CONFERENCIA MAGISTRAL

A magnetic resonance approach

to metabolic disorders in childhood
A.J. Barkovich

A MAGNETIC RESONANCE APPROACH TO METABOLIC DISORDERS IN CHILDHOOD

Summary. Introduction. Inborn errors of metabolism are a difficult group of disorders to diagnose. Clinical signs and symptoms
overlap among the different diseases, biochemical tests are often nonspecific or unrevealing, and genetic analyses are time
consuming, expensive, and often not commercially available. Imaging can aid in diagnosing some disorders, but the imaging
findings of many disorders seem to overlap, particularly as more and more disorders are identified and described.
Development. Several groups have proposed an imaging approach based upon pattern recognition on magnetic resonance
imaging, supplemented by metabolic data from proton magnetic resonance spectroscopy and microstructural data, as assessed by
diffusion weighted imaging. Conclusion. When properly used, this pattern recognition approach in conjunction with the other
imaging tools can be very useful for separating the complex group of metabolic disorders into more manageable groups.
Indeed, sometimes this approach allows a specific diagnosis to be made. [REV NEUROL 2006; 43 (Supl 1): S5-16]
Key words. Brain diseases. Canavan disease. Diffusion magnetic resonance imaging. Drug toxicity. Inborn errors of metabolism.
Leukodystrophy. Magnetic resonance imaging. Magnetic resonance spectroscopy. Mitochondrial diseases.

INTRODUCTION PATTERN-BASED APPROACHES

Inborn errors of metabolism are a complicated and difficult
group of disorders for most clinicians. Many such disorders
have been described and more are being discovered every year.
Part of the confusion in dealing with these disorders stems from
the fact that they are classified in many different ways; some
authors classify them by biochemical characteristics or the bio-
chemical pathway that is affected, some by the cellular
organelle in which the affected protein or biochemical pathway
is located, others by clinical characteristics, and still others by
the gene that is affected. Affected patients are not easy to segre-
gate by clinical criteria, as patients with metabolic disorders can
present at nearly any age, depending on the precise mutation to
the affected gene, and their clinical signs and symptoms are
almost invariably nonspecific in separating one disorder from
another. The imaging features of these disorders, too, can be
particularly confusing if not approached methodically. The
imaging appearances of many disorders overlap and, often, vary
with the stage and the variant of the disease. The white matter is
very commonly involved, sometimes primarily (a true leukody-
strophy) or sometimes by Wallerian degeneration resulting from
damage to cortical neurons. Involvement of the deep cerebral
nuclei (the thalami and basal ganglia) may be the result of gray
matter or white matter injury (about half of the basal nuclei is
composed of white matter). In the end stage, all disorders look
much the same, with atrophy of the cerebral cortex, reduced
volume of gliotic or demyelinated white matter, and shrunken
basal ganglia. Therefore, it is important to image the patient
early in the course of the disease, when the pattern of injury
may be more characteristic.
Aceptado: 14.09.06.
Professor of Radiology, Neurology, Pediatrics, and Neurosurgery Neurora-
diology. UCSF. San Francisco, CA, Estados Unidos.
Correspondencia: A. James Barkovich, M.D. Professor of Radiology, Neu-
rology, Pediatrics, and Neurosurgery Neuroradiology. Rm L-371. UCSF.
505 Parnassus Ave. San Francisco, CA 94143-0628, USA. E-mail: jim.
barkovich@radiology.ucsf.edu
© 2006, REVISTA DE NEUROLOGÍA
TO METABOLIC DISEASE
The use of a systematic approach based on the pattern of brain
involvement is useful in the analysis of neurometabolic disor-
ders by imaging [1-3]. Such an approach has been described by
van der Knaap and Valk; they initially created a database and
subsequently developed a computer-assisted pattern recognition
program to help in the imaging-guided identification of the dis-
orders [3,4]. The information entered into the database detailed
precisely what parts of the brain where involved in the disease
process and the specific type of involvement in each affected
area. In each cerebral lobe, they recorded the presence or
absence of involvement of the cerebral cortex, subcortical U
fibers (arcuate fibers), deep white matter, and periventricular
white matter. Next, they assessed the presence or absence of
involvement of multiple smaller structures: the internal capsule
(anterior and posterior limb), external capsule, caudate nucleus,
putamen, globus pallidus, thalamus, corpus callosum (rostrum,
genu, body, and splenium), cerebellar cortex, cerebellar white
matter, cerebellar dentate nucleus, cerebellar peduncles, hilus of
the dentate nucleus, midbrain, pons, and medulla. Subsequently,
they analyzed which areas were predominantly involved (i.e.,
supratentorial or infratentorial, cerebral lobe, subcortical or deep
or periventricular white matter), degree of left-right symmetry,
extension of the lesions (small or large, isolated or confluent),
appearance (swelling, atrophy, cystic degeneration), signal
intensity, homogeneity, and demarcation from surrounding brain
(sharp, vague, or mixed). Finally, they listed extra characteris-
tics: calcium, hemorrhage, contrast enhancement, enlargement
of subarachnoid spaces, presence of absence of cerebellar atro-
phy, and myelination (normal, delayed, absent). This data was
carefully recorded for nearly 1,500 patients with known diag-
noses; it was subsequently entered into their database.
The availability of this database greatly facilitates their abili-
ty to establish diagnoses. When a new patient is encountered, the
magnetic resonance imaging (MRI) is analyzed, the MRI data is
entered into their database, and it is compared with the data
already entered for known metabolic disorders. With some
patients, an exact (or nearly exact) match is achieved and the

