Neuro-radiology

Neuroradiology (1992) 34:52-56

9 Springer-Verlag 1992

MRI of patients with cerebral palsy and normal CT scan
P. van Bogaert 1, D. Baleriaux 2, C. Christophe 2, and H. B. Szliwowski ~
Departments of 1 Neurology (Pediatric Neurology) and
Radiology (Neuroradiology), H6pital Erasme, Universit6 Libre de Bruxelles, Brussels, Belgium
Received: 5 November 1990

Summary. Three children with clinical evidence of cerebral palsy (CP) and normal cerebral computed tomography (CT) scans were evaluated by magnetic resonance
imaging (MRI) to identify CT-undetectable white matter
lesions in the watershed zones of arterial territories. The
two patients with spastic diplegia showed bilateral lesions
either in the subcortical regions or in the occipital periventricular regions. The patient with congenital hemiplegia
exhibited unilateral lesions in the periventricular region.
We conclude that MRI is more informative than CT for
the evaluation of patients with CR
Key words: Cerebral palsy - Congenital hemiplegia Spastic diplegia - Magnetic resonance imaging - Computed tomography - Periventricular leukomalacia

Spastic diplegia (SD) and congenital hemiplegia are the
most common forms of cerebral palsy (CP), defined as a
permanent impairment of movement and posture resulting from a non-progressive brain disorder.
SD is a spastic paresis of the lower extremities with
variable but lesser involvement of the upper limbs. It is
one of the most characteristic clinical presentations of
periventricular leukomalacia (PVL) [1]. This condition
occurs preferentially in premature infants as the result of
ischemic infarction of the periventricular white matter [2].
Nevertheless, SD can occur in full-term infants with normal birth weight, and in these cases the causes and the
neuropathologic correlates seem to be more varied [3].
Congenital hemiplegia is defined as unilateral motor
disability. It has many etiologies but results most often
from unilateral ischemic infarction in the distribution of
the middle cerebral artery or in the periventricular region

[1].
CP can be documented by sonography during the neonatal period [4, 5] or by CT scanning [6-13]. Unfortunately, the results of neonatal sonograms are not always available or contributory, and the CT scan can be normal
[7-13]. We have investigated the value of MRI in three

children with CP and absence of abnormalities on their
CT scans.

Patients and methods
Patient 1 is a 3-year-old boy born by cesarian section for
acute fetal distress at 39weeks gestational age. Birth
weight was 2280 g. Apgar score was 0 at 1 min and 4 at
5 min. After resuscitation, he was treated for 3 weeks in a
neonatal intensive care unit. A few seizures were noted
during his first 3 days of life. Cranial sonography and CT
scanning performed on the 5th day were normal. His psychomotor development was retarded, and he began to
walk at the age of 30 months. His head circumference,
length and weight are all below the third percentile. He
speaks short sentences with a marked dysarthria. Examination of the cranial nerves is otherwise normal. Motor
function testing discloses a spastic paresis predominating
at the lower limbs with a characteristic spastic gait. Milder
involvement of the upper limbs is noted for finger movements. His development quotient on the Brunet-Lezine
scale is 80.
Patient 2 is a 5-year-old boy, born prematurely at
33 weeks gestational age by cesarian section because of
maternal hemorrhage; his birth weight was 1180 g. Apgar
score was 5 at i min and 8 at 5 min. A few hours after birth
he developed severe respiratory distress from hyaline
membrane disease, requiring prolonged artificial ventilation in a neonatal intensive care unit. Cranial sonogram on
the 19th day of life demonstrated a left subependymal hemorrhage. His milestones were delayed but constantly progressive; he began to walk at 21 months of age and to talk at
24 months of age. Surgery was performed to correct an alternating convergent strabismus at 27 months of age. Head
circumference, length and weight are in the third percentile. The clinical picture is that of an asymmetric SD which
predominates on the right side: fine motricity of the fingers
is better on the left side, the bilateral equinovarus deformity of the feet and hyperreflexia are more pronounced on
the right side, and a Babinski sign is present bilaterally.

53

I=l

It

Fig.la--e. Patient 1.

