J Inherit Metab Dis (2011) 34:763–780

DOI 10.1007/s10545-011-9317-5

ORIGINAL ARTICLE

Brain and spine MRI features of Hunter disease: frequency,

natural evolution and response to therapy
Renzo Manara & Elena Priante & Marco Grimaldi & Lucia Santoro & Luca Astarita &
Rita Barone & Daniela Concolino & Maja Di Rocco & Maria Alice Donati &
Simona Fecarotta & Anna Ficcadenti & Agata Fiumara & Francesca Furlan &
Irene Giovannini & Franco Lilliu & Rodica Mardari & Gabriele Polonara &
Elena Procopio & Angelica Rampazzo & Andrea Rossi & Graziolina Sanna &
Rossella Parini & Maurizio Scarpa

Received: 18 December 2010 / Revised: 8 March 2011 / Accepted: 9 March 2011 / Published online: 5 April 2011
# SSIEM and Springer 2011

Abstract evolution and the impact of intravenous enzymatic replace-

Backgroud Hunter disease is a rare X-linked mucopolysac- ment therapy (ERT).
charidosis. Despite frequent neurological involvement, Methods Sixty nine brain MRI examinations of 36 Italian
characterizing the severe phenotype, neuroimaging studies patients (mean-age 10.4 years; age-range 2.2-30.8; severe
are scarce. phenotype in 22 patients) were evaluated. Twenty patients
Objectives To determine frequency and severity of neuro- had multiple MRIs (median follow-up 3.1 years, range
radiological mucopolysaccharidosis-related features; to cor- 1–16.9): among them 15 had MRIs before and after ERT,
relate them with clinical phenotype; to evaluate their natural six had repeated MRIs without being on ERT and five
while on ERT. Perivascular, subarachnoid and ventricle
Communicated by: Ed Wraith space enlargement, white matter abnormality (WMA)
Competing interest: None declared. burden, pituitary sella/skull/posterior fossa abnormalities,
periodontoid thickening, spinal stenosis, dens hypoplasia,
Dr. M. Ermani, (Neurology Department, University Hospital of
Padua) performed the statistical analysis.
R. Manara (*) : R. Mardari R. Barone : A. Fiumara
Neuroradiologic Unit, University Hospital of Padua, Referral centre for inherited metabolic diseases,
via Giustiniani 2, Department of Pediatrics, University of Catania,
35128 Padova, Italy Catania, Italy
e-mail: renzo.manara@sanita.padova.it
D. Concolino
E. Priante : I. Giovannini : A. Rampazzo : M. Scarpa Department of Pediatrics,
Centre for Rare Diseases, Department of Pediatrics, Magna Graecia University of Catanzaro,
University of Padua, Catanzaro, Italy
Padua, Italy
M. Di Rocco
M. Grimaldi Unit of Rare Diseases, Department of Pediatrics,
Neuroradiology Section, Department of Radiology, Gaslini Institute Genoa,
San Gerardo Hospital, Genova, Italy
Monza, Italy
M. A. Donati : E. Procopio
L. Santoro : A. Ficcadenti Metabolic and Neuromuscular Unit, Meyer Children Hospital,
Metabolic diseases, Institute of Maternal-Infantile Sciences, University of Florence,
Polytechnic University of Marche, Florence, Italy
Ancona, Italy
F. Furlan : R. Parini
L. Astarita : S. Fecarotta Rare Metabolic Diseases Unit Fondazione Mariani,
Department of Pediatrics, Federico II University, Pediatric Department, San Gerardo Hospital,
Naples, Italy Monza, Italy
764
myelopathy, vertebral and intervertebral disc abnormalities
were graded by means of dedicated scales.
Results Perivascular spaces enlargement (89%), WMAs
(97%), subarachnoid space enlargement (83%), IIIrd-
ventricle dilatation (100%), pituitary sella abnormalities
(80%), cranial hyperostosis (19%), craniosynostosis (19%),
enlarged cisterna magna (39%), dens hypoplasia (66%),
periodontoid thickening (94%), spinal stenosis (46%),
platyspondylia (84%) and disc abnormalities (79%) were
frequently detected. WMAs, IIIrd-ventricle dilatation and
hyperostosis correlated with the severe phenotype (p<
0.05). Subarachnoid spaces and ventricle enlargement,
WMAs and spinal stenosis progressed despite ERT, while
other MR features showed minimal or no changes.
Conclusions The spectrum of brain and spine MRI abnor-
malities in Hunter disease is extremely wide and requires a
thorough evaluation. WMAs, atrophy/communicating hy-
drocephalus and spinal stenosis progress over time and
might represent possible disease severity markers for new
treatment efficacy assessment.
Introduction
Mucopolysaccharidoses (MPS) comprise a heterogeneous
group of rare recessive lysosomal storage diseases all
caused by impaired lysosomal glycosaminoglycans
(GAG) catabolism. Hunter disease (MPS type II,
OMIM#309900) is characterized by X-linked inheritance
and is caused by mutations of the gene encoding
iduronate-2-sulphatase. The accumulation of the upstream
metabolites (heparan and dermatan sulphate) within a variety
of cells, tissues and organs is responsible for progressive
multisystemic involvement.
Clinical manifestations include typical facial dysmor-
phism and skeletal deformities, severe airway obstruction,
cardiomyopathy, organomegaly, joint stiffness and, in most
patients, neurological decline. According to the phenotypic
expression, patients are traditionally divided into two
groups. In the severe form, clinical features appear between
F. Lilliu : G. Sanna
Department of Biomedical Sciences and Biotechnologies,
Cagliari, Italy
G. Polonara
Radiology Section, Department of Clinical Sciences,
Polytechnic University of Marche,
Ancona, Italy
A. Rossi
Pediatric Neuroradiology Department,
G. Gaslini Children’s Hospital,
Genoa, Italy
J Inherit Metab Dis (2011) 34:763–780
2 and 4 years of age, the neurological involvement is
progressive and death usually occurs in the first or second
decade. In the attenuated form clinical signs and symptoms
have a slightly later onset, neurological deficit is minimal or
even absent and patients may survive till late adulthood
(Martin et al. 2008).
Until recently, therapy for Hunter disease has been
palliative. In 2006 the Food and Drug Administration has
approved intravenous Enzyme Replacement Therapy
(ERT) with recombinant human iduronate-2-sulphatase
(idursulfase; Elaprase, Shire Human Genetic Therapies,
Cambridge, MA, USA), which had been shown to
improve many of the signs and symptoms as well as
the general wellbeing of patients with Hunter disease
(Wraith et al. 2008). The true impact of idursulfase
therapy on the Central Nervous System (CNS) manifes-
tations is however still debated and data are scarce.
The complex multisystem involvement of Hunter
patients requires a multidisciplinary management and
follow up (Wraith et al. 2008). This will include the
neuroimaging since CNS involvement needs to be early
diagnosed and strictly monitored, especially to prevent
communicating hydrocephalus- or spinal stenosis-related
consequences. Magnetic Resonance Imaging (MRI) repre-
sents undoubtedly the most sensitive method to detect brain
and spinal abnormalities in mucopolysaccharidosis (Faerber
and Poussaint 2002). Several MRI findings have been so
far described in Hunter patients but all the evidence is
based on case reports and small and heterogeneous MPS
cohorts (Lee et al. 1993; Parsons et al. 1996; Seto et al.
2001; Matheus et al. 2004; Müller-Forell et al., 2007;
Vedolin et al. 2007a and b, Finn et al. 2008). Actually, we
lack systematic neuroimaging studies including data on
frequency and evolution of specific MRI abnormalities as
well as their correlation with the clinical phenotype.
Furthermore, the nature of these alterations is still debated
and most of the physiopathological hypotheses, that have
been thoroughly proposed (Lee et al. 1993; Parsons et al.
1996; Seto et al. 2001; Matheus et al. 2004; Vedolin et al.
2007a and b; Finn et al. 2008; Wraith et al. 2008) do not
explain the observed neuroradiological findings. Finally, it
is undetermined which features are merely markers of the
disease and which might be used for establishing the
efficacy of newly introduced therapies.
This cross-sectional and longitudinal retrospective
study reports on cerebral and spinal MRI features in a
large Italian group of Hunter patients in order to 1)
define their frequency and correlation with the clinical
phenotype, 2) delineate their natural course as well as the
impact of ERT; 3) be able to hypothesise on potential
physiopathological mechanisms and 4) identify which
MRI features might be used to ascertain efficacy of
potential new therapies.
J Inherit Metab Dis (2011) 34:763–780 765