REV NEUROL 2006; 43 (Supl 1): S5-S16 S5

A.J. BARKOVICH
diagnosis is established. If the match is not precise, the patient
may be put into a group in which several diagnoses remain pos-
sible based upon clinical and imaging criteria. By entering the
patient into a smaller group of potential diagnoses, the amount
of biochemical and genetic testing that must be performed is
substantially reduced, saving time and money. Sometimes, the
data is clearly different from all known syndromes, but matches
one or more previously entered patients with unknown diag-
noses. This sort of observation is sometimes the first step in the
identification of a new syndrome. Examples of disorders identi-
fied in this manner include hypomyelination with atrophy of the
basal ganglia and cerebellum [5], megalencephaly with leukoen-
cephalopathy and cysts [6], and cerebellar ataxia with cerebral
hypomyelination (vanishing white matter disease) [7].
A similarly systematic, but less sophisticated, approach to
the analysis of metabolic disorders, based upon the pattern of
brain involvement on neuroimaging studies, is presented in this
manuscript. This analysis will be primarily based upon the ear-
ly appearance of disorders on MRI. Disorders with similar MRI
patterns can sometimes be differentiated on the basis of their
microstructural characteristics, which can be analyzed by diffu-
sion weighted imaging, or by in vivo biochemical analysis,
which can be grossly performed by proton magnetic resonance
spectroscopy. This approach does not replace a thorough bio-
chemical and genetic work-up, but it provides an excellent ini-
tial screening process; proper analysis of the MR studies can
reduce the extent of the biochemical and genetic analyses.
First analysis: white matter versus gray matter
The first important determination to be made is whether the dis-
ease involves primarily gray matter, primarily white matter, or
both gray and white matter. If the gray matter is primarily
involved, both the cortex and deep nuclei should be scrutinized to
determine whether involvement is primarily cortical, primarily
deep nuclear, or both. Sometimes, disorders that primarily affect
cortical gray matter will show cortical swelling with effaced sulci
early in the course of the disease; more commonly, cortical thin-
ning is found, with prominent cortical sulci [8-11]. In later stages,
cortical thinning is the rule in all such disorders. The cerebral
white matter will often have an abnormal appearance in these
patients, as Wallerian degeneration of axons causes diminished
white matter volume and mild to moderate hyperintensity on
FLAIR and T2 weighted images. This white matter appearance
can often be differentiated from that of primary white matter dis-
orders if the study is performed early in the course of the disease,
as primarily affected white matter will often be edematous and,
therefore, brighter and more voluminous (causing compressed,
smaller sulci) than the white matter that has undergone secondary
(Wallerian) degeneration. In the acute phase, disorders primarily
affecting deep gray matter will typically show edema with FLAIR
hyperintensity and prolonged T1 and T2 relaxation times on MRI
in the involved structures (the major exception to this is pan-
tothenate kinase associated neuropathy, in which iron accumula-
tion in the globus pallidus results in T2 hypointensity [12,13]) and
may show loss of volume with gliosis (resulting in slit-like con-
figuration with T2 hyperintensity) in more chronic stages.
Disorders primarily affecting white matter cause marked sig-
nal abnormality before any volume loss is apparent. Some white
matter disorders cause spongiform changes, result in intra-
myelinic edema, or have an inflammatory component in the ear-
ly stages; these conditions cause edema, with accompanying
S6
Table I. Metabolic disorders causing T2 or FLAIR hyperintensity of the
corpus striatum.
Leigh syndrome (includes pyruvate dehydrogenase deficiency,
respiratory complex I and complex II disorders)
Wilson disease
Glutaric aciduria type I
Juvenile Huntington disease
Molybdenum co-factor deficiency
Propionic acidemia
mass effect, upon adjacent structures [14]. Alternatively, many
white matter disorders, such as adrenoleukodystrophy and fibri-
noid leukodystrophy (Alexander disease), start locally and
advance over time to involve adjacent areas. Neither of these
appearances is seen in the white matter of gray matter disorders.
White matter diseases can result in devastation of the involved
areas, with necrosis and cavitation of the affected regions and
subsequent ex vacuo dilation of the ventricles, whereas the
abnormal white matter in gray matter disorders appears less
severely damaged. Finally, the clinical presentation of patients
with cortical gray matter disorders (seizures, dementia in early
stages) differs from that of deep gray matter disorders (chorea,
athetosis, dystonia) and both differ from the presentation of
white matter disorders (spasticity, hyperreflexia, ataxia). Although
such clinical criteria are sometimes misleading, they can some-
times be helpful to get started on the right track.
MRI in gray matter disorders
As discussed above, gray matter disorders need to be further
analyzed in order to determine whether the disorder primarily
involves the cerebral cortex or the deep gray matter nuclei. This
is most easily determined by examining the deep gray nuclei to
look for abnormal signal intensity on T2 weighted or FLAIR
images. Some disorders will manifest cortical edema in the
acute phase, appearing as T2 or FLAIR hyperintensity; if imag-
ing is not performed until later stages of the disease, cortical
involvement will be manifested as sulcal enlargement, cortical
thinning, and, sometimes, abnormal signal intensity.
Deep gray matter involvement
When only deep gray matter is involved, it is important to iden-
tify the specific structures that are affected and their signal
intensities. Involvement of the striatum (caudate and putamen)
is typically seen in mitochondrial disorders –primarily Leigh
syndrome, mitochondrial encephalopathy with lactic acidosis and
stroke-like symptoms (MELAS), and the glutaric acidurias–,
propionic acidemia, Wilson disease, juvenile Huntington dis-
ease, molybdenum cofactor deficiency, asphyxia, and hypo-
glycemia (Table I). Associated white matter or cortical injury is
frequently present in many of these disorders. If isolated globus
pallidus involvement shows T2 prolongation, succinate semi-
aldehyde dehydrogenase deficiency, methylmalonic acidemia,
guanidinoacetate methyltransferase deficiency (a creatine syn-
thesis disorder), isovaleric acidemia, pyruvate dehydrogenase
deficiency –due to mutation of the dihydrolipoamide acetyl-
transferase (E2) component–, carbon monoxide poisoning, or
the chronic phase of kernicterus should be considered (Table II)
REV NEUROL 2006; 43 (Supl 1): S5-S16

b
Figure 1. L-2-hydroxyglutaric aciduria: a) Axial SE 2500/30 image shows
bilateral hyperintensity (arrows) of the globus pallidus. Note that the peri-
ventricular white matter around the frontal horns and trigones is myelina-
ted, but the subcortical white matter is not; b) Coronal FSE 3000/100 ima-
ge better shows the hyperintensity of the subcortical white matter with
sparing of the periventricular white matter.
[2,15]. If T2 or FLAIR hyperintensity of the globus pallidus is
seen in association with subcortical white matter demyelination,
sparing of the periventricular white matter, and involvement
of the cerebellar dentate nuclei, L-2-hydroxyglutaric aciduria
(Fig. 1) [16] and Kearns-Sayre syndrome [17] should be consid-
ered. When the MRI shows atrophy of the dorsal brain stem and
cerebellar dentate nuclei, one should consider dentatorubral and
pallidoluysian atrophy [18,19]. If T1 hyperintensity of the
globus pallidus is seen associated with normal T2 signal, and
the patient is not receiving hyperalimentation, consider chronic
hepatic disease [20]. If both T1 and T2 hyperintensity are seen in
REV NEUROL 2006; 43 (Supl 1): S5-S16
CONFERENCIA MAGISTRAL
Figure 2. Citrullinemia: axial SE 2500/80 image shows abnormal hyperin-
tensity of the globus pallidus (small arrows) and the insular cortex (large
arrows).
Table II. Metabolic disorders causing T2 or FLAIR hyperintensity of the globus
pallidus.
Methylmalonic acidemia
Succinic semialdehyde dehydrogenase deficiency
Urea cycle disorders
Guanidinoacetate methyltransferase deficiency
Pyruvate dehydrogenase (E2) deficiency
Systemic lupus erythematosis
Hemolytic-uremic syndrome
Bilirubin toxicity
Isovaleric acidemia
Carbon monoxide toxicity
Cyanide toxicity
a neonate or young infant, consider acute hyperbilirubinemia
of infancy [21-23], systemic lupus erythematosis [24,25], and
hemolytic-uremic syndrome [26,27]; in the presence of edema
involving the external and extreme capsules and the claustrum,
hemolytic-uremic syndrome is the most likely [27]. If the
globus pallidus, the insula and perirolandic cortex are all hyper-
intense on T2 weighted or FLAIR images (Fig. 2), the diagnosis
of a urea cycle disorder should be suggested [28,29]. If the
involvement of the globus pallidus is manifested on MR as T2
shortening or T2 shortening with central T2 prolongation (Fig.
3), the diagnosis of pantothenate kinase associated neuropathy
S7
A.J. BARKOVICH

Figure 3. Pantothenate kinase associated neuropathy (formerly Hallervor-

den-Spatz disease): coronal FSE 3000/120 image shows marked hypo-
intensity of the globus pallidus with central hyperintensity; this is the so-
called ‘eye of the tiger’ appearance.

(formerly called Hallervorden-Spatz disease) can be made with

some confidence [12,13].
Figure 4. Pelizaeus-Merzbacher disease: axial SE 2500/80 image shows
minimal myelination restricted to the internal capsule; this is the pattern
Cerebral cortical involvement of an infant aged maybe 6 months, but this child is 3 years old! Pelizaeus-
Merzbacher disease results in a nearly complete lack of myelination, fol-
If the pattern of the imaging study indicates that the metabolic lowed by slow loss of myelin.
disorder is primarily one of cortical involvement (cortical thin-
ning with enlarged cortical sulci), consideration should be given
to such disorders as the neuronal ceroid lipofuscinoses [8,9,30],
the mucolipidoses [31], glycogen storage diseases [32], or GM1
gangliosidosis [2,15].

White matter disorders

White matter disorders can be divided into [1] those disorders in
which white matter never myelinates completely (hypomyelina-
tion) and those disorders in which myelin forms and is subse-
quently destroyed (demyelination). In the latter group, it is
important to determine whether myelin destruction is in the
periventricular, deep or subcortical white matter and what part of
the white matter (specific gyrus or lobe) is affected.

Hypomyelination
The pattern of a lack of myelination, or hypomyelination, is
seen in very few disorders. This group of disorders is known as
the hypomyelinating leukoencephalopathies; several of these
diseases are listed below. The hypomyelination pattern is most
commonly found in Pelizaeus-Merzbacher disease, a disorder
that affects the PLP1 gene that codes for the production of pro-
teolipid protein, one of the major structural proteins of myelin
[33,34]. The appearance of the brain in this disorder is that of a
normal, much more immature brain. For example, the MRI of
a 5-year-old child with this disorder might be mistaken for that
of a 5-month-old infant (Fig. 4). Similar appearances can be
seen in patients with Pelizaeus-Merzbacher-like disease (identi-
cal to PMD other than the causative genes [35,36]) leukodystro- Figure 5. Alexander disease: axial SE 2500/80 image shows abnormal
phies with trichothiodystrophy and photosensitivity [37,38], in hyperintensity of the frontal white matter including the subcortical arcu-
ate fibers. The heads of the caudate nuclei and the anterior putamina are
patients with Tay syndrome [39], in patients with the 18q-syn- also affected, along with the external and extreme capsules. This pattern
drome (deletion of a large portion of the long arm of chromo- in a macrocephalic patient is specific for Alexander disease.

