Kenneth
D. Pearsen,
MD
#{149} Alisa
D. Gean-Marton,
Phenylketonuria: Brain with Clinical Fifteen patients with biochemically documented phenylketonuria (PKU) were studied with use of magnetic resonance (MR) imaging with spinecho T2-weighted pulse sequences. The resulting images demonstrated varying degrees of symmetric high signal intensity of the white matter within the posterior cerebral hemispheres. Involvement of the antenior hemispheres was seen only in cases with severe signal intensity changes. There was no involvement of the cerebral cortex, brain stem, or cerebellum. Moreover, no anatomic structural abnormalities were observed. Mild cortical atrophy was observed in eight of the 15 patients. There was no significant correlation between the patients’ IQ scores and the level of MR signal intensity changes. Although MR imaging routinely shows relatively distinct abnormalities in patients with PKU, the clinical severity of the disease does not parallel its imaging severity.
Index terms: Brain, abnormalities, Brain, MR studies, 13.1214 #{149} Brain, ter, 13.879 #{149} Phenylketonuria Radiology
1990;
13.599. white mat-
177:437-440
MD
#{149} Harvey
L Levy,
MR Imaging Correlation’
P
autosomal
recessive
is an
genetic
dis-
order resulting from a deficiency in phenylalanine hydroxylase, the enzyme that converts phenylalanine to tynosine (i). The gene for phenylalanine hydroxylase has been cloned and mapped on chromosome 12. A number of mutations that cause PKU have been identified (2). The block in phenylalanine
elevated
degradation
levels
its organic
results
of phenylalanine
acid
metabolites
in
and in blood
and
tissue (1). The clinical phenotype of PKU is limited to the brain. Mental retardation is the most common characteristic of PKU in untreated patients, but other neurologic abnormalities such as autism, seizures, lack of coordination, hyperactive behavior, and hyperreflexia are also seen (3). Treatment with a phenylalanine-nestricted diet controls the biochemical abnormalities and will prevent mental retardation
if begun
in early
infancy
(4,5). Consequently, routine newborn screening for PKU is conducted to identify affected infants so that early presymptomatic treatment can be mitiated (6). Treatment begun after the appearance of developmental delay and other signs of neurologic damage may reverse some of the clinical features
but
will
not
restore
normal
intelligence (7). Despite this extensive knowledge about PKU, the pathogenesis of the brain damage is unknown. The most striking From the Division partment of Radiology K.R.D.) and the Joseph
of Neuroradio!ogy, (K.D.P., A.D.G.M.,
I
tories,
Department
P. Kennedy
of Neurology
Jr.
DeLabora-
(H.L.L.),
Mas-
sachusetts General Hospital, Harvard Medical School, Boston; and the Biochemical Genetics Unit of the Genetic Service, Children’s Hospita!, Harvard Medical School, Boston, (H.L.L.). Received April 4, 1990; revision requested May 2; revision
received
Supported
by
grant
K.D.P.,
NS-05096. 65
14534. C RSNA,
July
National
Caversham
1990
Address
ii;
accepted
Institutes
reprint Woods,
July
13.
of Health
requests Rochester,
to NY
and
most
frequent
#{149} Kenneth
R. Davis,
MD
of the
(PKU)
HENYLKETONURIA
MD
neuro-
pathologic finding, however, has been white matter hypomyelination (8). Occasionally the myelmn reduction is so prominent and diffuse as to mimic Schilders disease (9,10), but, as shown in Figures 1 and 2, it more often involves several different areas of the brain and is accompanied by foci of gliosis (1 1-13). The myelin reduction in PKU appears to result from reduced myelin synthesis rather than excessive myelin loss, since myelin
degradation products are not present (13). This has led to the characterization of this finding as dysmyelmnation rather than demyelination (14). Gray matter changes in PKU have been reported to be absent, with the notable exception of a study by Bauman and Kemper (15) in which three patients had a higher density of cells, decreased
dendritic
cortical
neurons,
reduced
Nissl
arborization
and
of
atrophy
granules
with
in Betz
cells.