Patients and methods The following MRI features were scored:

– enlargement of Virchow-Robin perivascular spaces (PVSs)
We enrolled 36 Italian patients affected by Hunter disease (35
separately in corpus callosum and supratentorial white
males and 1 female, age range 2.8-32.2 years, mean age
matter according to Lee (Lee et al. 1993). A sieve-like
12.6 years). The diagnosis was made by urinary glycosamino-
appearance of supratentorial white matter was scored as 4
glycans measurements followed by demonstrating a deficient
(Fig. 1).
activity of iduronate-2-sulphatase in cultured fibroblasts or in
– white matter abnormality (WMA) burden in T2 weighted
leukocytes. Multiple sulphatase deficiency was excluded by
and Fluid Attenuated Inversion Recovery images (Fig. 2).
detection of normal activity of other lysosomial sulphatases.
– enlargement of subarachnoid cerebro-spinal fluid (CSF)
The phenotype was judged as “attenuated” in 14 (age
spaces according to Lee (Lee et al. 1993) (Fig. 3).
range 4.6-32.2 years; mean age 15.9 years) and “severe” in
– enlargement of IIIrd ventricle according to Lee (Lee et
22 patients (age range 2.8-20.1 years; mean age 10.5 years)
al. 1993) (Fig. 4).
by paediatricians expert in metabolic disorders according to
the classification recently published by Muenzer (Muenzer Enlargement of subarachnoid and ventricle CSF spaces
et al. 2009). Briefly, patients who presented with CNS were considered as markers of atrophy and or communi-
involvement, resulting in impaired cognitive abilities, cating hydrocephalus.
behavioural problems and neuropsychological developmen-
Skull and pituitary sella
tal delay or regression were judged to be affected by severe
phenotype, whereas patients who exhibited little or none
– cranial bones abnormalities, e.g. pituitary sella abnormal-
neurological involvement were considered “attenuated
ities (Fig. 5), cranial vault hyperostosis (Fig. 6a) as
phenotype”. Thirty-three patients received weekly infusions
proposed by Avrahami (Avrahami et al. 1987), craniosy-
of idursulfase (mean treatment duration 3.25 years, range
nostosis (scaphocephaly or trigonocephaly) (Fig. 6b) and
0.5-5.8 years); four subjects underwent bone marrow
platybasia as proposed by Koenisberg (Koenisberg et al.
transplantation; one patient had both treatments.
2005).
Two experienced neuroradiologists, blind to clinical
data, jointly evaluated retrospectively all brain and spine
MRI studies performed from 1991 to 2009. This included a Posterior fossa
total of 69 brain scans. Cervical spine studies were
available in 32 patients. All images had been acquired with – enlarged cisterna magna, small posterior fossa and
1.0T or 1.5T scanners and were judged of good quality, Chiari I malformation (Fig. 7).
despite the large time span considered. For 12 patients, the
neuroradiologists evaluated the images separately and inter-
Craniovertebral junction
observer reliability was calculated (Table 1).
Twenty patients had more than one MRI examination
– dens hypoplasia (only in subjects older than 2 years)
allowing longitudinal analysis (mean follow-up 4.1 years;
(Fig. 8)
median follow-up 3.1 years; range 1–16.9 years). Among
– atlanto-occipital assimilation
these patients, 15 subjects had MRI examinations before
– basilar invagination
and after ERT introduction (mean follow-up 3 years;
– characteristic soft tissue thickening around the odon-
median follow-up 2.4 years; range 1–8.7 years; mean
toid peg (measured from posterior cortical rim of the
follow-up after ERT beginning 1.7 years; range 1–
dens until the membrane tectoria) (Fig. 9);
2.9 years); six patients had multiple MRI examinations
– spinal stenosis according to the Space Available for the
without receiving ERT (mean follow-up 5.4 years; median
Cord (SAC) scale (Tierney et al. 2002) (Fig. 9). In
follow-up 2.5 years; range 1.2–17 years); three of them
particular the SAC was determined by subtracting the
eventually received ERT. Five patients had multiple
sagittal spinal-cord diameter from the corresponding
examinations while on ERT (mean follow-up 2.3 years;
sagittal spinal-canal diameter; a previous decompres-
median follow-up 2.0 years; range 1.5–3.6 years) .
sive surgery was scored as 5.
According to the available literature and our experience on
– mielopathy; syringomyelia was scored 3.
MRI abnormalities in mucopolysaccharidoses, we identified
several neuroradiological alterations, which were scored by
means of: 1) scales already present in literature (e.g. dilatation Spine
of IIIrd ventricle), 2) scales adapted to our study (e.g. spinal
stenosis) and 3) scales specifically created (e.g. intersomatic – cervical platyspondilia (in subjects older than 15 months)
disc abnormalities, white matter abnormality burden) (Table 1). (Fig. 10a)
766 J Inherit Metab Dis (2011) 34:763–780

Table 1 Scores of brain and spine MRI features used in the present study

MRI feature Scale

Enlargement of Virchow-Robin PVSs inside cor- 0 = none

pus callosum 1 = <10 PVSs with a diameter <3 mm
2 = >10 PVSs with a diameter <3 mm
3 = PVSs with a diameter of >3 mm
Enlargement of Virchow-Robin PVSs outside 0 = none
corpus callosum 1 = <10 PVSs with a diameter <3 mm
2 = >10 PVSs with a diameter <3 mm
3 = PVSs with a diameter of >3 mm
4 = sieve-like appearance of supratentorial white matter.
White matter signal abnormalities in T2-w and 0 = absent
FLAIR images 1 = mild
2 = moderate
3 = severe
Subarachnoid CSF spaces enlargement 0 = absent
1 = mild (widening of the sylvian and interhemispheric fissures <3 mm, but not all sulci are
involved)
2 = moderate (widening of all fissures and sulci between 3–5 mm with visualization of the
entire mesencephalic cistern)
3 = severe (widening of all fissures and sulci >5 mm with definite loss of cortex and white
matter)
IIIrd ventricle enlargement 0 = absent (<3 mm)
1 = mild (<5 mm but no temporal horn dilatation)
2 = moderate (<10 mm with temporal horn dilatation)
3 = severe (>10 mm)
Pituitary sella abnormalities (j-shaped, ω-shaped 0 = normal
or enlarged sella) 1 = present
Hyperostosis 0 = absent (<10 mm)
1 = present (>10 mm)
Craniosynostosis (scaphocephaly or 0 = absent
trigonocephaly) 1 = present
Platybasia 0 = absent (<125°)
1 = present (>125°)
Enlarged cisterna magna 0 = absent
1 = present
Chiari I malformation 0 = absent
1 = borderline (cerebellar tonsills herniation <5 mm)
2 = present (cerebellar tonsills herniation >5 mm)
Small posterior fossa 0 = absent
1 = present
Dens dysplasia 0 = absent
1 = mild (dysplasic dens reaching atlas arch)
2 = severe (dysplasic dens not reaching atlas arch)
Periodontoid tissue thickening 0 = normal (<4 mm)
1 = mild (<10 mm)
2 = severe (>10 mm)
Spinal stenosis 0 = normal (SAC 10.4-2.5 mm)
1 = mild stenosis (SAC 2.5-1 mm)
2 = severe stenosis (SAC <1 mm)
3 = mild cord compression (spinal cord diameter >5 mm)
J Inherit Metab Dis (2011) 34:763–780 767