S8 REV NEUROL 2006; 43 (Supl 1): S5-S16


Figure 6. X-linked adrenoleukodystrophy: axial SE 2500/90 image shows
marked hyperintensity of the callosal splenium and the parietal white
matter, extending anteriorly into the internal and external capsules. Note
the sparing of the subcortical arcuate fibers (arrows).
Table III. Metabolic disorders causing early involvement of subcortical
white matter.
Alexander disease
Kearns-Sayre syndrome
Megalencephalic leukoencephalopathy with subcortical cysts
Galactosemia
Mitochondrial disorders
some 18) [40,41], and in patients with Salla disease (a disorder
of sialic acid transport) [42,43]. If there is a question about the
diagnosis of Pelizaeus-Merzbacher disease, proton MR spec-
troscopy may help, as it shows an elevated NAA peak [44].
Loss of myelin (demyelination and dysmyelination)
When myelin develops but is subsequently damaged, it is not
clear from imaging whether the myelin that was initially formed
was normal or abnormal. The former condition (destruction of
normal myelin) is sometimes referred to as demyelination to
differentiate if from breakdown of abnormal myelin, sometimes
called dysmyelination. By the use of diffusion tensor imaging,
it may eventually be possible to differentiate these two entities
[45], but the differentiation cannot currently be made.
When such myelin breakdown and loss occurs, the white
matter becomes more hypointense on T1 weighted images and
more hyperintense on T2 weighted images. In these situations,
the brain should be analyzed to determine whether the region
primarily affected is the periventricular white matter, the deep
REV NEUROL 2006; 43 (Supl 1): S5-S16
CONFERENCIA MAGISTRAL
Figure 7. X-linked adrenoleukodystrophy, very early stage. Brain involve-
ment of this disorder typically starts in the corpus callosum (arrows) as
seen in this patient.
white matter or the subcortical white matter. Table III lists some
disorders with early involvement of the subcortical white mat-
ter. If the subcortical white matter is involved, it should be care-
fully analyzed to see if the subcortical U fibers (the arcuate
fibers at the cortical-white matter junction) are affected. If
so, an attempt should be made to find out whether the patient
has macrocephaly. Bilateral, symmetrical, frontal white matter
involvement involving the U fibers in a macrocephalic patient is
quite specific for Alexander disease, particularly if it extends
posteriorly to involve the caudate heads and putamen (Fig. 5)
[46]. Bilateral, diffuse and symmetric, peripheral white matter
involvement without involvement of the deep or periventricular
white matter and without macrocephaly should raise suspicion
for organic acidurias or early Kearns-Sayre syndrome [2,17].
Diffuse white matter abnormality involving the subcortical U
fibers and associated with subcortical cysts in a macrocephalic
infant or young child suggests megalencephalic leukoen-
cephalopathy with subcortical cysts (MLC) [6,47,48].
It is uncommon to see early myelin injury restricted to the
deep white matter without periventricular white matter involve-
ment. Table IV lists some disorders with early involvement of
the deep and periventricular white matter. If early involvement
affects both periventricular and deep white matter, the thalami
should be specifically analyzed. High attenuation on computer-
ized tomography (CT) or short T1 (hyperintensity) or T2 (hypo-
intensity) on MR bilaterally in the thalami strongly suggests
globoid cell leukodystrophy (Krabbe disease) or GM2 gan-
gliosidosis [2,49-51]. If the thalami are normal, the brain stem
should be evaluated for involvement of specific tracts, particu-
larly the corticospinal tracts. If specific tracts (the corticospinal
tracts, in particular) are involved, peroxisomal disorders such
S9
A.J. BARKOVICH
as X-linked adrenoleukodystrophy (Fig. 6) should be strongly
considered [52] –of note, the earliest radiologic sign of adreno-
leukodystrophy is T2 hyperintensity in the central part of the cal-
losal splenium, not in the hemispheric white matter; therefore, so
if adrenoleukodystrophy is suspected clinically or if there is a
family history, the corpus callosum should be intensely scruti-
nized (Fig. 7)–. If not, consideration should be given to meta-
chromatic leukodystrophy, cerebellar ataxia with cerebral hypo-
myelination (vanishing white matter disease), phenylketonuria,
Lowe syndrome (oculocerebrorenal syndrome), mucolipidosis
type IV, merosin deficient congenital muscular dystrophies, and,
in the proper clinical setting, damage from radiation or chemo-
therapy [2,15]. Among these, vanishing white matter disease
should be considered if the clinical history is one of periodic
acute worsening after trauma or infection and if areas of cystic
degeneration (seen as hypointense on FLAIR images) develop in
the hemispheric white matter [7,53]. Lowe syndrome should be
suspected in many small cysts are seen in the affected white mat-
ter, particularly in the periventricular region [54,55]. If the dorsal
brain stem, internal capsules, cerebral peduncles, and cerebellar
white matter are affected in a newborn, maple syrup urine dis-
ease should be considered; this diagnosis can be confirmed by
the finding of reduced diffusion on diffusion weighted images
and the presence of a broad peak from branched chain ketoacids
at 0.9 ppm on proton MR spectroscopy –for details, see the later
section of uses of MR spectroscopy [56,57]–.
Few metabolic disorders affect the periventricular white mat-
ter while sparing the deep white matter. Observation of this pat-
tern should raise the possibility of prenatal or (in prematurely
born infants) neonatal or early postnatal white matter injury.
Other causes of periventricular white matter injury include infec-
tion (specifically ventriculitis), small periventricular hemorrhag-
ic venous infarctions, and (rarely) demyelinating disease [58].
Some metabolic disorders have nonspecific white matter
patterns; this group includes disorders with involvement of
both superficial and deep or periventricular white matter,
those with unilateral white matter involvement, those with dif-
fuse white matter involvement, and those with bilateral asym-
metric white matter involvement. Among the disorders includ-
ed under this category are the collagen vascular diseases, such
as systemic lupus erythematosis (which tend to involve the
white matter bilaterally and asymmetrically [59,60]), congen-
ital muscular dystrophies with neurologic impairment [61-64],
and demyelinating diseases, such as multiple sclerosis and
acute disseminated encephalomyelitis (which affect the white
matter bilaterally and asymmetrically and may affect deep
cerebral nuclei) [65,66]. End stage white matter disease of any
cause results in diffuse (superficial and deep), bilateral white
matter damage that is completely nonspecific.
Metabolic disorders affecting both gray and white matter
Disorders involving both gray and white matter are first separat-
ed by the type of gray matter involvement: those involving only
the cerebral cortex and those involving deep gray matter (with
or without cortical involvement).
Cerebral cortex involvement
Those disorders involving only cortical gray matter can be sub-
divided depending on involvement of long bones and the spinal
column. If the long bones are normal, the cortex should be ana-
lyzed for malformations of cortical development (MCD). If an
S10
Table IV. Metabolic disorders causing early involvement of periventricular
and deep white matter with sparing of subcortical white matter.
X-linked adrenoleukodystrophy
Krabbe disease (globoid cell leukodystrophy)
Metachromatic leukodystrophy
GM2 gangliosidoses
Childhood ataxia with CNS hypomyelination
(vanishing white matter disease)
Lowe syndrome (oculocerebrorenal syndrome)
Mucolipidosis type IV
Merosin deficient congenital muscular dystrophy
Damage from radiation or chemotherapy
MCD is present in addition to a lack of myelination, the differ-
ential diagnosis includes the generalized peroxisomal disorders
[67,68], congenital cytomegalovirus disease [69-71], and con-
genital muscular dystrophies with cerebral involvement. Con-
genital muscular dystrophies will typically have pontine hypo-
plasia and cerebellar dysplasia, as well [63] (Fig. 8)–. If no cor-
tical dysplasia is present, differential considerations include Al-
pers disease and Menkes disease, both of which cause consider-
able cerebral cortical destruction [17]. If the bones are abnor-
mal, the differential diagnoses include primarily storage dis-
eases, such as the mucopolysaccharidoses and lipid storage
disorders [72-75]. A description of the bony findings of these
disorders is beyond the scope of this manuscript.
Cerebral nuclear involvement
If deep gray matter is involved, the first task is to determine pre-
cisely which nuclei are affected (Table I). If the thalami are
involved, differential considerations include Krabbe disease and
the GM1 and GM2 gangliosidoses; the thalami in these disor-
ders display high attenuation on CT and short T1 and T2 relax-
ations times (hyperintense on T1 weighted images and hypo-
intense on T2 weighted images) on MR. Krabbe disease is dis-
tinguished from the others by the presence of abnormal T2
hyperintensity along the corticospinal tracts and in the cere-
bellar dentate nuclei (Fig. 9). Another disorder with thalamic
involvement is profound neonatal hypoxic-ischemic injury,
which typically involves the ventrolateral thalami along with
posterior putamina and perirolandic cortices [76-78]; these
patients typically have a characteristic history of perinatal dis-
tress and neonatal encephalopathy that simplifies the diagnosis
[79-81]. Another consideration when thalamic involvement is
identified is autosomal dominant acute necrotizing encephalitis
[82,83], particularly if T2 hyperintensity is lso seen in the dor-
sal brain stem. Thalami may also be affected in mitochondrial
disorders, Wilson disease, and Canavan disease; typically other
deep gray matter nuclei will be affected, as well (putamina in
mitochondrial disorders and Wilson disease, globus pallidus in
Canavan disease).
Globus pallidus involvement in association with diffuse
white matter disease including the subcortical, deep, and periven-
tricular regions suggests a diagnosis of Canavan disease (Fig. 10)
[84]. Association of globus pallidus involvement with affected
REV NEUROL 2006; 43 (Supl 1): S5-S16
 CONFERENCIA MAGISTRAL