If myelmn loss is related to the mental retardation in PKU, those with the lowest intelligence who had received little or no dietary treatment would be expected to have the most severe gray and white matter changes, whereas those who were treated early and have normal intelligence would be expected to have little or no change. To examine this relationship, we studied the magnetic resonance
(MR)
images
of the
brain
in 15 patients with PKU. We compared the MR findings with the biochemical severity of PKU, the dietary treatment status, and the IQ level for each patient. MATERIALS Fifteen confirmed Fourteen
AND
patients PKU patients
ma pheny!alanine mg/dL), while underwent use
of
levels greater one had atypical
detailed
including
recent either
METHODS
with biochemically underwent MR imaging. had classic PKU (plas-
the
clinical
intelligence Wechsler
than 20 PKU. All
evaluations,
testing
with
Intelligence
Scale for Children (WISC) or, if over age 16 years, the Wechsler Adult Intelligence Scale (WAIS). Plasma phenylalanine 1evels were measured with use of an amino acid analyzer (16) or a fluorometric assay (17). nine
The most recent levels obtained
plasma phenylalawere used for pa-
tients on a normal diet, and, for patients receiving dietary treatment, the confirmatory level obtained before initiation the diet was used. The patient’s intelli-
Abbreviation:
PKU
=
of
phenylketonuria.
437
A
‘TX
‘4
4 4
-
‘
#{234}1P
.k
‘4
,
4
Figure
1. Silver-stained demonstrates myelin
tion
coronal pallor.
brain
sec-
3.
Figures
3, 4. Patient
severe
increased
periatriai
signal
zone
centrum
8.
(3) T2-weighted intensity of the
(arrows).
(4)
MR
image
4. axial MR periventricular
image
demonstrates
of 35-year-old white matter,
severe
patient demonstrates most evident in the signal intensity of the
increased
semiovaie.
-P.--
:
:
4 L’
(‘.
Figure
2. Magnified demonstrates
matter lin-eosin
specimen spongiosis.
of white (Hematoxy-
stain.)
gence capacity was categorically graded as normal (IQ 90 and above), mildly abnormal (IQ 70-85), moderately retarded (IQ 50-70), or severely retarded (IQ less than
5.
50).
MR imaging tients
was performed
by means
Medical
Milwaukee)
technique
proton
with
density
on 13 pa-
of a 1 .5-T system
Systems,
echo
and
with
spin-
use of Ti-weighted, T2-weighted
field
trix. One (Technicare
of view, patient [GE
and
formed
by
(A.D.G.M, assessed
and
anomalies, control
location
matter
the
and
atrophy. were
immediate
Radiology
#{149}
at 0.6 T Mu-
used. that
images
weighted ventricular
12. T2-weighted
no significant
axial MR image signal intensity
matter
only),
deep bral
white
axial
white
matter
of a 35-year-old posteriorly.
of gray
congenital Age-matched The most
the white
periatrial
signal matter
conintenin-
semild white
moderate
matter
hemispheres),
to involve
the
MR
image
signal
severely
(also
of the or severe
anterior
involving
posterior
cere-
(extending
cerebral
hemi-
spheres).
RESULTS
were
severity
with use of T2-weighted was graded as absent,
(involving
ma-
1 (Fonar, Melwas per-
changes,
sistent MR finding, sity of periventricular
438
and
(5) Patient
boy shows
neuroradiologists
and subjects
creased quences,
X 256
was evaluated Medical Systems,
K.R.D.), for
white
two
5, 6.
of a 13-year-old
abnormalities. retarded
moderately
(6) Patient
man
shows
mild
re-
15. T2increased
pen-
se-
a 256
waukee]) and one at 0.35 ville, NY). The examination
6.
Figures tarded
(GE
quences (repetition time of 300 or 1,500 msec; echo time of 30, 60, or 120 msec), a 23-cm
,
The Table shows the clinical and biochemical characteristics and the level of white matter signal intensity of the patients. All but one of the T2-weighted images intensity
matter
showed
symmetric high in the peniventniculan of the posterior cerebral
signal white hemi-
spheres (Fig 3), with extension into the frontal lobes in advanced cases only (Fig 4). There was also mild contical atrophy in eight patients, but this was statistically unrelated to the severity of the white matter abnormalities. No signal abnormalities
were
noted
in the
brain
stem,
cere-
bellum, or cortex. The basal ganglia were uniformly spared. No congenital anatomic anomalies were present. The coefficient of correlation between the severity of the white matten signal abnormalities and the patients’ IQ levels was .5, indicating no
November
1990
A ‘I.,,
4, ..
7.
8.
Figures tients
7-9. have
T2-weighted normal
Clinical Intensity
axial
images
intelligence
and
of patient
moderate
9.