Table 1 (continued)

MRI feature Scale

4 = severe cord compression (spinal cord diameter <5 mm)

5 = previous decompressive surgery
Myelopathy 0 = absent
1 = non-specific T2-hyperintensities
2 = T2-hyperintense lesion
3 = syringomyelia
Atlanto-occipital assimilation 0 = absent
1 = present
Basilar invagination 0 = absent
1 = present
Platyspondylia 0 = absent
1 = present
Other vertebral anomalies (schisis or vertebral 0 = absent
body fusion) 1 = present
Intervertebral disc abnormalities 0 = none
1 = reduction in height
2 = partially absence
3 = absence of intervertebral disc

PVSs: perivascular spaces; FLAIR: fluid attenuated inversion recovery; CSF: cerebro-spinal fluid; SAC: space available for the cord

– other vertebral anomalies, such as schisis or vertebral Statistical analysis

body fusion (Fig. 10c,d)
– intervertebral disc abnormalities (Fig. 10b).
Frequency and severity of these abnormalities were
determined with a cross-sectional analysis of our cohort;
when more than 1 MRI examination was available, the most
recent examination was used (age range 2,2-30.8 years; mean
age: 10,4 years).
The variables with normal distribution were analysed using
the Student T test, while for variables not normally distributed
the Mann Whitney U test was used. Cathegorial variables
were analysed using the Chi square of Pearson or the Fisher
Exact test when required. The linear correlation was tested
using the Spearman Rho. Interobserver agreement was tested
by using the Kappa of Cohen coefficient. Significance level
was set at p<0.05.

Fig. 1 Perivascular spaces

(PVSs) enlargement spectrum.
Bilateral enlarged PVSs in the
posterior region of centrum
semiovale, score 3 (a) and in the
basal ganglia and in juxta-
cortical regions, score 3 (b);
giant PVS in the right lenticular
nucleus, score 3 (c); typical
enlarged PVSs in the corpus
callosum, score 3 (d); sieve-like
appearance of the posterior part
of cerebral hemispheres, score 4
(e); enlarged PVSs in the mid-
brain (f and g)
768
Fig. 2 White matter signal abnormality spectrum. T2w axial images at the level of lateral ventricles showing minimal/absent, score 0 (a), mild (<1/3,
score 1) (b), moderate (<2/3, score 2) (c) and severe (>2/3, score 3) (d) supratentorial white matter involvement
Patient consent
Written informed consent was obtained from all patients
(or guardians of patients) participating in the study.
Results
Cross-sectional analysis
Frequency and severity of brain and spine MRI features are
shown in Table 2.
Among pituitary abnormalities, four patients had a marked
enlargement of the sella regardless of the severity of ventricular
and subarachnoid enlargement. The presence of an enlarged
cisterna magna did not correlate with the presence and the
severity of enlargement of supratentorial CSF spaces (either
ventricle [p=0,58, Mann-Whitney test] or subarachnoid spaces
[p=0.41, Mann-Whitney test]). Regarding the craniovertebral
junction no association was found between dens hypoplasia
severity and the presence of periodontoid thickening or spinal
stenosis (p=0.31 and p=0.68 respectively, Chi-square test). On
Fig. 3 Atrophy/communicating
hydrocephalus in 3 patients with
severe phenotype. Marked CSF
spaces enlargement with promi-
nent involvement of subarach-
noid (a) ventricular (b) or both
(c) compartments
J Inherit Metab Dis (2011) 34:763–780
the other hand, we found a direct correlation between age and
periodontoid thickening (p=0.047, Spearman’s Rho test) and a
inverse correlation between age and sella abnormalities (p<
0.005, Student t-test) and dens hypoplasia (p<0.0005, Student
t-test).
The extent of WMAs (p=0.016, Mann-Whitney test), IIIrd
ventricle enlargement (p=0.0005, Mann-Whitney test) and the
presence of occipital hyperostosis (p=0.019, Chi-square test)
correlated with a severe phenotype while the presence of an
enlarged cisterna magna was more frequent in patients with
attenuated phenotype (p=0.016, Chi-square test).
MRI findings of patients who underwent bone marrow
transplantation did not significantly differ from the whole
group of Hunter patients.
Longitudinal analysis
Table 3 reports MRI changes between first and last MRI
examinations of all 20 patients with an MRI follow-up.
Table 4 summarizes MRI changes in those 15 patients
with an MRI examination shortly before and after ERT
introduction, while Table 5 compares MRI changes without
J Inherit Metab Dis (2011) 34:763–780 769

Fig. 4 Atrophy/communicating
hydrocephalus spectrum at level
of the IIIrd ventricle Mild, score
1 (a) and severe, score 3 (b)
enlargement of IIIrd ventricle

ERT (6 patients) with MRI changes during ERT (5
patients).
WMAs, subarachnoid CSF spaces and IIIrd ventricle
enlargement worsened in several cases, though these MRI
features were graded at maximum score at the first
examination in 5/19, 5/19 and 7/19 subjects, respectively.
These changes were observed regardless of ERT introduc-
tion (Fisher’s exact test).
Also spinal stenosis at cranio-vertebral junction showed
a comparable worsening in the whole group and in those
patients examined before/after ERT introduction, without or
while on ERT (Fisher’s exact test).
Cranial vault hyperostosis and periodontoid thicken-
ing progression was observed during the follow-up of
patients not receiving ERT but comparison with other
groups did not reach statistical significance (Fisher’s
exact test).
PVSs enlargement score changes were seldom observed,
likely due to different positioning of slices in follow-up
examinations, without significant differences among groups
(Chi-square test).
In few cases dens hypoplasia and pituitary sella
abnormalities showed an amelioration regardless of ERT
(Chi-square test and Fisher’s exact test, respectively).