a b the affected regions, and a

characteristic peak at 0.9
ppm on proton MR spec-
troscopy (see below) [56,
57,85]. Carbon monoxide
and cyanide toxicity typical-
ly involve the cerebral cortex
and striatum and, sometimes,
the cerebellum [86-89]. Ta-
a ble V lists some disorders
with involvement of both
white matter and the globus
pallidus.
White matter disease
associated with striatal (pu-
tamen and caudate) involve-
ment suggests Leigh syn-
drome, MELAS, propionic
acidemia, glutaric aciduria
type 1 (glutaryl CoA de-
hydrogenase deficiency),
Figure 8. Congenital muscular dystrophy with brain involvement: a) Axial SE 2500/80 image shows abnormal
hyperintensity in the frontal and parietal white matter. The cortical gyral pattern in the frontal lobes (arrows) is dys- molybdenum cofactor defi-
plastic; b) Axial image through the cerebellum shows dysplastic cortex with multiple subcortical cysts (arrows). ciency, isolated sulfite oxi-
dase deficiency, hypomyeli-
a b nation with atrophy of the
basal ganglia and cerebel-
lum, toxic exposure, later
infantile or childhood pro-
found hypoxic-ischemic in-
jury, or childhood hypo-
glycemia (Table VI). Regions
of involvement in Leigh syn-
drome vary with the underly-
ing molecular cause of the
disorder, although no consis-
tently reproducible geno-
type-phenotype associations
have been identified. In
MELAS, cortical lesions are
more common than basal
ganglia lesions, and are usu-
ally present by the time basal
ganglia involvement is iden-
tified. Two imaging features
Figure 9. Globoid cell leukodystrophy (Krabbe disease): a) Axial SE 3000/120 in this infant shows abnormal hyper- can help to differentiate the
intensity (arrows) along the corticospinal tracts in the posterior limbs of the internal capsules; b) Axial image in the cortical lesions of MELAS
posterior fossa shows abnormal hyperintensity in the hila of the cerebellar dentate nuclei (white arrows) and in
the medullary pyramids (black arrowheads).
from those of ischemic in-
farcts: the locations do not
correspond to vascular terri-
subcortical white matter but sparing of periventricular white mat- tories; and premature senescent calcifications in the globus pal-
ter suggests a later phase of Kearns-Sayre syndrome (globus pal- lidus are typically identified on CT [90]. Glutaric aciduria type 1
lidus are spared in early phases) or L-2-hydroxyglutaric aciduria is typically associated with enlarged subarachnoid spaces, partic-
(Fig. 1); the latter will often show involvement of the cerebellar ularly in the anterior sylvian fissures, and central white matter T2
dentate nuclei [16]. Diffuse white matter involvement, with spar- hyperintensity. Isolated sulfite oxidase deficiency is rapidly pro-
ing the subcortical white matter during the early stages of the dis- gressive and causes multicystic encephalomalacia of the cerebral
ease, suggests methylmalonic acidemia, maple syrup urine dis- white matter to develop within the first weeks after birth [91].
ease, carbon monoxide toxicity, or cyanide toxicity. MRI of Cockayne disease will show calcification of the striatum, as well
maple syrup urine disease during the acute neonatal phase of the as characteristic facies and other aspects of the syndrome.
disease typically shows involvement of the corticospinal tracts in Proper analysis of the MR scans using this pattern system
the centrum semiovale, internal capsules, cerebral peduncles, can facilitate the work-up of patients with inborn errors of
dorsal pons, and cerebellar white matter, reduced diffusivity in metabolism.

REV NEUROL 2006; 43 (Supl 1): S5-S16 S11

A.J. BARKOVICH

PROTON MR
a b
SPECTROSCOPY
IN METABOLIC
DISORDERS
At the University of California
at San Francisco, we have
found that proton MR spec-
troscopy (MRS) can be an
extremely useful adjunct when
a patient is being imaged for
clinical suspicion of an inborn
error of metabolism. In some
cases, the spectroscopy serves
mainly to help verify a diagno-
sis already suspected by clinical
suspicion or by the analysis of
pattern on MRI. For example,
the pattern of brain involvement
in neonatal maple syrup urine Figure 10. Canavan disease in a 12-month-old infant: a) Axial SE 2500/80 image shows diffuse white matter
disease is very specific. Edema hyperintensity, including the subcortical, deep, and periventricular white matter. Abnormal hyperintensity is also
(FLAIR and T2 hyperintensity) seen in the globus pallidus (g); b) Calculated diffusivity map shows abnormally reduced diffusivity in the white
matter and globus pallidus.
is present in the dorsal brain
stem, the cerebellar white matter, the internal capsules and cere- Table V. Metabolic disorders with globus pallidus and white matter involve-
bral corticospinal tracts, and the globus pallidus (Fig. 11) [85]. ment.
Proton MRS helps to confirm the findings when it shows a charac- Canavan disease
teristic broad peak centered at 0.9 ppm (Fig. 11d) that represents
branched chain amino acids and ketoacids [57]. Similarly, Cana- Methylmalonic acidemia
van disease has a characteristic pattern of diffuse white matter Kearns-Sayre syndrome
edema, involving periventricular, deep, and subcortical white mat-
ter, in addition to bilateral globus pallidus and, sometimes, thala- L-2-hydroxyglutaric aciduria
mic involvement [84,92]. Proton MRS helps to confirm this sus- Maple syrup urine disease
pected diagnosis by revealing a very large NAA peak at 2.01 ppm.
In other disorders, MRS is essential for making the diagnosis. Carbon monoxide poisoning
The clinical features and MRI imaging pattern of patients with Cyanide toxicity
nonketotic hyperglycinemia, for example, is nonspecific. Proton
MRS, however, shows an elevated glycine peak at 3.55 ppm on
both long and short echo spectra. Using a short echo time, one Table VI. Metabolic disorders with striatal (caudate and putaminal) and
might confuse the peak with that of myo-Inositol, but if the peak white matter involvement.
persists at long echo times such as 270 or 288 ms, it represents
Leigh syndrome
glycine and strongly suggests the diagnosis. Many other disor-
ders have spectroscopic findings that are nearly specific in mak- Mitochondrial encephalopathy with lactic acidosis
ing the diagnosis: a single peak for pyruvate (2.36 ppm) and a and stroke-like symptoms (MELAS)
doublet for lactate (1.33 ppm) in pyruvate dehydrogenase defi- Other mitochondrial leukoencephalopathies
ciency; a singlet for succinate at 2.40 ppm in succinate dehydro-
genase deficiency; a singlet at 1.3 ppm in Sjögren-Larsson syn- Propionic acidemia
drome; two singlets (at 3.67 and 3.74 ppm) in untreated galac- Glutaric acidemia type I
tosemia; and a singlet at 7.36 ppm in phenylketonuria [15]. In
creatine deficiency syndromes, MRS is the most important tech- Isolated sulfite oxidase deficiency
nique in making the diagnosis. Creatine deficiency syndromes Late infantile/childhood profound hypoxic-ischemic injury
include disorders due to defects in synthesis –guanidinoacetate
methyltransferase (GAMT) deficiency [93,94] and arginine:gly- Childhood hypoglycemia
cine amidinotransferase (AGAT) deficiency [95]– and defects in
the transport of creatine across the blood-brain barrier (X-linked
creatine transporter defect) [96,97]. Patients with GAMT defi- Proton MRS can also be important in assessing response to
ciency often have abnormal hyperintensity of the globus pallidus therapy of metabolic disorders. Response to oral creatine sup-
on FLAIR and T2 weighted images, but patients with AGAT defi- plementation by following brain creatine levels with MRS has
ciency and creatine transporter defects can only be detected by already been discussed in the previous paragraph. Proton MRS
appreciated the low or absent creatine peak on proton MRS of the also allows monitoring of the brain phenylalanine levels in
brain (Fig. 12). As oral creatine supplementation leads to striking patients with phenylketonuria (PKU), although it requires the
improvements in neurological exam and intellectual development use of a very short echo time (20 ms or less), because of the
[98], detection of this disorder with MRS is extremely important. short T2 relaxation time of the phenyl protons, and a large field

S12 REV NEUROL 2006; 43 (Supl 1): S5-S16

 CONFERENCIA MAGISTRAL

a b develop the serious form of ALD.