L.

a Normal CT scan. Comparison with
b,c (SE T2-weighted images) and
d, e (IR Tl-weighted images). There are
multiple scattered lesions in the frontal and
occipital subcortical white matter with involvement of the centrum semiovale

e

Patient 3 is an 8-year-old girl, born after 41 weeks gestation with a birth weight of 2540 g. She cried spontaneously but there was a question of mild hypothermia on
her 1st day of life. During the first months, her parents reported that her left wrist nearly always remained closed,
and that she manipulated objects with her right hand preferentially. She began to walk at 15 months of age with an
abnormal gait characterized by circumduction of her left
leg. Clinical examination at 2 years of age revealed a clear
picture of left hemiplegia with spontaneous positioning of
the left forearm in flexion and pronation, hyperextensibility of the fingers, clumsiness with prehensile movement
of objects with the left hand, equinovarus deformity of the
left foot, left hyperreflexia and left Babinski sign. Her congenital hemiplegia has remained moderate until now with
physical therapy. Development of cognitive functions has
been normal.
MRI was performed on a superconducting 0.5 T system. For all patients, T2-weighted images were obtained

in either axial or coronal planes with a standard spin echo
(SE) technique (TE: 40 and 95 ms; TR: 2500 ms); T1weighted images were obtained in the axial plane with an
inversion recovery (IR) technique (TE: 35 ms, TI: 300 ms,
TR: 1500 ms).
CT scans were performed for all patients with a 2-ram
slice thickness and a 2-mm interslice gap. Images were
viewed with a narrow window.
Results

The CT scans were normal in all cases (Figs. la, 2 a and
3 a). In patient 1, MRI demonstrated scattered lesions in
the subcortical white matter of the frontal and occipital
lobes. These lesions were hyperintense on T2-weighted
images (Fig. 1 b, c). The same lesions appeared less extensive and hypointense on Tl-weighted images (Fig. 1 d, e).
In patient 2, T2-weighted images exhibited periventricular hyperintense lesions in front of the trigone of the

54

Fig.2a-d. Patient 2. a Normal CT scan. Comparison with b,c (coronal SE T2-weighted
images) and d (axial IR Tl-weighted image). There are bilateral lesions which predominate in the left occipital periventricular region

lateral ventricle. However, coronal images demonstrated
clear asymmetry of the lesions, which were more extensive on the left side (Fig. 3 b, c). Tl-weighted images in the
axial plane confirmed this asymmetric involvement of the
occipital periventricular regions as hypointense areas
(Fig.3 d).
Patient 3, with congenital left hemiplegia, exhibited
two right lateral periventricular lesions: the more extensive is located in front of the trigone while a smaller one is
seen in the frontal region (Fig.4b-c). These lesions are
hyperintense on T2-weighted images and hypointense on
Tl-weighted images.

Discussion
In three patients with CR MRI identified abnormalities
not visible on CT scan. These abnormalities consist of
hyperintense white matter lesions on T2-weighted images
with corresponding hypointense but less extensive abnormalities on Tl-weighted images. In all cases, the abnormalities are situated in the watershed zones between the

anterior, middle and posterior cerebral arteries and presumably result from an ischemic injury.
The two patients with SD exhibit bilateral white matter
lesions. The localization of these watershed infarcts differs in the first case when compared to the second one. In
the first patient, a full-term child with SD, the lesions are
located predominantly in the subcortical white matter and
appear to correspond to a mild parasagittal cerebral injury
without cortical involvement. Indeed parasagittal cerebral injury is characteristic of the full-term child with intrauterine asphyxia, and consists in necrosis of the cortex
and the subcortical white matter especially in the parietooccipital region [1]. In the second case of SD, we suggest
that the lesions correspond to mild PVL, since the involvement of the occipital radiation at the trigone of the lateral
ventricle is a constant autopsy finding in PVL [2]. It is of
great interest in this patient that his asymmetric neurologic impairment (predominantly right paresis) correlates
well with the asymmetric MRI abnormalities which are
more pronounced in the left periventricular region.
Several reports have already described the MRI appearance of PVL, but most of them address premature in-

55

Fig. 3 a-c. Patient 3. a Normal CT scan. Comparison with b (SE T2-weighted image) and c (IR Tl-weighted image). There are right periventricular lesions