Fig. 5 Pituitary sella abnormal-

ity spectrum. Magnification has
been kept relatively constant in
order to show size differences
(the brainstem is similar in all
images): typical J-shaped (a)
or ω-shaped (b) pituitary sella;
small sella in a hyperostotic
sphenoid bone (c); enlarged
sella with slight prolapse of the
optic chiasm (d)
770
Fig. 6 Cranial vault bone abnor-
malities. Thecal hyperostosis (a);
craniosynostosis with dolicho-
cephaly (b)
Discussion
Hunter disease is one of the MPS in which the CNS is
primarily involved. This is the first large study aimed at
determining frequency and evolution of known MPS-related
MRI features in Hunter disease.
Fig. 7 Posterior fossa abnormality spectrum. Magnification has been
kept relatively constant in order to show posterior fossa differences in size
(the brainstem is similar in all images): small posterior fossa with
impingement of neural structures (a); medium-sized posterior fossa with
modest impingement of cerebellar tonsils in the foramen magnum; CSF
J Inherit Metab Dis (2011) 34:763–780
Enlargement of Virchow-Robin perivascular spaces (PVSs)
Virchow-Robin PVSs enlargement is a known but non-
specific feature of MPS since it has also been detected in a
number of different conditions such as myotonic dystrophy,
neurosarcoidosis, parkinsonism, chronic alcoholism and in
space is visible below the cerebellar vermis (b); giant posterior fossa with
megacisterna magna (c); megacisterna magna with dural sepimentation
and relative cerebellar hypoplasia on coronal images (d, e); Chiari 1
malformation with cerebellar tonsils below the foramen magnum (f)
J Inherit Metab Dis (2011) 34:763–780
Fig. 8 Dens hypoplasia spectrum. Normal finding with a well recognizable dens, score 0 (a); mild dens hypoplasia with the odontoid process reaching
the anterior arch of the atlas, score 1 (b); severe dens hypoplasia, score 2 (c); normal dens with severe spinal canal stenosis (d)
healthy subjects, as well. PVSs in healthy subjects tend to
increase in number and size with age, however, in contrast
to PVSs in MPS, they rarely involve the corpus callosum
(Groeschel et al. 2006). At present, the rare available neuro-
pathologic reports do not clarify the pathophysiologic
mechanisms leading to enlarged PVSs. It is unclear whether
PVSs enlargement is due to the storage of GAG around the
vessels (Parsons et al. 1996) or to hampered liquid
reabsorption caused by deposits of mucopolysaccharides
in the leptomeninges (Matheus et al. 2004). Both mecha-
nisms are supposed to lead to a progressive enlargement of
PVSs thus giving a “sieve-like” aspect in the late phases of
disease (Matheus et al. 2004).
Our data confirm that the PVSs enlargement is frequent
in Hunter disease, since more than 90% of patients
presented these findings, especially at the level of corpus
Fig. 9 Atlo-occipital junction and spectrum of spinal canal stenosis
due to periodontoid thickening. Normal sized spinal canal, score 0 (a),
mild stenosis with reduced CSF perimidollar space (<2.5 mm, score 1)
771
callosum. Nonetheless, no correlation was found between
the clinical phenotype and the size and location of
Virchow-Robin PVSs since they were massively present
also in the attenuated form as demonstrated in our series as
well as in the literature (Shimoda-Matsubayashi et al. 1990;
Parsons et al. 1996; Shinomiya et al. 1996). In contrast to
previous studies which suggested a reduction (Lee et al.
1993) or an increase (Parsons et al. 1996) of the PVSs
enlargement, in our study age did not correlate with the
severity of PVSs enlargement. Furthermore, during follow-up
only limited changes in PVSs were observed. Indeed, a slight
worsening was detected in three out of 18 patients while a
slight improvement was found in two patients. These changes,
however, might be due to different slice positioning. There-
fore, our data strongly suggests that PVSs do not significantly
evolve in Hunter disease.
(b), severe stenosis with reduced CSF spaces (<1 mm, score 2) (c) and
mild spinal cord compression with absent CSF spaces and reduced
spinal cord sagittal diameter, score 3 (d)
772
Fig. 10 Spine abnormalities.
Typical platyspondylia with
bullet-shaped vertebral bodies
(a); intervertebral discs with
reduced height in their central
part due to nucleus pulposus
hypotrophy, score 1 (b); anterior
arch schisis of the atlas in a
9-year-old boy which is more
visible on CT scan (c) than on
MRI (D); wedge-shaped L1
vertebral body (this deformity is
not rarely encountered at dorso-
lumbar level in MPS)
White matter signal abnormalities
WMAs are frequently observed in Hunter disease, partic-
ularly in the periventricular regions. A combination of
gliosis, loss of axon and myelin probably represents the
pathological substrate of these abnormalities (Lee et al.
1993; Vedolin et al. 2007a and b), but direct pathological
findings are still lacking.
Several visual semiquantitative scores of white matter signal
abnormalities have been conceived for physiological ageing or
vascular dementia (Fazekas et al. 1987; Scheltens et al. 1993,
Wahlund et al. 2001). No grading systems has been developed
for evaluating WMAs in lysosomal storage diseases, where
the white matter involvement is often diffuse and not clearly
related to ischemia. Since most of our examinations were on
films, we applied a semi-quantitative scale thus confirming a
primary supratentorial white matter involvement in Hunter
disease. The burden of WMAs was graded moderate or severe
in nearly 75% of cases and directly correlated with the severe
clinical phenotype. These findings, therefore, confirm the
correlation between WMAs burden and the presence of
cognitive impairment, already outlined in previous studies
(Vedolin et al. 2007a; Gabrielli et al. 2004).
J Inherit Metab Dis (2011) 34:763–780
Atrophy and communicating hydrocephalus
One of the most challenging aspect in Hunter disease
management is the differentiation between progressive
brain atrophy and an incipient communicating hydroceph-
alus, which would require a prompt surgical intervention.
Both manifestations might share the same clinical and
instrumental findings. Firstly, the expected worsening of
neurological status in a patient with the severe phenotype,
which is often stepwise, might conceal the neurological
deterioration due to hydrocephalus. Secondly, fundoscopy
and visual evoked potential studies may be inconclusive
because of poor compliance and because chronic papil-
ledema may occur in Hunter patients even in the absence of
increased intracranial pressure, probably due to deposition
of GAG within the sclera (Beck and Cole 1984). Thirdly, in
both conditions CT and MRI examinations detect an
enlargement of subarachnoid and ventricular spaces. Final-
ly, neuroimaging cannot easily differentiate between
Hunter-related periventricular WMAs and transependimal
leakage due to impaired CSF reabsorption in communicat-
ing hydrocephalus. All the above may easily lead to a
delayed or inappropriate surgical treatment.
J Inherit Metab Dis (2011) 34:763–780 773

Table 2 Frequency and severity of neuroradiological findings

0 1 2 3 4 n. a. Cohen’s K

Enlargement of Virchow-Robin PVSs inside CC 4/35 19/35 3/35 9/35 1 1

11% 54% 9% 26%
Enlargement of Virchow-Robin PVSs outside CC 6/36 5/36 5/36 18/36 2/36 1
17% 14% 14% 50% 5%
White matter signal abnormalities 1/35 8/35 12/35 14/35 1 0.71 (p=0.0002)
3% 23% 34% 40%
Subarachnoid CSF spaces enlargement 6/36 6/36 6/36 18/36 0.77 (p<0.0001)
17% 17% 17% 50%
IIIrd ventricle enlargement 0/36 5/36 13/36 18/36 0.74 (p<0.0001)
14% 36% 50%
Pituitary sella abnormalities 7/36 29/36 1
19% 80%
Hyperostosis 29/36 7/36 1
81 % 19 %
Craniosynostosis 29/36 7/36 1
81% 19%
Platybasia 36/36 0/36 1
100%
Enlarged cisterna magna 22/36 14/36 1
61% 39%
Chiari I malformation 34/36 1/36 1/36 1
94% 3% 3%
Small posterior fossa 33/36 3/36 1
92% 8%
Dens dysplasia 11/35 14/35 10/35 1 0.75 (p=0.0001)
31% 40% 29%
Periodontoid tissue thickening 2/35 31/35 2/35 1 1
6% 89% 6%
Spinal stenosis 19/35 9/35 4/35 3/35 1 0.85 (p<0.0001)
54% 26% 11% 9%
Myelopathy 34/35 0/35 0/35 1/35 1 1
97% 3%
Atlanto-occipital assimilation 36/36 0/36 1
100%
Basilar invagination 35/35 0/35 1 1
100%
Platyspondylia 5/32 27/32 4 1
16% 84%
Other vertebral anomalies 28/32 4/32 4 1
87% 12%
Intervertebral disc abnormalities 6/29 21/29 2/29 0/29 7 1
21% 72% 7%

n.a: not available; PVSs: perivascular spaces; CC: corpus callosum; CSF: cerebro-spinal fluid Cohen’s K: Cohen’s kappa coefficient
The score ranges from 0 (normal) to 1–4 according to the scale used for the evaluation . The thick vertical line discriminates between normal and
abnormal findings.
Missing data and Cohen’s K coefficient as a measure of Inter-observer agreement obtained from a sample of 12 patients are also reported.