Waiting until the child has rapidly
advancing or severe cerebral
involvement is not recommended,
either, as the procedure may lead
to acceleration of the disease
[101-103]. Proton MRS provides
important information in this
regard, as it shows abnormalities
before the MR imaging study
becomes abnormal in some chil-
dren [104] and in normal appear-
ing white matter peripheral to the
regions of T2 prolongation and
enhancement; this area of abnor-
mal MRS may represent zone C.
Indeed, it is suggested that reduc-
tion in the NAA/choline ratio to
below 5.0 is predictive of disease
c d progression within the next 2 to
3 years [101]. Therefore, proton
MRS should be performed in all
individuals with ALD, particular-
ly those with stable MR imaging
exams and in siblings (of patients
with ALD) who are known to
have the ALD mutation but are
as yet asymptomatic. A list of
disorders in which proton spec-
troscopy is particularly useful is
provided in table VII.

DIFFUSION WEIGHTED
IMAGING IN METABOLIC
DISORDERS
Figure 11. Maple syrup urine disease in a young infant: utility of proton MR spectroscopy and diffusion
weighted imaging: a,b) Axial SE 3000/120 images show abnormal hyperintensity in the globus pallidus and The literature about diffusion
posterior limbs of the internal capsules (black arrows) in a, and along the corticospinal tracts in the cerebral weighted imaging in metabolic
hemispheres (black arrows) in b. c) Axial calculated diffusivity image at the same level as (b) shows reduced
diffusion (hypointensity) in the corticospinal tracts in the cerebral hemispheres (black arrows); d) Proton MR disorders is contradictory and
spectroscopy (TE: 26 ms) shows a large peak at 0.9 ppm (arrow) that represents branched chain ketoacids. rather confusing because the
This peak is characteristic of maple syrup urine disease. The spectrum is otherwise normal for age. results depend upon the acuity
of the disease at the time of the
of view because cerebral phenylalanine level is typically very study. During an acute exacerbation of metachromatic leukody-
low (less than 1 mM/L). The resonances of the phenyl protons strophy or globoid cell leukodystrophy, for example, diffusion
are visualized in a single peak at 7.36 ppm; this peak has large weighted imaging shows reduced diffusion in the affected
enough amplitude that it can be quantified. Direct analysis of white matter, while in later, more quiescent phases, diffusivity
the brain is important because phenylalanine levels in the brain in increased [105]. This likely results from the fact that acute
do not correlate closely with those in the blood [99,100]. cellular injury, whether from toxins, ischemia, trauma, or infec-
In some disorders, proton MRS can aid in making a diagnosis tion, results in the slowing of the motion of water molecules
and instituting therapy before clinical signs and symptoms devel- (reduced diffusivity). Later, after axons, cell bodies, or myelin
op. An excellent example of this is X-linked adrenoleukodystro- have been damaged, there are fewer barriers to water motion
phy (ALD). In this disorder, the central issue is whether a boy car- and diffusivity is increased. Other disorders cause a specific
rying the biochemical/genetic defect should undergo hematopoiet- type of myelinopathy (myelin vacuolization [105]) that results
ic stem cell transplantation, a form of treatment reserved for the in a marked decrease in diffusivity. An example of this type of
rapidly progressive cerebral disease (an not for the later-onset disorder is neonatal onset maple syrup urine disease. During
variant known as adrenomyeloneuropathy). If bone marrow trans- the acute phase of the disease, myelin (mature and immature)
plantation is performed early in the course of the rapidly progres- becomes edematous and the spirals become slightly separated;
sive disease, it may prevent further deterioration or even lead to during this phase of the disease, all myelinated parts of the
improvement [101-103]. However, the therapy is not recommend- brain show up to an 80% reduction in diffusivity, resulting in
ed for asymptomatic children, as the morbidity and mortality asso- markedly increased signal intensity on diffusion weighted images
ciated with transplantation is high and more than 50% will never and markedly decreased signal intensity on diffusivity images

REV NEUROL 2006; 43 (Supl 1): S5-S16 S13

A.J. BARKOVICH
Table VII. Utility of proton spectroscopy in metabolic disorders.
Identification of a specific peak
Canavan disease
Maple syrup urine disease
Nonketotic hyperglycinemia
Succinate dehydrogenase deficiency
Sjögren-Larsson syndrome
Phenylketonuria
Pyruvate dehydrogenase deficiency
Absence of a normal peak
Creatine deficiency disorders
Identification of a nonspecific peak
Mitochondrial disorders (lactate)
Alexander disease (myo-Inositol)
–also called apparent diffusion coefficient images (Fig. 11c)
[56]–. Several other disorders show markedly reduced diffu-
sion in the acute state, including nonketotic hyperglycinemia,
Kearns-Sayre syndrome, and Canavan disease (Fig. 10),
although the precise mechanism by which the diffusivity de-
creases is not known. The precise timing of the changes in dif-
fusivity with respect to onset of the disease and its histologic
progression is not know, but it is well established that, after
the acute stage of encephalopathy, myelin will break down
and diffusivity will increase.
REFERENCES
1. Van der Knaap M, Breiter S, Naidu S, Hart A, Valk J. Defining and cat-
egorizing leukoencephalopathies of unknown origin: MR imaging
approach. Radiology 1999; 213: 121-33.
2. Barkovich AJ. Toxic and metabolic brain disorders. In Barkovich AJ,
ed. Pediatric neuroimaging. 4 ed. Philadelphia: Lippincott Williams &
Wilkins; 2005. p. 76-189.
3. Van der Knaap MS, Valk J. Pattern recognition in white matter disor-
ders. In van der Knaap MS, Valk J, eds. Magnetic resonance of myeli-
nation and myelin disorders. 3 ed. Berlin: Springer; 2005. p. 881-904.
4. Van der Knaap MS, Valk J, De Neeling N, Nauta JJ. Pattern recogni-
tion in magnetic resonance imaging of white matter disorders in chil-
dren and young adults. Neuroradiology 1991; 33: 478-93.
5. Van der Knaap M, Naidu S, Pouwels P, Bonacita S, Van Coster R,
Lagae L, et al. New syndrome characterized by hypomyelination with
atrophy of the basal ganglia and cerebellum. AJNR Am J Neuroradiol
002; 23: 1466-74.
6. Van der Knaap MS, Barth PG, Stroink H, Van Nieuwenhuizen O, Arts
WFM, Hoogenraad F, et al. Leukoencephalopathy with swelling and a
discrepantly mild clinical course in eight children. Ann Neurol 1995;
37: 324-34.
7. Van der Knaap MS, Barth PG, Gabreëls FJM, Franzoni E, Begeer JH,
Stroink H, et al. A new leukoencephalopathy with vanishing white
matter. Neurology 1997; 48: 845-55.
8. Autti T, Raininko R, Vanhanen SL, Santavuori P. Magnetic resonance
techniques in neuronal ceroid lipofuscinoses and some other lysosomal
diseases affecting the brain. Curr Opin Neurol 1997; 10: 519-24.
9. Autti T, Raininko R, Vanhanen SL, Santavuori P. MRI of neuronal
ceroid lipofuscinosis. I. Cranial MRI of 30 patients with juvenile neu-
ronal ceroid lipofuscinosis. Neuroradiology 1996; 38: 476-82.
10. Vanhanen SL, Raininko R, Autti T, Santavuori P. MRI evaluation of the
brain in infantile neuronal ceroid-lipofuscinoses. II. MRI findings in
21 patients. J Child Neurol 1995; 10: 444-50.
S14
Figure 12. Creatine deficiency due to arginine:glycine amidinotransferase
(AGAT) mutation: proton MR spectrum from the patient’s basal ganglia
region shows only two peaks, choline (Ch) and N-acetylaspartate (NAA).
The creatine peak, normally at 3.03 ppm (see figure 11d) is absent.
CONCLUSION
The analysis and diagnosis of metabolic disorders is extremely
complicated and difficult. If a practical methodological approach
is used in the interpretation of MRI studies of the brain, these
studies can be extremely useful in helping to narrow the differ-
ential diagnosis of affected patients. Proper understanding and
analysis of proton MR spectroscopy and diffusion weighted
imaging can further help, in some patients, to narrow the diagno-
sis. The use of this approach can significantly simplify the job of
the neurologist or geneticist trying to make the proper diagnosis.
11. Vanhanen SL, Raininko R, Santavuori P. Early differential diagnosis of
infantile neuronal ceroid lipofuscinosis, Rett syndrome, and Krabbe
disease by CT and MR. AJNR Am J Neuroradiol 1994; 15: 1443-453.
12. Hayflick S, Westaway S, Levinson B, Zhou B, Johnson M, Ching K, et
al. Genetic, clinical and radiographic delineation of Hallervorden-
Spatz syndrome. N Engl J Med 2003; 348: 33-40.
13. Hayflick SJ, Penzien JM, Michl W, Sharif UM, Rosman NP, Wheeler
PG. Cranial MRI changes may precede symptoms in Hallervorden-
Spatz syndrome. Pediatr Neurol 2001; 25: 166-9.
14. Moser HW. Adrenoleukodystrophy: from bedside to molecular biolo-
gy. J Child Neurol 1987; 2: 140-50.
15. Van der Knaap MS, Valk J. Magnetic resonance of myelination and
myelin disorders. 3 ed. Berlin: Springer; 2005.
16. D’Incerti L, Farina L, Moroni I, Uziel G, Savoiardo M. L-2-hydroxyg-
lutaric aciduria: MRI in seven cases. Neuroradiology 1998; 40: 727-33.
17. Barkovich AJ, Good W, Koch TK, Berg BO. Mitochondrial disorders:
analysis of their clinical and imaging characteristics. AJNR Am J Neu-
roradiol 1993; 14: 1119-37.
18. Imamura A, Ito R, Tanaka S. High intensity proton and T2 weighted
MR signals in the globus pallidus in juvenile type of dentatorubral and
pallidoluysian atrophy. Neuropediatrics 1994; 25: 234-7.
19. Miyazaki M, Kato T, Hashimoto T, Harada M, Kondo I, Kuroda Y. MR
of childhood onset dentatorubral-pallidoluysian atrophy. AJNR Am J
Neuroradiol 1995; 16: 1834-6.
20. Fukuzawa T, Matsutani S, Maruyama H, Akiike T, Saisho H, Hattori T.
Magnetic resonance images of the globus pallidus in patients with
idiopathic portal hypertension: a quantitative analysis of the relation-
ship between signal intensity and the grade of portosystemic shunt. J
Gastroenterol Hepatol 2006; 21: 902-7.
21. Govaert P, Lequin M, Swarte R, Robben S, De Coo R, Weisglas-Kupe-
rus N, et al. Changes in globus pallidus with (pre)term kernicterus.
Pediatrics 2003; 112: 1256-63.
REV NEUROL 2006; 43 (Supl 1): S5-S16