1 (9 years
to severe
and Biochemical Characteristics in Patients with PKU
old) (7), patient
increased
and Level
6 (8 years
periventnicular
of White
signal
Matter
(y)
1
9
29.8
Early
(on)
119
Moderate
2 3
14 17
33.3 30.0
Early Early
(off) (on)
102 100
Moderate Mild
4 5 6 7 8 9
51.6 4i.0 29.0 88.0 29.5 38.7 26.4 17.1
Early Early Early Early Never Never Late Early
(on) (off) (on) (on)
11
8 24 8 10 35 29 31 21
12
13
26.9
Early
42.8 22.2 25.0
Late Never Never
13 14 15
Level
(mg/dL)
26 35 35
Early
diet
discontinued I Poorly
began
before
age
Diet*
3 weeks;
at age 5 years. Late diet compliant on diet.
began
on
or off
after
age
refers
IQ
occurred
(off)
Mild Severe Severe Mild Severe Moderate Mild Mild
(on)t
67
Normal
67 20 20
Moderate Moderate Mild
diet
thus
MR changes
continued
or was
and
or the
of the
MR
ed varies
images
(Figs
5, 6), and
five
pa-
tients with essentially normal intelligence displayed moderate (patients 1, 2) to severe (patients 5, 6, and 8) white matter abnormalities (Figs 3, 4,
7-9).
The
remaining
seven
patients
demonstrated parallel clinical and MR imaging severity. The white matter abnormalities visualized at MR also did not significantly correlate with the diet history of the patients. The cortical eight patients correlate with
Volume
177
atrophy observed did not significantly either diet history
Number
#{149}
2
in or
age
was
not
signifi-
cantly correlated with either the IQ or the MR findings. Although the age when myelination of the peniatnial zone is complet-
greatly
normal brains, mal myelination
zone
in individuals the
with
presence of nonin the periatnial
in all age-matched
control
sub-
jects compared with the abnormal signals in nearly all our imaged PKU patients suggests that this finding is not merely a result of normally debayed myelination. We are aware,
however,
that
increased
signal
inten-
sity of the periatrial zone on T2weighted images may persist into adulthood in some individuals with normal intelligence. No deaths have
IQ
patients
patient
pop-
no neuropathoavailable.
either
the
of the
patient.
with
dietary
normal
were
treated diet from had moderate
MR changes patient’s
far in our
This MR study of the brain in PKU revealed a symmetric T2 high signal intensity in the peniventnicular white matter of the posterior cerebral hemispheres in all but one patient. When the changes were severe, the high signal intensity extended into the frontal lobes. There was no clear conrelation between the severity of the
gence who phenylalanine born period
IQ. The
pa-
DISCUSSION
1 year.
significant correlation (P > .1). Three patients with mild mental retardation (patients 10, 1 1, and 12) and one with severe mental retardation (patient 15) had normal or near-normal
(9). Al! three
Signal
history has
old)
ulation; consequently, gic correlation is yet
Intensity
95 90 90 90 88 85 73 72
to whether
5 (24 years
of White
Matter
Phenylalanine
Patient
10
4
Blood
patient
intensity.
Signal Level
Age
old) (8), and
(patients
Two
intelliwith the low the newto severe
2, 8), while
one
untreated patient with severe mental retardation (patient 15) and another with mild retardation (patient 12) had little or no change at MR imaging. To our knowledge, MR imaging characteristics of the brain in PKU
have
not
although that there
ties
in four
been Allen were
previously
described,
et al (18) no brain
patients.
reported abnormali-
However,
that
study was performed with a lowstrength (0.35-T) MR unit. In another study, the MR image of the brain obtamed in one patient with maple syrup urine disease, a genetic disorder of branched-chain amino acid metabolism, showed an increased T2 signal intensity pattern (19) similar to that observed in our study of PKU. The
Radiology
439
#{149}
MR
changes
in maple
syrup
disease correlated with hypomyelination noted pathologic examination.
It is likely
that
signal
abnormalities
served
in PKU
hypomyelination. clear whether
the
is consonant
bution
of myelin
ies
represent
with loss
of
matter
have
ob-
areas
However, distribution
the
served white logic edge,
white we
also
urine
the regions at neuro-
of
it is not we ob-
the
within
correlated IQ level
Acknowledgment: bregt for coordinating
We thank Deborah the study of these
tients
for
specifically
mention
1.
the
(12,20),
which
white
was
in
14 of
our 15 patients. Conversely, neuropathologic studies have identified fected areas of the brain that were found such sule, mains
2.
matter
observed
3.
af-
to be normal in our study, as the brain stem, internal capand cerebellum (12-14). It reto be seen whether these dif-
ferences
reflect
ing MR imaging neunopathobogic
the
limitations
of us-
dysmyelination
Radiology
#{149}
degree observed
of white at MR
matter was
arranging
many
other
Lobpa-
13.