Due to the difficulties in differentiating brain atrophy
from communicating hydrocephalus these conditions were
considered together. In our series atrophy/communicating
hydrocephalus was extremely frequent: with regard to their
last MR examination all patients presented an enlargement
of the third ventricle while more than 80% showed an
enlargement of subarachnoid spaces. In as many as half of
the cases they were graded severe. The clinical impact of
these features is still debated. Some authors have reported a
correlation between severity of brain atrophy and cognitive
774 J Inherit Metab Dis (2011) 34:763–780

Table 3 MRI changes between

first and last MRI examinations Unchanged Worsened Improved Not available
in 20 patients. Mean follow-up
4.1 years; median follow-up Enlargement of Virchow-Robin PVSs inside CC 13**/18 3*/18 2/18 2
3.1 years; range 1–16.9 years Enlargement of Virchow-Robin PVSs outside CC 17**/19 1/19 1*/19 1
White matter signal abnormalities 12***/19 7/19 1
Subarachnoid CSF spaces enlargement 10*/19 9**/19 1
IIIrd ventricle enlargement 11*/19 8**/19 1
Pituitary sella abnormalities 18*/20 2**/20
Hyperostosis 17*/20 3**/20
Craniosynostosis 20***/20
Platybasia 16*/16 4**
Enlarged cisterna magna 19***/20 1/20
Chiari I malformation 20***/20
Small posterior fossa 20***/20
Dens dysplasia 16**/19 1/19 2*/19 1
Periodontoid tissue thickening 14**/18 3*/18 1/18 2
PVSs: perivascular spaces; CC: Spinal stenosis 13*/20 7**/20
corpus callosum; CSF: cerebro-
Myelopathy 17*/18 1°/18 2**
spinal fluid;
Atlanto-occipital assimilation 17**/17 3*
syringomyelia in Chiari I
malformation; Basilar invagination 20***/20
* indicates the number of patients Platyspondylia 15*/15 5**
who underwent bone marrow Other vertebral anomalies 14*/15 1/15 5**
transplantation (* one patient, ** Intervertebral disc abnormalities 12*/15 2/15 1/15 5**
two patients, *** three patients).

impairment in MPS (Vedolin et al. 2007a), while others did
not find any association (Gabrielli et al. 2004; Barone et al.
1999). Our data showed a highly significant correlation
between the degree of third ventricle enlargement and the
clinical phenotype. These findings are consistent with the
quantitative data published by Fan, who found the ventricle
enlargement as the sole direct parameter correlating with
the presence of cognitive impairment (Fan et al. 2010). On
the other hand, we failed to detect a similar correlation
between subarachnoid space enlargement and cognitive
impairment. The discrepancies between severe phenotype
and the two measurements (CSF enlargement of ventricles
and subarachnoid spaces) might be due to the inadequacy
of the semiquantitative scores used for the subarachnoid
spaces enlargement.
Alternatively, the possible concomitance of the two
different conditions (i.e. diffuse enlargement of CSF spaces
in atrophy and variable and unpredictable enlargement of
subarachnoid spaces in communicating hydrocephalus)
might have interfered with the correlation, acting as a
confounding factor.
The pathogenesis of cerebral atrophy/communicating
hydrocephalus is controversial. According to some authors
the communicating hydrocephalus is determined by an
impaired CSF reabsorption due to deposits of GAG in the
Pacchionian granulations (Matheus et al. 2004). Other authors
postulated that abnormal morphology of the skull base may
hamper venous blood outflow, thus leading to hydrocephalus
(Vedolin et al. 2007b). On the other hand, atrophy could be
caused by the progressive accumulation of GAG in blood
vessels, leading to ischemic damage and consequent neuro-
nal death and gliosis (Lee et al. 1993). According to other
authors the direct intracellular storage of GAG is the cause of
brain parenchyma disease, which determines atrophy and the
ex-vacuo enlargement of CSF spaces (Parsons et al. 1996).
In summary, subarachnoid and ventricle CSF enlarge-
ment is a very frequent finding, its pathogenesis is still
poorly understood and the lack of reliable diagnostic
markers of communicating hydrocephalus necessitates
further studies.
Sella dysplasia, cranial hyperostosis, craniosynostosis
Pituitary sella abnormalities (“J” or “omega” shaped sella)
have been widely documented in patients affected by MPS
(Parsons et al. 1996). Our data confirm that the sellar region
is characteristically affected in Hunter disease as a j-shaped
sella was found in more than 70% of patients and 4 subjects
showed a significant enlargement of the sella turcica. The
latter feature did not correlate with the presence of severe
hydrocephalus and can therefore not be considered as a
marker of increased intracranial pressure.
Differently from sella abnormalities, cranial hyperostosis
is not considered a neuroradiological feature of Hunter
J Inherit Metab Dis (2011) 34:763–780 775

Table 4 MRI changes in 15

patients examined before and Unchanged Worsened Improved Not available
after enzyme replacement thera-
py introduction (one patient also Enlargement of Virchow-Robin PVSs inside CC 9/13 3/13 1*/13 2
underwent bone marrow trans- Enlargement of Virchow-Robin PVSs outside CC 13*/14 1/14 1
plantation) (median follow-up White matter signal abnormalities 8*/14 6/14 1
2.4 years; range 1–8.7 years;
median follow-up after ERT Subarachnoid CSF spaces enlargement 8*/14 6/14 1
beginning 1.6 years; range IIIrd ventricle enlargement 7*/14 7/14 1
1–2.9 years) Pituitary sella abnormalities 15*/15
Hyperostosis 14*/15 1/15
Craniosynostosis 15*/15
Platybasia 14*/14 1
Enlarged cisterna magna 15*/15
Chiari I malformation 15*/15
Small posterior fossa 15*/15
Dens dysplasia 12*/14 1/14 1/14 1
Periodontoid tissue thickening 10*/13 2/13 1/13 2
Spinal stenosis 11*/15 4/15
Myelopathy 14*/15 1/15
Atlanto-occipital assimilation 13*/13 2
PVSs: perivascular spaces; CC:
corpus callosum, CSF: cerebro- Basilar invagination 15*/15
spinal fluid; Platyspondylia 12*/12 3
*indicates the patient who un- Other vertebral anomalies 12*/12 3
derwent bone marrow transplan- Intervertebral disc abnormalities 10*/12 2/12 3
tation and subsequently ERT.