22. Martich-Kriss V, Kollias S, Ball WS. MR findings in kernicterus.
AJNR Am J Neuroradiol 1995; 16: 819-21.
23. Penn A, Enzmann D, Hahn J, Stevenson D. Kernicterus in a full term
infant. Pediatrics 1994; 93: 1003-6.
24. Lim MK, Suh CH, Kim JK, Cho YK, Choi SH, Kang JH, et al. Sys-
temic lupus erythematosus: brain MR imaging and single-voxel hydro-
gen 1 MR spectroscopy. Radiology 2000; 217: 43-9.
25. Oku K, Atsumi T, Furukawa S, Horita T, Sakai Y, Jodo S, et al. Cerebral
imaging of magnetic resonance imaging and single photon emission
computed tomography in systemic lupus erythmatosus with central
nervous system involvement. Rheumatology (Oxford) 2003; 42: 773-7.
26. Sherwood JW, Wagle WA. Hemolytic uremic syndrome: MR findings
of CNS complications. AJNR Am J Neuroradiol 1991; 12: 703-4.
27. Hager A, Staudt M, Klare B, Von Einsiedel H, Krägeloh-Mann I.
Hemolytic-uremic syndrome with involvement of basal ganglia and
cerebellum. Neuropediatrics 1999; 30: 210-3.
28. Takanashi J, Barkovich A, Cheng S, Kostiner D, Baker J, Packman S.
Brain MR imaging in acute hyperammonemic encephalopathy arising
from late-onset ornithine transcarbamylase deficiency. AJNR Am J
Neuroradiol 2003; 24: 390-3.
29. Takanashi J, Barkovich AJ, Cheng SF, Weisiger K, Zlatunich CO,
Mudge C, et al. Brain MR imaging in neonatal hyperammonemic
encephalopathy resulting from proximal urea cycle disorders. AJNR
Am J Neuroradiol 2003; 24: 1184-7.
30. Autti T, Raininko R, Santavuori P, Vanhanen SL, Poutanen VP, Haltia M.
MRI of neuronal ceroid lipofuscinosis. II. Postmortem MRI and histo-
pathological study of the brain in 16 cases of neuronal ceroid lipofusci-
nosis of juvenile or late infantile type. Neuroradiology 1997; 39: 371-7.
31. Frei KP, Patronas NJ, Crutchfield KE, Altarescu G, Schiffmann R.
Mucolipidosis type IV: characteristic MRI findings. Neurology 1998;
51: 565-9.
32. Melis D, Parenti G, Casa RD, Sibilio M, Romano A, Di Salle F, et al.
Brain damage in glycogen storage disease type I. J Pediatr 2004; 144: 637.
33. Van der Knaap MS, Valk J. The reflection of histology in MR imaging
of Pelizaeus-Merzbacher disease. AJNR Am J Neuroradiol 1989; 10:
99-103.
34. Van der Knaap MS, Valk J. Pelizaeus-Merzbacher disease and X-
linked spastic paraplegia type 2. In van der Knaap MS, Valk J, eds.
Magnetic resonance of myelin, myelination, and myelin disorders. 3
ed. Berlin: Springer; 2005. p. 272-80.
35. Bugiani M, Al Shahwan S, Lamantea E, Bizzi A, Bakhsh E, Moroni I,
et al. GJA12 mutations in children with recessive hypomyelinating
leukoencephalopathy. Neurology 2006; 67: 273-9.
36. Uhlenberg B, Schuelke M, Rüschendorf F, Ruf N, Kaindl AM, Hen-
neke M, et al. Mutations in the gene encoding gap junction protein
alpha-12 (Connexin 46.6) cause Pelizaeus-Merzbacher-like disease.
Am J Hum Genet 2004; 75: 251-60.
37. Battistella PA, Peserico A. Central nervous system dysmyelination in
PIBI(D)S syndrome: a further case. Childs Nerv Syst 1996; 12: 110-3.
38. Porto L, Weis R, Schulz C, Reichel P, Lanfermann H, Zanella FE.
Tay’s syndrome: MRI. Neuroradiology 2000; 42: 849-51.
39. Ostergaard JR, Christensen T. The central nervous system in Tay syn-
drome. Neuropediatrics 1996; 27: 326-30.
40. Linnankivi T, Autti T, Pihko S, Somer M, Tienari P, Wirtavuori K, et al.
18q-syndrome: brain MRI shows poor differentiation of gray and white
matter on T2-weighted images. J Magn Reson Imaging 2003; 18: 414-9.
41. Loevner LA, Shapiro RM, Grossman RI, Overhauser J, Kamholz J.
White matter changes associated with deletions of the long arm of
chromosome 18 (18q-syndrome): a dysmyelinating disorder? AJNR
Am J Neuroradiol 1996; 17: 1843-8.
42. Linnankivi T, Lonnqvist T, Autti T. A case of Salla disease with involve-
ment of the cerebellar white matter. Neuroradiology 2003; 45: 107-9.
43. Sonninen P, Autti T, Varho T, Hämäläinen M, Raininko R. Brain involve-
ment in Salla disease. AJNR Am J Neuroradiol 1999; 20: 433-43.
44. Takanashi J, Inoue K, Tomita M, Kurihara A, Morita F, Ikehira H, et al.
Brain N-acetylaspartate is elevated in Pelizaeus-Merzbacher disease
with PLP1 duplication. Neurology 2002; 58: 237-41.
45. Song SK, Sun SW, Ramsbottom MJ, Chang C, Russell J, Cross AH.
Dysmyelination revealed through MRI as increased radial (but un-
changed axial) diffusion of water. Neuroimage 2002; 17: 1429.
46. Van der Knaap MS, Naidu S, Breiter SN, Blaser S, Stroink H, Springer
S, et al. Alexander disease: diagnosis with MR imaging. AJNR Am J
Neuroradiol 2001; 22: 541-52.
47. Topçu M, Saatci I, Topcuoglu MA, Kose G, Kunak B. Megalencephaly
and leukodystrophy with mild clinical course: a report on 12 new cas-
es. Brain Dev 1998; 20: 142-53.
48. Van der Knaap MS, Kamphorst W, Barth PG, Kraaijeveld CL, Gut E,
Valk J. Phenotypic variation with leukoencephalopathy with vanishing
white matter. Neurology 1998; 51: 540-7.
REV NEUROL 2006; 43 (Supl 1): S5-S16
CONFERENCIA MAGISTRAL
49. Kwan E, Drace J, Enzmann D. Specific CT findings in Krabbe disease.
AJR Am J Roentgenol 1984; 143: 665-70.
50. Koelfen W, Freund M, Jaschke W, Koenig S, Schultze C. GM2 gan-
gliosidosis (Sandhoff’s disease): two year follow-up by MRI. Neurora-
diology 1994; 36: 152-4.
51. Mugikura S, Takahashi S, Higano S, Kurihara N, Kon K, Sakamoto K.
MR findings in Tay-Sachs disease. J Comput Assist Tomogr 1996; 20:
551-5.
52. Barkovich AJ, Ferriero DM, Bass N, Boyer R. Involvement of the ponto-
medullary corticospinal tracts: a useful finding in the diagnosis of X-link-
ed adrenoleukodystrophy. AJNR Am J Neuroradiol 1997; 19: 95-100.
53. Schiffmann R, Moller JR, Trapp BD. Childhood ataxia with diffuse cen-
tral nervous system hypomyelination. Ann Neurol 1994; 35: 331-40.
54. Charnas L, Bernar J, Pezeshkpour GH, Kalakas M, Harper GS, Gahl
WA. MRI findings and peripheral neuropathy in Lowe’s syndrome.
Neuropediatrics 1988; 19: 7-9.
55. Schneider JF, Boltshauser E, Neuhaus TJ, Rauscher C, Martin E. MRI and
proton spectroscopy in Lowe syndrome. Neuropediatrics 2001; 32: 45-8.
56. Cavalleri F, Bernardi A, Burlina A, Ferrari F, Mavilla L. Diffusion-
weighted MRI of maple syrup urine disease encephalopathy. Neurora-
