon reflexes. Laboratory findings were
within normal limits except for hemoglobin of 6.4 g/dL.
Prothrombin time, partial thromboplastin time, and antithrombin III were all within normal limits.
Cranial CT on admission revealed a right subdural
hematoma (Fig. 1a). Cerebral angiography was not carried
out. The patient was started on glycerol, dexamethasone, and
phenobarbital. Three days after admission, she exhibited leftsided clonic convulsions for which a continuous infusion of
lidocaine was initiated. On day 5 of admission, a repeat CT
demonstrated some improvement in the right subdural
hematoma. Diffuse low density areas were noted in the white
matter of bilateral cerebral hemispheres. Increased density

Correspondence: Kimitaka Takitani MD, Department of Pediatrics,
Osaka Medical College, 2-7 Daigakumachi, Takatsuki, Osaka 5698686, Japan. Email: ped016@poh.osaka-med.ac.jp
Received 25 October 2001; revised 27 February 2002; accepted
25 April 2002.

was present within the straight sinus and the confluence of
the sinus (Fig. 1b).
On day 6, MRI was carried out which demonstrated high
signal intensity within the confluence of the sinus on T2- and
proton-density weighted images (Fig. 2). The right subdural
hematoma was well visualized, but the MRI did not show
relevant changes in the white matter.
Seizures decreased in frequency and severity. Glycerol,
dexamethasone, and lidocaine were discontinued. On day 14
of admission, she could orally take milk. She continued to
improve with phenobarbital treatment and was discharged on
day 30. At discharge, she had no further seizures, was alert,
but had a mild left hemiplegia. Follow-up CT showed no low
density areas in either frontal lobes, but they were present in
the right occipital lobe (Fig. 3). At 12 months of age, her
development was normal and the mild left hemiplegia had
resolved.

Discussion
In neonates, venous thrombosis has been associated with
acute systemic disease, including congenital heart disease,
shock, metabolic acidosis, sepsis, or acute dehydration, and
undetermined causes.2 In older infants and children, cerebral
venous thrombosis have demonstrated etiologic associations
with acute dehydration and local infection such as
meningitis, otitis media, or scalp cellulitis. More recently,
venous thrombosis has resulted from hypercoagulate states
caused by protein C deficiency or the mutation in the factor V.3
Cerebral venous thrombosis in children is a rare disorder.4
The clinical diagnosis of dural sinus thrombosis is difficult
due to the non-specifity of the clinical presentation.
Symptoms can include headache, nausea, vomiting, disorientation, seizures, decreased level of consciousness, or speech
and motor disturbances. The CT findings of dural sinus
thrombosis can be divided into direct and indirect signs.

White matter lesion due to sinus thrombosis

681

Fig. 3 Follow-up non-contrast computed tomography scan
obtained at discharge shows low density area in the right occipital
lobe on day 35 of admission.

Fig. 1 (a) Cranial computed tomography (CT) scans on admission demonstrates a right subdural hematoma. (b) Non-contrast
CT scans obtained 5 days after admission shows low density
areas in the white matter of bilateral cerebral hemispheres and
increased density within the straight sinus and the confluence of
the sinus.

Fig. 2 (a) T2-weighted image (2000/100) (a repetition time/an
echo time). (b) Proton density weighted image (2000/30). Axial
images reveal intense signal in the confluence of the sinus
(arrow).

Direct signs include the ‘cord sign’ and the ‘empty delta’
sign are specific for this condition and consist of the actual
visualization of a thrombus within a vein or sinus.5 Indirect
signs are less specific, but may suggest venous thrombosis.
These include low density areas, hemorrhage, small
ventricles, gyral enhancement, and dilated collateral veins.6
Specific CT findings are not always present and can be
sometimes overlooked.6,7 In addition, bony structures may
cause artifacts which impair the evaluation of the sinuses.
Conventional or digital subtraction angiography may be
considered as confirmatory tests,8 but these are invasive and
should not be used without a specific indication.6 Magnetic
resonance angiography provides excellent visualization of
dural sinuses and cerebral veins non-invasively.9 This is a
very useful imaging technique to evaluate venous sinus
thrombosis. Magnetic resonance angiography, as well as
cerebral angiography, was not carried out in the present case.
However, we consider the diagnosis of dural sinus
thrombosis by clinical symptoms and findings of MRI at the
acute phase.
Magnetic resonance imaging in this case demonstrated
high signal intensity within the confluence of the sinus due to
clotted blood. This was noted on T2- and proton-density
weighted images. The thrombus was most likely related to
the subdural hematoma. Previous reports have demonstrated
non-specific low density areas on CT as findings with venous
infarction,5 venous congestion1 or ischemic change.10 In our
case, the thrombus may have caused the hypoperfusion
lesions in the bilateral cerebral hemispheres on CT examinations, because these are not confined to the blood supply of a
single artery. Although the MRI findings were not consistent
with venous infarction suggested by the low densities on
unenhanced CT, it is probable that the diffuse hypoperfusion
lesion of the cerebral white matter due to the thrombus of the
confluence of the sinus is reversible.
The CT scan prior to discharge demonstrated a low
density area only in the right occipital lobe. The lesions in the
other lobes were diminished. We speculate that there is a
difference of the degree of damage in each lobe. At discharge,
follow-up MRI demonstrated recanalization of the sinus.
Repeated CT showed no diffuse cerebral white matter
lesions. We speculate that a reversible hypoperfusion lesion

682

K Takitani et al.

of cerebral white matter following venous congestion due to
thrombosis of the dural sinus may have caused the low
density areas seen on the CT on day 5. The present case suggests
that, in neonates and children whose CT show diffuse low
densities, the presence of cerebral venous thrombosis should
be considered.

References
1 Fujimoto K, Shimomura T, Okumura Y, Sakaki T. CT shows
ipsilateral cerebral hemispheric low density area after acute
subdural hematoma in infants. Nervous System Children 1999;
24: 504–8 (in Japanese).
2 Barron TF, Gusnard DA, Zimmerman RA, Clancy RR.
Cerebral venous thrombosis in neonates and children. Pediatr.
Neurol. 1992; 8: 112–16.
3 Vielhaber H, Ehrenforth S, Koch HG, Scharrer I, van der
Werf N, Nowak-Gottl U. Cerebral venous sinus thrombosis in

infancy and childhood: role of genetic and acquired risk factors
of thrombophilia. Eur. J. Pediatr. 1998; 157: 555–60.
4 deVeber G, Andrew M, Adams C et al. Cerebral sinovenous
thrombosis in children. N. Engl. J. Med. 2001; 345: 417–23.
5 Kim KS, Walczak TS. Computed tomography of deep cerebral
venous thrombosis. J. Comput. Assist. Tomogr. 1986; 10: 386–90.
6 Hulcelle PJ, Dooms GC, Mathurin P, Cornelis G. MRI
assessment of unsuspected dural sinus thrombus. Neuroradiology 1989; 31: 217–22.
7 Synder TC, Sachdev HS. MR imaging of cerebral dural sinus
thrombosis. J. Comput. Assist. Tomogr. 1986; 10: 889–91.
8 Hanigan WC, Rossi LJ, McLean JM, Wright RM. MRI of
cerebral vein thrombosis in infancy: a case report. Neurology
1986; 36: 1354–6.
9 Wang AM. MRA of venous sinus thrombosis. Clin. Neurosci.
1997; 4: 158–64.
10 Eick JJ, Miller KD, Bell KA, Tutton RH. Computed
tomography of deep cerebral venous thrombosis in children.
Radiology 1981; 140: 399–402.