disease. Nonetheless, according to the scale proposed by
Avrahami (Avrahami et al. 1987), a cranial hyperostosis
was detected in nearly 20% of subjects, mostly in the severe
phenotype. Also craniosynostosis, namely scaphocephaly
and trigonocephaly, was detected in about 20% of cases.
These features appear to be more frequent in Hunter
patients compared to the general population (Hockley et
al. 1988). According to murine MPS I studies, accumulat-
ing GAG in growing cartilage is thought to cause a bone
maturation defect characterized by impaired osteoclast
activity and decreased cartilage reabsorption (Wilson et al.
2009). This may also explain the above mentioned skull
abnormalities. These features should therefore be consid-
ered as manifestations of the pathological process which
affects the whole skeletal system in MPS, as commonly
documented by X-ray examinations at level of claviculas,
ribs and tibias (Martin et al. 2008).
Posterior fossa abnormalities
Similarly to the pituitary region, the posterior fossa seems
to be often involved in Hunter patients. A megacisterna
magna was detected in nearly 40% of patients, while Chiari
I malformation or a small posterior fossa were detected in 6
and 8% of subjects, respectively.
Both abnormalities can be observed also in general
population though with a lower frequency. In a recent study
a megacisterna magna has been found in 0.2% of 681
children affected by headache (Schwedt et al. 2006), thus
underlying the markedly increased prevalence of this
feature in Hunter disease (Parsons et al. 1996; Seto et al.
2001). Even if the megacisterna magna can be considered
an enlargement of subarachnoid spaces, no correlation was
found between its presence and the enlargement of the
subarachnoid space and ventricles. Interestingly, its pres-
ence correlated with an attenuated phenotype, where
atrophy and hydrocephalus are less common.
Bejjani estimated the prevalence of Chiari I malforma-
tion in the general population between 0.5% and 5%
(Bejjani 2001) while the prevalence of a small posterior
fossa in normal subjects has not been reported. The
pathophysiological mechanism that leads to posterior fossa
abnormalities in Hunter disease is not known, nor is its
clinical consequence; in our cohort one patient presented
Chiari 1 malformation and subsequently developed an
extensive syringomyelia at the age of five. Longer follow up
studies may help to clarify their potential prognostic signif-
icance and possible complications such as syringomyelia or
CSF reabsorption defects.
Cranio-cervical junction abnormalities
Several cranio-cervical junction abnormalities have been
described in MPS, such as dens displasia, atlanto-axial
instability or subluxation, periodontoid tissue thickening,
spinal stenosis and compressive cervical myelopathy (Finn
776 J Inherit Metab Dis (2011) 34:763–780

Table 5 MRI changes in 6 patients without ERT and in five patients on ERT

Unchanged Worsened Improved Not available

Gr. 1 Gr. 2 Gr. 1 Gr. 2 Gr. 1 Gr. 2 Gr. 1 Gr. 2

Enlargement of Virchow-Robin PVSs inside CC 6**/6 4/5 1/5

Enlargement of Virchow-Robin PVSs outside CC 4*/6 5/5 1/6 1*/6
White matter signal abnormalities 5**/6 3/5 1/6 2/5
Subarachnoid CSF spaces enlargement 2/6 3/5 4**/6 1/5 1/5
IIIrd ventricle enlargement 2/6 2/5 4**/6 3/5
Pituitary sella abnormalities 4/6 5/5 2**/6
Hyperostosis 4/6 5/5 2**/6
Craniosynostosis 6**/6 5/5
Platybasia 4/4 5/5 2**
Enlarged cisterna magna 6**/6 4/5 1/5
Chiari I malformation 6**/6 5/5
Small posterior fossa 6**/6 5/5
Dens dysplasia 5*/6 4/5 1*/6 1/5
Periodontoid tissue thickening 4*/6 5/5 2*/6
Spinal stenosis 4*/6 3/5 2*/6 2/5
Myelopathy 4/4 5/5 2**
Atlanto-occipital assimilation 5*/5 5/5 1*
Basilar invagination 6**/6 5/5
Platyspondylia 4/4 5/5 2**
Other vertebral anomalies 4/4 3/4 1/4 2** 1
Intervertebral disc abnormalities 4/4 2/4 1/4 1/4 2** 1

PVSs: perivascular spaces; CC: corpus callosum, CSF: cerebro-spinal fluid

* indicates the number of patients who underwent bone marrow transplantation (* one patient, ** two patients).
Gr.1: group 1 (six patients, median follow-up 2.5 years; range 1.2–17 years); Gr.2: group 2 (5 patients, median follow-up 2.0 years; range 1.5–3.6 years).