diology 2002; 44: 499-502.
57. Jan W, Zimmerman R, Wang Z, Berry G, Kaplan P, Kaye E. MR diffu-
sion imaging and MR spectroscopy of maple syrup urine disease dur-
ing acute metabolic decompensation. Neuroradiology 2003; 45: 393-9.
58. Barkovich AJ. Pediatric neuroimaging. 4 ed. Philadelphia: Lippincott
Williams & Wilkins; 2005.
59. Graham JW, Jan W. MRI and the brain in systemic lupus erythemato-
sus. Lupus 2003; 12: 891-6.
60. Lalani TA, Kanne JP, Hatfiled GA, Chen P. Imaging findings in sys-
temic lupus erythematosus. Radiographics 2004; 24: 1069-86.
61. Aida N, Tamagawa K, Takada K, Yagishita A, Kobayashi N, Chiku-
maru K, et al. Brain MR in Fukuyama congenital muscular dystrophy.
AJNR Am J Neuroradiol 1996; 17: 605-14.
62. Leite CC, Reed UC, Otaduy MCG, Lacerda MTC, Costa MOR, Fer-
reira LG, et al. Congenital muscular dystrophy with merosin deficien-
cy: H1 MR spectroscopy and diffusion-weighted MR imaging. Radiol-
ogy 2005; 235: 190-6.
63. Barkovich AJ. Neuroimaging manifestations and classification of congen-
ital muscular dystrophies. AJNR Am J Neuroradiol 1998; 19: 1389-96.
64. Van der Knaap MS, Smit LME, Barth PG, Catsman-Berrevoets CE,
Brouwer OF, Begeer JH, et al. MRI in classification of congenital mus-
cular dystrophies with brain abnormalities. Ann Neurol 1997; 42: 50-9.
65. Baum P, Barkovich AJ, Koch T, Berg BO. Deep gray matter involve-
ment in children with acute disseminated encephalomyelitis. AJNR
Am J Neuroradiol 1994; 15: 1275-83.
66. Leake JA, Albani S, Kao AS, Senac MO, Billman GF, Nespeca MP, et
al. Acute disseminated encephalomyelitis in childhood: epidemiologic,
clinical and laboratory features. Pediatr Infect Dis J 2004; 23: 756-64.
67. Braverman N, Dodt G, Gould SJ, Valle D. Disorders of peroxisome
biogenesis. Hum Mol Genet 1995; 4: 1791-8.
68. Faust PL, Banka D, Siriratsivawong R, Ng VG, Wikander TM. Peroxi-
some biogenesis disorders: the role of peroxisomes and metabolic dys-
function in developing brain. J Inherit Metab Dis 2005; 28: 369.
69. Barkovich AJ, Linden CL. Congenital cytomegalovirus infection of the
brain: imaging analysis and embryologic considerations. AJNR Am J
Neuroradiol 1994; 15: 703-15.
70. Boppana S, Fowler K, Vaid Y, Hedlund G, Stagno S, Britt W, et al.
Neuroradiographic findings in the newborn period and long-term out-
come in children with symptomatic congenital cytomegalovirus infe-
cion. Pediatrics 1997; 99: 409-14.
71. Steinlin MI, Nadal D, Eich GF, Martin E, Boltshauser EJ. Late intra-
uterine cytomegalovirus infection: clinical and neuroimaging findings.
Pediatr Neurol 1996; 15: 249-53.
72. Barone R, Nigro F, Triulzi F, Musumeci S, Fiumara A, Pavone L. Clin-
ical and neuroradiological follow-up in mucopolysaccaridosis type III
(Sanfilipo syndrome). Neuropediatrics 1999; 30: 270-4.
73. Barone R, Parano E, Trifiletti R, Fiumara A, Pavone P. White matter
changes mimicking a leukodystrophy in a patient with mucopolysac-
charidosis: characterization by MRI. J Neurol Sci 2002; 195: 171-5.
74. Lee C, Dineen TE, Brack M, Kirsch JE, Runge VM. Mucopolysaccha-
ridoses: characterization by cranial MR imaging. AJNR Am J Neurora-
diol 1993; 14: 1285-92.
75. Murata R, Nakajima S, Tanaka A. MR imaging of the brain in patients
with mucopolysaccharidosis. AJNR Am J Neuroradiol 1989; 10: 1165-70.
76. Barkovich AJ. MR and CT evaluation of profound neonatal and infan-
tile asphyxia. AJNR Am J Neuroradiol 1992; 13: 959-72.
77. Barkovich AJ, Hallam D. Neuroimaging in perinatal hypoxic-ischemic
injury. MRDD Research Reviews 1997; 3: 28-41.
78. Barkovich AJ, Westmark KD, Ferriero D, Sola A, Partridge C. Perina-
S15
A.J. BARKOVICH
tal asphyxia: MR findings in the first 10 days. AJNR Am J Neuroradiol
1995; 16: 427-38.
79. Barnett A, Mercuri E, Rutherford M, Haataja L, Frisone M, Henderson
S, et al. Neurological and perceptual-motor outcome at 5-6 years of
age in children with neonatal encephalopathy: relationship with neona-
tal brain MRI. Neuropediatrics 2002; 33: 242-8.
80. Mercuri E, Rutherford M, Barnett A, Foglia C, Haataja L, Counsell S,
et al. MRI lesions and infants with neonatal encephalopathy. Is the
Apgar score predictive? Neuropediatrics 2002; 33: 150-6.
81. Rutherford M, Pennock J, Schwieso J, Cowan F, Dubowitz L. Hypoxic
ischaemic encephalopathy: early and late magnetic resonance findings in
relation to outcome. Arch Dis Child Fetal Neonatal Ed 1996; 75: 145-51.
82. Neilson DE, Eiben RM, Waniewski S, Hoppel CL, Varnes ME,
Bangert BA, et al. Autosomal dominant acute necrotizing encephalo-
pathy. Neurology 2003; 61: 226-30.
83. Neilson DE, Feiler HS, Wilhelmsen KC, Lynn A, Eiben RM, Kerr DS,
et al. Autosomal dominant acute necrotizing encephalopathy maps to
2q12.1-2q13. Ann Neurol 2004; 55: 291-4.
84. Brismar J, Brismar G, Gascon G, Ozand P. Canavan disease: CT and
MR imaging of the brain. AJNR Am J Neuroradiol 1990; 11: 805-10.
85. Brismar J, Aqueel A, Brismar G. Maple syrup urine disease. AJNR Am
J Neuroradiol 1990; 11: 1219-28.
86. Rachinger J, Fellner FA, Stieglbauer K, Trenkler J. MR changes after
acute cyanide intoxication. AJNR Am J Neuroradiol 2002; 23: 1398-401.
87. Rosenberg NL, Myers JA, Martin WR. Cyanide-induced parkinsonism:
clinical, MRI, and 6-fluorodopa PET studies. Neurology 1989; 39: 142-4.
88. Kinoshita T, Sugihara S, Matsusue E, Fujii S, Ametani M, Ogawa T.
Pallidoreticular damage in acute carbon monoxide poisoning: diffu-
sion-weighted MR imaging findings. AJNR Am J Neuroradiol 2005;
26: 1845-8.
89. O’Donnell P, Buxton PJ, Pitkin A, Jarvis LJ. The magnetic resonance
imaging appearances of the brain in acute carbon monoxide poisoning.
Clin Radiol 2000; 55: 273.
90. Van der Knaap MS, Valk J. Mitochondrial encephalopathy with lactic
acidosis and stroke-like episodes. In van der Knaap MS, Valk J, eds.
Magnetic resonance of myelination and myelin disorders. 3 ed. Berlin:
Springer; 2005. p. 204-11.
91. Dublin A, Hald J, Wootton-Gorges S. Isolated sulfite oxidase deficien-
cy: MR imaging features. AJNR Am J Neuroradiol 2002; 23: 484-5.
92. Topçu M, Erdem G, Saatci I, Aktan G, Limlek A, Renda Y, et al. Clini-