fants with evidence of P V L on sonograms [14, 15] or CT
scans [16]. Flodmark et al. [16] concluded that M R I is
m o r e precise than CT scanning in the evaluation of PVL.
The abnormalities demonstrable on CT scans (reduction
in the amount of periventricular white matter, widening of
sutci, ventriculomegaly, irregular ventricular outline) are
m o r e precisely delineated on I R images. Moreover, on T2weighted images these authors described hyperintense lesions located in the occipital lobes beyond the periventricular region without involvement of the centrum
semiovale and the subcortical white matter. In their study,
there were no cases of CP with normal CT scan. More recently, K o e d a et al. [17] reported their M R I observations
in p r e t e r m and t e r m infants with SD. In one t e r m infant
they found isolated periventricular lesions extending to
the subcortical white matter. These lesions appeared hypointense on T l - w e i g h t e d images and hyperintense on
T2-weighted images; CT scan results were not mentioned
in their cases.
O u r patient with congenital hemiplegia (no. 3) exhibits
lateral periventricular lesions which may correspond to
unilateral PVL. No other report has emphasized the importance of M R I in the evaluation of infants with congenital hemiplegia. We believe that the physiopathology of
this condition could be better understood by use of MRI,
especially for patients with normal CT scans.
In conclusion, M R I evaluation of children with CP m a y
disclose white matter lesions not visible on CT scan images. These lesions are most evident on T2-weighted images as hyperintense areas. M R I is useful in all conditions,
such as the leukodystrophies, in which the white matter is
affected [18]. Clinical evaluation of these children remains of primary importance. Other investigations must
be p e r f o r m e d if a progressive neurologic deterioration is
suspected.

References

1. Volpe JJ (1987) Neurology of the newborn. Saunders, Philadelphia, pp 209-279
2. Shuman RM, Selednik LJ (1980) Periventricular leukomalacia: a
one-year autopsy study. Arch Neuro137: 231-235
3. Veelken N, Hagberg B, Hagberg G, Olow I (1983) Diplegic cerebral palsy in Swedish term and preterm children: differences in
reduced optimality, relations to neurology and pathogenic factors. Neuropediatrics 14:20-28
4. Bowerman RA, Donn SM, Di Pietro MA, D'Amato CJ, Hicks
SP (1984) Periventricular leukomalacia in the pre-term newborn
infant: sonographic and clinical features. Radiology 151:383-388
5. Dubowitz LMS, Bydder GM, Mushin J (1985) Developmental
sequence of periventricular leukomalacia: correlation of ultrasound, clinical, and nuclear magnetic functions. Arch Dis Child
60:349-355
6. Flodmark O, Roland EH, Hill A, Whitfietd MF (1987) Periventricular leukomalacia: radiologic diagnosis. Radiology 162: 119124
7. Taudorf K, Melchior JC, Pedersen H (1984) CT findings in spastic cerebral palsy: clinical, aetiological and prognostic aspects.
Neuropediatrics 15:120-124
8. Koch B, Braillier D, Eng G, Binder H (1980) Computerized tomography in cerebral-palsied children. Dev Med Child Neurol
22:595-607
9. Kulakowski S, Larroche JC (1980) Cranial computerized tomography in cerebral palsy. An attempt at anatomo-clinical and
radiological correlations. Neuropediatrics 11:339-353
10. Kotlarek F, Rodewig R, Brull D, Zeumer H (1981) Computed tomographic findings in congenital hemiparesis in childhood and
their relation to etiology and prognosis. Neuropediatrics 12: 101109
ii. Cohen ME, Duffner PK (1981) Prognostic indicators in hemiparetic cerebral palsy. Ann Neurol 9:353-357
12. Molteni B, Oleari G, Fedrizzi E. Bracchi M (1987) Relation
between CT patterns, clinical findings and etiological factors
in children born at term, affected by congenital hemiparesis.
Neuropediatrics 18:75-80

56
13. Wiklund LM, Uvebrant P, Flodmark O (1990) Morphology of
cerebral lesions in children with congenital hemiplegia: a study
with computed tomography. Neuroradiology 32:179-186
14. Dubowitz LMS, Bydder GM (1985) Nuclear magnetic resonance
imaging in the diagnosis and follow-up of neonatal cerebral injury. Clin Perinatol 12:243-260
15. Wilson DA, Steiner RE (1986) Periventricular leukomalacia:
evaluation with MR imaging. Radiology 160:507-511
16. Flodmark O, Lupton B, Li D, Stimac GH, Roland EH, Hill A,
Whitfield MF, Norman MG (1989) MR imaging of periventricular leukomalacia in childhood. AJR 152:583-590
17. Koeda T, Suganuma I, Kohno Y, Takamatsu T, Takeshita K
(1990) MR imaging of spastic diplegia: comparative study between preterm and term infants. Neuroradiology 32:187-190

18. Kumar AJ, Rosenbaum AE, Naidu S, Wener L, Citrin CM, Lindenberg R, Kim WS, Zinreich SJ, Molliver ME, Mayberg HS,
Moser HW (1987) Adrenoleukodystrophy: correlating MR imaging with CT. Radiology 165:497-504

Dr. P.van Bogaert
Department of Neurology
H6pital Erasme
Universit6 Libre de Bruxelles
Route de Lennik, 808
B-1070 Brussels
Belgium