et al. 2008). These abnormalities represent a critical aspect
in MPS, since lesion or compression of neural structures at
the bulbar-spine junction may lead to breathing centres
sufferance. Quadriparesis and respiratory arrest by trivial
cervical trauma have been reported in patients with pre-
existent spinal stenosis (Lipson 1977; Hite et al. 2000).
Cervical stenosis might also cause urinary retention and
sensory disturbances (Koyama et al. 1994; Al Sawaf et al.
2008). Cranio-cervical junction is thought to be less
severely affected in Hunter disease than in other forms of
MPS, such as the Morquio (MPS IV) and Maroteaux-Lamy
(MPS VI) syndromes. However, decompression surgery has
been reported also in Hunter patients (O’Brien et al. 1997).
Although no patient had signal abnormalities due to spinal
cord compression, spinal stenosis was found in nearly half
of our cases, thus underscoring the need of a careful
examination of this region.
Similarly, even if the odontoid process is reported to be
“well-formed” in most Hunter patients (Wraith et al. 2008),
about 65% of our patients presented with dens hypoplasia .
Since the ossification of the tip of the dens should be
completed at late infancy we cannot ascertain if dens peg
MR abnormalities in young subjects are due to a delayed
maturation or to an irreversible defect. The inverse
correlation between age and the presence of dens hypopla-
sia as well as the follow-up data (see below) seem to
suggest that this feature might improve over time.
Cranio-cervical junction abnormalities may be due to
several pathogenic mechanism. Thickening of the dura and
hypertrophy of the paraspinal ligaments are thought to be
due to progressive storage of GAG (Young et al. 1980;
Taccone et al. 1993; O’Brien et al. 1997; Kachur and Del
Maestro 2000; Seto et al. 2001). In addition, chronic
inflammation due to incomplete odontoid peg development
and concomitant atlanto-axial instability might cause a
progressive thickening of periodontoid soft tissue (Stevens
et al. 1991; Parsons et al. 1996; Hite et al. 2000).
Pathological specimens have revealed the prevalence of
fibrocartilaginous granulation tissue in the periodontoid
tissue of MPS IV (Stevens et al. 1991) while dural balloon
cells due to GAG abnormal storage have been found in
MPS VI (Young et al. 1980) No pathological findings are
J Inherit Metab Dis (2011) 34:763–780
Table 6 MRI sequence
and brain/spine MRI features MRI sequence
best visualized with
Brain FLAIR axial images
Brain/cervical sagittal T1
Brain/cervical sagittal T2
PVSs: perivascular spaces;
FLAIR: Fluid Attenuated Inver-
sion Recovery; CSF:
cerebro-spinal
available in Hunter patients. In our study we did not find
any correlation between the severity of dens hypoplasia and
the thickening measured at the odontoid peg. In fact, spinal
cord compression was observed even without dens hypo-
plasia, thus revealing that atlanto-axial instability is not the
sole determinant of the thickening at the odontoid peg (see
Fig 8d). On the other hand, ligamentous laxity is not
evaluated by standard MRI and a definitive conclusion
might require specific evaluations of the atlanto-axial
instability by plain flexion and extension cervical spine
radiographs.
Vertebral and disk alterations
Multiple bone and joint involvement due to GAG deposits
in growing cartilage and connective tissue is a common
finding in Hunter disease. Spine is not exempt and
morphologic abnormalities of vertebral bodies and inter-
somatic discs are often seen on neuroradiological examina-
tions (Parsons et al. 1996). Eighty-five per cent of our
patients had platyspondylia with typical bullet shaped
vertebral bodies. This appearance is normal in children
until the age of 12–14 months. In Hunter disease it
probably represents a halting of the normal maturation of
the vertebral body. Since the growth of vertebral body
777
Brain and spine MR features
Enlargement of Virchow-Robin PVSs outside corpus callosum
White matter signal abnormalities
Subarachnoid CSF spaces enlargement
IIIrd ventricle enlargement
Hyperostosis
Craniosynostosis
Enlargement of Virchow-Robin PVSs inside corpus callosum
Pituitary sella abnormalities
Hyperostosis
Enlarged cisterna magna, Chiari I malformation, Small posterior fossa
Dens dysplasia
Periodontoid tissue thickening
Spinal stenosis
Platyspondylia
Enlargement of Virchow-Robin PVSs inside corpus callosum
Pituitary sella abnormalities
Hyperostosis
Enlarged cisterna magna, Chiari I malformation, Small posterior fossa
Dens dysplasia
Periodontoid tissue thickening
Spinal stenosis
Myelopathy
Platyspondylia
Intervertebral disc abnormalities
mostly depends on the superior and inferior cartilaginous
plates, platyspondylia is likely due to the effects of
deposition of GAG in the growing cartilage, which
interferes with normal ossification processes, as proposed
for the pituitary and skull abnormalities. The involvement
of the cartilaginous plates might also explain the inter-
somatic disc abnormalities, i.e. height reduction or partial
absence, as observed in nearly 80% of patients.
Vertebral and disk changes may result in gibbus
formation, shortness of the neck, and restricted mobility
as well as in a reduced height.
Time course of MR features
The longitudinal analysis revealed that brain atrophy and/or
communicating hydrocephalus and WMA burden were the
brain MRI features soften showing progression during the
course of the disease, especially in the severe phenotype
(Fig 11). During a mean life-span of 4 years 9 out of 19
patients had a further enlargement of subarachnoid spaces,
eight out of 19 showed a progression of third ventricle
enlargement and seven out of 19 showed an increase of
WMA burden. These brain MRI changes reflect the well
known neurological deterioration. Notably, several severe
Hunter patients had, already at first examination, the
778
Fig. 11 MRI changes during
follow up. Dramatic ventricle
enlargement in a 3-year-old child
in 1-year span from score 2 to 3
(a); white matter signal abnor-
malities worsening from score 2
at the age of 20 months to score 3
at the age of 6 years (b)
highest obtainable scores for these abnormalities. There-
fore, these figures likely underestimate the real rate of
progression. In addition, a progression of MRI abnormal-
ities was also observed in the attenuated phenotype, where
a neurological involvement was not to be expected. This
fact supports the opinion that a dichotomic classification in
“severe” and “mild” phenotypes is a gross simplification
(Wraith et al. 2008) and does not fit with a disease where
the clinical manifestations are distributed in a continuous
spectrum. For this reason, a more detailed scoring system of
disease severity is probably needed in order to better represent
the complexity of the phenotype and the variations occurring
during disease progression.
Our findings allow some further considerations. Accord-
ing to previous studies (Lee et al. 1993) WMAs are thought
to appear subsequently to Virchow-Robin PVSs enlarge-
ment. WMAs were found to progress during the follow-up
in nearly half of our patients and showed an evolution
which did not correlate with presence, severity and
evolution of PVSs enlargement. Therefore, these two MRI
features, even though both very common in Hunter disease,
do present an independent course and likely recognize a
different pathogenesis.
A recent large cross-sectional study on patients affected
by different subtypes of MPS reported no correlation
between disease duration and white matter abnormalities
J Inherit Metab Dis (2011) 34:763–780
or severity of atrophy (Vedolin et al. 2007a, b). The cross-
sectional analysis of Hunter patients performed in our study
also did not find correlation between age and WMAs and
atrophy/communicating hydrocephalus severity, even when
patients were subgrouped according to the clinical pheno-
type. The contradiction between the longitudinal (progres-
sion of the disease) and the cross-sectional (no correlation
with age) analyses seem to reveal that disease duration and
clinical phenotype do not completely account for the above
mentioned abnormalities course and that other confounding
factors might act on the MR phenotype. Delayed or
impaired myelination, neuronal loss with atrophy and
wallerian axonal degeneration, ependymal or meningeal
barrier dysfunction and increased intracranial pressure
might concur to MRI white matter appearance and CSF
enlargement spectrum thus likely hindering the effect of
disease duration.
Quantitative measurements of CSF enlargement and
WMA burden or the use of new techniques such as
diffusion tensor and magnetization transfer imaging might
help to better understand the natural course of these MR
features.
Among the extracerebral MR features we observed a
progression of occipital hyperostosis, periodontoid tissue
thickening and spinal stenosis in three out of 20, three out of
19 and seven out of 20 patients, respectively. The latter
J Inherit Metab Dis (2011) 34:763–780
features further underscore the importance of a periodic
neuroradiological follow-up of the atlanto-occipital junction
in order to plan a proper decompressive surgery, when needed.
Interestingly, we did not find any correlation between the
presence of dens hypoplasia and the progression of spinal
stenosis, thus sustaining the hypothesis that the thickening
at the odontoid peg is not determined exclusively by the
presence of atlanto-axial instability. Moreover, dens hypo-
plasia may improve over time, so that this feature should
not be regarded as an unfavorable prognostic factor for the
development of spinal stenosis.
ERT effects in Hunter disease