cal and magnetic resonance imaging features of L-2-hydroxyglutaric
acidemia: report of three cases in comparison with Canavan disease. J
Child Neurol 1996; 11: 373-7.
ABORDAJE DE LOS TRASTORNOS METABÓLICOS
EN LA INFANCIA POR RESONANCIA MAGNÉTICA
Resumen. Introducción. Los errores congénitos del metabolismo
constituyen un grupo de trastornos cuyo diagnóstico no resulta fácil.
Las sintomatologías de las distintas enfermedades se solapan, con
frecuencia las pruebas bioquímicas no son específicas o bien son
poco reveladoras, y los análisis genéticos son laboriosos, caros y
muchas veces no están disponibles en el mercado. Las técnicas de
diagnóstico por imagen pueden ser útiles en algunos trastornos, pero
a medida que aumenta el número de enfermedades identificadas y
descritas, en muchos casos los resultados de las exploraciones por
imagen parecen solaparse. Desarrollo. Varios grupos han propuesto
un abordaje por técnicas de imagen basado en el reconocimiento de
patrones en la resonancia magnética (RM), lo cual se complementa
con datos sobre el metabolismo procedentes de la espectroscopia por
RM y datos microestructurales proporcionados por la RM por difu-
sión ponderada. Conclusión. Si se utiliza adecuadamente, este méto-
do de reconocimiento de patrones empleado conjuntamente con otras
herramientas de imagen puede ser muy útil para dividir el complejo
grupo de trastornos metabólicos en otros grupos más fáciles de ma-
nejar. De hecho, esta aproximación permite, a veces, llegar a un diag-
nóstico específico. [REV NEUROL 2006; 43 (Supl 1): S5-16]
Palabras clave. Enfermedad de Canavan. Enfermedades cerebra-
les. Enfermedades mitocondriales. Errores congénitos del metabo-
lismo. Espectroscopia por resonancia magnética. Imágenes de re-
sonancia magnética por difusión. Leucodistrofia. Resonancia mag-
nética. Toxicidad de los medicamentos.
S16
93. Schulze A, Hess T, Wevers R, Mayatepek E, Bachert P, Marescau B, et
al. Creatine deficiency syndrome caused by guanidinoacetate methyl-
transferase deficiency: diagnostic tools for a new inborn error of
metabolism. J Pediatr 1997; 131: 616-31.
94. Stockler S, Isbrandt D, Hanefeld F, Schmidt B, Von Figura K. Guanidi-
noacetate methyltransferase deficiency: the first inborn error of crea-
tine metabolism in man. Am J Hum Genet 1996; 58: 914-22.
95. Item C, Stockler-Ipsiroglu S, Stromberger C, Muhl A, Alessandri M,
Bianchi M, et al. Arginine:glycine amidinotransferase deficiency: the
third inborn error of creatine metabolism in humans. Am J Hum Genet
2001; 69: 1127-33.
96. Salomons G, Van Dooren S, Verhoeven N, Cecil K, Ball W, Degrauw
T, et al. X-linked creatine-transporter gene (SLC6A8) defect: a new
creatine-deficiency syndrome. Am J Hum Genet 2001; 68: 1497-500.
97. Degrauw T, Salomons G, Cecil K, Chuck G, Newmeyer A, Schapiro
M, et al. Congenital creatine transporter deficiency. Neuropediatrics
2002; 33: 232-8.
98. Stockler S, Hanefeld F, Frahm J. Creatine replacement in guanidinoac-
etate methyltransferase deficiency, a novel inborn error of metabolism.
Lancet 1996; 348: 789-90.
99. Moats RA, Mosely KD, Koch R, Nelson M Jr Brain phenylalanine
concentrations in phenylketonuria: research and treatment of adults.
Pediatrics 2003; 112: 1575-9.
100. Möller HE, Weglage J, Bick U, Wiedermann D, Feldmann R, Ullrich
K. Brain imaging and proton magnetic resonance spectroscopy in
patients with phenylketonuria. Pediatrics 2003; 112: 1580-3.
101. Eichler F, Barker P, Cox C, Edwin D, Ulug A, Moser H, et al. Proton
MR spectroscopic imaging predicts lesion progression on MRI in X-
linked andrenoleukodystrophy. Neurology 2002; 26: 901-7.
102. Eichler F, Itoh R, Barker P, Mori S, Garrett E, Van Zijl P, et al. Proton
MR spectroscopic and diffusion tensor brain MR imaging in X-linked
adrenoleukodystrophy: initial experience. Radiology 2002; 225: 245-52.
103. Wilken B, Dechent P, Brockmann K, Finsterbusch J, Baumann M,
Ebell W, et al. Quantitative proton magnetic resonance spectroscopy of
children with adrenoleukodystrophy before and after hematopoietic
stem cell transplantation. Neuropediatrics 2003; 34: 237-46.
104. Tzika A, Ball W Jr, Vigneron D, Dunn RS, Nelson SJ, Kirks D. Child-
hood adrenoleukodystrophy: assessment with proton MR spectroscopy.
Radiology 1993; 189: 467-80.
105. Van der Knaap MS, Valk J. Diffusion-weighted imaging. In van der
Knaap MS, Valk J, eds. Magnetic resonance of myelination and myelin
disorders. 3 ed. Berlin: Springer; 2005. p. 839-53.
ABORDAGEM AOS DISTÚRBIOS METABÓLICOS
NA INFÂNCIA POR RESSONÂNCIA MAGNÉTICA
Resumo. Introdução. Os erros congénitos do metabolismo consti-
tuem um grupo de distúrbios cujo diagnóstico não é fácil. As sinto-
matologias das diferentes doenças sobrepõem-se; frequentemente
os exames bioquímicos ou não são específicos ou são pouco revela-
dores e as análises genéticas são morosas, caras e muitas vezes
não estão disponíveis no mercado. As técnicas de diagnóstico por
imagem podem ser úteis em alguns distúrbios, mas à medida que o
número de doenças identificadas e descritas aumenta, em muitos
casos, os resultados dos exames por imagem parecem sobrepor-se.
Desenvolvimento. Vários grupos propuseram uma abordagem por
técnicas de imagem baseada no reconhecimento de padrões na res-
sonância magnética (RM), a qual é completada com dados sobre o
metabolismo resultantes da espectroscopia por RM e dados micro-
estruturais proporcionados pela RM por difusão ponderada. Con-
clusão. Se for adequadamente utilizado, este método de reconheci-
mento de padrões utilizado conjuntamente com outras ferramentas
de imagem pode ser muito útil para dividir o complexo grupo de
distúrbios metabólicos noutros grupos mais fáceis de manejar. De
facto, esta abordagem permite, por vezes, chegar a um diagnóstico
específico. [REV NEUROL 2006; 43 (Supl 1): S5-16]
Palavras chave. Doença de Canavan. Doenças cerebrais. Doenças
mitocondriais. Erros congénitos do metabolismo. Espectroscopia
por ressonância magnética. Imagens da ressonância magnética por
difusão. Leucodistrofia. Ressonância magnética. Toxicidade dos me-
dicamentos.
REV NEUROL 2006; 43 (Supl 1): S5-S16