Scriver CR, Kaufman 5, Woo SLC. The hyperphenylalaninemias. In: Scniver CR, Beaudet AL, Sly WS, Val!e D, eds. The metabolic basis of inherited disease. 6th ed. New York: McGraw-Hill, 1989; 495546. Levy HL. Invited editorial: molecular genetics of phenylketonuria and its implications. Am J Hum Genet 1989; 45:667-670. Scriver CR, C!ow CL. Phenylketonunia: epitome of human biochemical genetics. N Eng! J Med 1980; 303:1336-1342, 1394-
Holtzman
NA,
Termination
6.
7.
8.
9.
Welcher
of restricted
phenylketonuria:
DW,
diet
Mellitis
15.
Bauman
Brain
1954;
1959;
82:1-9.
ML, Kemper
and histoanatomic brain in untreated
nuria.
TL.
Morphologic
observations of the human phenylketo-
Acta Neuropathol
(Base!)
1982;
58:55-63.
Spackman
DH, Stein
Automatic recording the chromatography
Chem 17.
WH, Moore apparatus of amino
MW,
Robins
E.
in the determination in serum. J Lab C!in
anine 59:885-890.
con-
18.
19.
20.
Allen
S.
for use in acids. Anal
1958; 30:1190-1206.
McCaman
method
ED.
trolled study. N Engl J Med 1975; 293:1121-1124. Guthnie R, Susi A. A simple phenylalanine method for detecting phenylketonuria in large populations of newborn infants. Pediatrics 1963; 32:338-343. Bickel H, Gerrard J, Hickmans EM. The influence of phenylalanine intake on the chemistry and behavior of a phenylketonuric child. Acta Paediatr 1954; 43:64-77. Malamud N. Neuropathology of phenylketonuria. J Neuropathol Exp Neuro! 1966; 25:254-268. Jervis GA. Phenylpyruvic oligophrenia (phenylketonuria). Assoc Res Nerv Ment Dis
Poser CM, Van Bogaert L. Neuropathologic observations in pheny!ketonunia.
in children
a randomized
of
Bechar M, Bornstein B, E!ian M, Sandbank U. Phenylketonuria presenting an intermittent progressive course. J Neurol Neurosurg Psychiatry 1965; 28:165-170. Alford EC Jr. Stevenson LD, Vogel FS, Engle RL Jr. Neuropatho!ogica! findings in phenylpyruvic oligophrenia (phenylketonuria). J Neuropathol Exp Neuro! 1950; 9:298-310.
14.
16.
Dobson JC, Williamson ML, Azen C, Koch R. Intellectual assessment of 1 1 1 fouryear-old children with phenylketonunia. Pediatrics 1977; 60:822-827.
Crome L. The association of pheny!ketonunia with leucodystrophy. J Neurol Neurosurg Psychiatry 1962; 25:149-153. Corsellis JAN. The pathological report a case of phenylpyruvic o!igophrenia. Neurol Neurosurg Psychiatry 1953;
16:139-143. 12.
adminis-
needs.
with
pathologically by white matter dysmyelination that can be visualized as increased white matter signal intensity on T2-weighted MR images.
the
5.
10.
1 1.
1400. 4.
in identifying findings on the
greater severity of brain disease in those patients who have been exammed neuropathologicalby. In summary, PKU, which leads to mental retardation in the absence of dietary treatment, is characterized
However,
with cior dietary
References
distni-
involvement
peniventnicular
and
trative
matter reported in neunopathostudies of PKU. To our knowlonly two neunopathologic stud-
of the
440
not significantly then the patient’s treatment. U
RJ, Gebarski
5, Aisen
Fluorometric
of phenyla!Med
A.
1962;
Magnetic
resonance (MR) brain imaging in genetic metabolic disease: diagnostic and therapeutic implications. Pediatr Res 1985; 19:387A. Uziel G, Salvoiardo M, Nardocci N. CT and MRI in maple syrup urine disease. Neurology 1988; 38:486-488. Crome L, Pare CMB. Phenylketonunia: a review and a report of the pathological findings in four cases. J Ment Sci 1960; 106:862-883.
33:259-282.
November
1990