ERT has proven to be effective in Hunter patients in terms of
organomegaly and joint stiffness reduction and in improving
quality of life (Wraith et al. 2008). Despite encouraging results
reported on two Hunter patients (Wang et al. 2009), the CNS
does not seem to benefit significantly from intravenous
enzyme administration likely because the enzyme does not
pass the blood-brain barrier (BBB) (Al Sawaf et al. 2008).
Even if our data are preliminary and based on few patients,
neuroradiological features seem to respond poorly to the
treatment since brain MRI features seem to worsen despite
ERT introduction. Interestingly, even those tissues (e.g.
pituitary sella, Virchow-Robin PVSs, periodontoid thicken-
ing, platyspondylia, discs abnormalities) which are not
beyond the BBB did not show a response to ERT. These
findings seem to indicate that GAG mediated damages at the
level of the structures without BBB (e.g. bone and meninges)
might be difficult to revert once they have occurred. These
data therefore suggest the need of an early ERT introduction
in order to prevent irreversible tissue changes.
MRI protocol
Hunter patients frequently require sedation when undergo-
ing MRI examination. The important anaesthetic risks
correlated to the disease itself impose to perform these
studies in centres staffed with anaesthesiologists experi-
enced with these disorders (Wraith et al. 2008). Further-
more, the MRI protocol should be kept at minimum for
monitoring the most significant abovementioned MR
features. According to our experience brain FLAIR axial
images and brain/cervical sagittal T1 and T2 thin images
provide all the required information in a time span shorter
than 15 minutes (Table 6).
Conclusion
In conclusion, a number of neuroradiological brain, spinal
cord, skull and vertebral MRI features must be considered
779
in MPS. WMAs, atrophy and/or communicating hydro-
cephalus and cranial vault hyperostosis correlate with the
Hunter disease severe phenotype while an enlarged cisterna
magna is more frequent in the attenuated phenotype.
ERT introduction does not seem to modify progression
of WMAs, atrophy/communicating hydrocephalus and
spinal stenosis, even though data are very preliminary and
a longer follow-up is needed. Therefore, these features have
to be strictly monitored both in routine practise and when
evaluating new therapies response. A thorough evaluation
of MRI exams, especially if performed in a centralized
system, will increase the knowledge of the natural and post-
treatment course of the CNS involvement in Hunter patients
and will help outline the differences from other forms of
MPS. A scale that considers WMAs, atrophy/communicat-
ing hydrocephalus, dens hypoplasia, periodontoid tissue
thickening, spinal stenosis and myelopathy might be a valid
tool for monitoring Hunter patients, while other MR
features seem to be less useful.
Acknowledgment We acknowledge Fondazione Pierfranco e Luisa
Mariani for their consistent economic support to the metabolic Center
in Monza.
We acknowledge Frits Wijburg and Chiara Briani for their critical
support.
References
Al Sawaf S, Mayatepek E, Hoffmann B (2008) Neurological findings
in Hunter disease: pathology and possible therapeutic effects
reviewed. J Inherit Metab Dis 31:473–480
Avrahami E, Katz A, Bornstein N, Korczyn AD (1987) Computed
tomographic findings of brain and skull in myotonic dystrophy. J
Neurol Neurosurg Psychiatry 50(4):435–438
Barone R, Nigro F, Triulzi F, Musumeci S, Fiumara A, Pavone L
(1999) Clinical and neuroradiological follow-up in mucopoly-
saccharidosis type III (Sanfilippo syndrome). Neuropediatrics 30
(5):270–274
Beck M, Cole G (1984) Disc oedema in association with Hunter’s
syndrome: ocular histopathological findings. Br J Ophthalmol
68:590–594
Bejjani GK (2001) Definition of the adult Chiari malformation: a brief
historical overview. Neurosurg Focus 11(1):E1
Faerber EN, Poussaint TY (2002) Magnetic resonance of metabolic
and degenerative diseases in children. Top Magn Reson Imaging
13(1):3–21
Fan Z, Styner M, Muenzer J, Poe M, Escolare M (2010) Correlation of
Automated Volumetric Analysis of Brain MR Imaging with
Cognitive Impairment in a Natural History Study of Mucopoly-
saccharidosis II. AJNR Am J Neuroradiol;[Epub ahead of print].
Fazekas F, Chawluk J, Alavi A, Hurtig H, Zimmerman R (1987) MRI
signal abnormalities at 1.5T in Alzheimer’s dementia and normal
aging. AJNR Am J Neuroradiol 8:421–426
Finn CT, Vedolin L, Schwartz IV et al (2008) Magnetic resonance
imaging findings in Hunter syndrome. Acta Paediatr Suppl 97
(457):61–68
Gabrielli O, Polonara G, Regnicolo L et al (2004) Correlation between
cerebral MRI abormalities and mental retardation in patients with
mucopolysaccharidosis. Am J Med Genet 125:224–231
780
Groeschel S, Chong WK, Surtees R, Hanefeld F (2006) Virchow-
Robin spaces on magnetic resonance images: normative data,
their dilatation, and a review of the literature. Neuroradiology
48:745–754
Hite SH, Peters C, Krivit W (2000) Correction of odontoid dysplasia
following bone-marrow transplantation and engraftment (in
Hurler syndrome MPS 1H). Pediatr Radiol 30(7):464–470
Hockley AD, Wake MJ, Goldin H (1988) Surgical management of
Craniosynostosis. Br J Neurosurg 2(3):307–313
Kachur E, Del Maestro R (2000) Mucopolysaccharidoses and spinal
cord compression: case report and review of the literature with
implications of bone marrow transplantation. Neurosurgery 47
(1):223–229
Koenigsberg RA, Vakil N, Hong TA et al (2005) Evaluation of
platybasia with MR imaging. AJNR Am J Neuroradiol 26(1):89–
92
Koyama K, Moda Y, Sone A, Tanaka H, Hino Y (1994) Neurogenic
bladder in Hunter's syndrome. J Med Genet 31(3):257–258
Lee C, Dineen TE, Brack M, Kirsch JE, Runge VM (1993) The
mucopolysaccharidoses: characterization by cranial MR imaging.
Am J Neuroradiol 4:1285–1292
Lipson SJ (1977) Dysplasia of the odontoid process in Morquio's
syndrome causing quadriparesis. J Bone Joint Surg Am 59
(3):340–344
Martin R, Beck M, Eng C et al (2008) Recognition and Diagnosis of
Mucopolysaccharidosis II (Hunter Syndrome). Pediatrics 121:
e377–e386
Matheus MG, Castillo M, Smith JK, Armao D, Towle D, Muenzer J
(2004) Brain MRI findings in patients with mucopolysaccharidosis
types I and II and mild clinical presentation. Neuroradiology
46:666–672
Muenzer J, Beck M, Eng CM et al (2009) Multidisciplinary
management of Hunter syndrome. Pediatrics 124(6):e1228–
e1239, Epub 2009 Nov 9. Review
Müller-Forell W, Frenking GS, Amraoui Y, Beck M (2007) Mucopo-
lysaccharidoses (MPS) clinical and neuroradiological aspects of
the different types. Clin Neuroradiol 17(3):141–158
O’Brien DP, Cowie RA, Wraith JE (1997) Cervical decompression in
mild mucopolysaccharidosis type II (Hunter syndrome). Childs
Nerv Syst 13:87–90
Parsons VJ, Hughes DG, Wraith JE (1996) Magnetic resonance
imaging of the brain, neck and cervical spine in mild Hunter’s
syndrome (mucopolysaccharidosis type II). Clin Radiol 51:719–
723
Scheltens P, Barkhof F, Leys D (1993) A semiquantitative rating scale
for the assessment of signal hyperintensities on magnetic
resonance imaging. J Neurol Sci 114:7–12
J Inherit Metab Dis (2011) 34:763–780
Schwedt TJ, Guo Y, Rothner AD (2006) "Benign" imaging abnor-
malities in children and adolescents with headache. Headache 46
(3):387–398
Seto T, Kono K, Morimoto K (2001) Brain magnetic resonance
imaging in 23 patients with mucopolysaccharidoses and the
effect of bone marrow transplantation. Ann Neurol 50:79–91
Shimoda-Matsubayashi S, Kuru Y, Sumie H, Ito T, Hattori N, Okuma
Y (1990) MRI findings in the mild type of mucopolysacchar-
idosis. II (Hunter’s syndrome). Neuroradiology 32:328–330
Shinomiya N, Nagayama T, Fujioka Y, Aoki T (1996) MRI in the mild
type of mucopolysaccharidosis II (Hunter's syndrome). Neurora-
diology 38(5):483–485
Stevens JM, Kendall BE, Crockard HA, Ransford A (1991) The
odontoid process in Morquio-Brailsford's disease. The effects of
occipitocervical fusion. J Bone Jt Surg Br 73(5):851–858
Taccone A, Tortori Donati P, Marzoli A, Dell'Acqua A, Gatti R, Leone
D (1993) Mucopolysaccharidosis: thickening of dura mater at the
craniocervical junction and other CT/MRI findings. Pediatr
Radiol 23(5):349–352
Tierney RT, Maldjian C, Mattacola CG, Straub SJ, Sitler MR (2002)
Cervical spine stenosis measures in normal subjects. J Athl Train
37(2):190–193
Vedolin L, Schwartz IV, Komlos M et al (2007a) Correlation of MR
imaging and MR spectroscopy findings with cognitive impair-
ment in mucopolysaccharidosis II. AJNR Am J Neuroradiol 28
(6):1029–1033(a)
Vedolin L, Schwartz IV, Komlos M, Schuch A, Azevedo AC, Vieira T
(2007b) Brain MRI in mucopolysaccharidosis: effect of aging
and correlation with biochemical findings. Neurology 69:917–
924(b)
Wahlund L, Barkhof F, Fazekas F (2001) A new rating scale for age-
related white matter changes applicable to MRI and CT. Stroke
32:1318–1322
Wang RY, Cambray-Forker EJ, Ohanian K et al (2009) Treatment
reduces or stabilizes brain imaging abnormalities in patients with
MPS I and II. Mol Genet Metab 98(4):406–411
Wilson S, Hashamiyan S, Clarke L et al (2009) Glycosaminoglycan-
mediated loss of cathepsin K collagenolytic activity in MPS I
contributes to osteoclast and growth plate abnormalities. Am J
Pathol 175(5):2053–2062
Wraith JE, Scarpa M, Beck M et al (2008) Mucopolysaccharidosis
type II (Hunter syndrome): a clinical review and recommenda-
tions for treatment in the era of enzyme replacement therapy. Eur
J Pediatr 167(3):267–277
Young R, Kleinman G, Ojemann RG et al (1980) Compressive
myelopathy in Maroteaux-Lamy syndrome: clinical and patho-
logical findings. Ann Neurol 8(3):336–340