Journal of Neurology, Neurosurgery, and Psychiatry 1984;47: 704-709

Hemi-dystonia secondary to localised basal ganglia
tumour
J NARBONA,* JA OBESO,* T TUNON, JM MARTINEZ-LAGE,* CD MARSDEN.t
From the Departments of Neurology (Neuropediatric and Movement Disorders Unit)* and Pathology,
Clinica Universitaria, University of Navarra Medical School, Pamplona, Spain, and University Department of
Neurology, Institute of Psychiatry and King's College Medical School, London, England. t
SUMMARY An 8-year-old boy with an 18 month history of left limb hemi-dystonia due to a right
lenticular nucleus astrocytoma originating in the putamen is reported. Subsequent
neuropathological study demonstrated that the tumour was mainly localised to the right lenticular
nucleus, with cystic necrosis in the infero-lateral putamen. Solid tumour also infiltrated the right
hypothalamus, the anterior commisure and the optic chiasm, and there was perivascular spread
into the globus pallidus, internal capsule and roof of the right lateral ventricle. This case, and the
few other published reports of symptomatic dystonia due to focal brain lesions verified pathologically, indicate that damage to the lenticular nucleus, and to the putamen in particular, can cause
limb dystonia in man.

The illness commenced with the left hand adopting
bizarre, unwanted postures on attempted action. Gradually
all manual activity became distorted by such involuntary
movements, and drawing with the left hand became
increasingly difficult. Three months after the onset of
symptoms in the left hand, the left foot also became
affected, beginning to plantar-flex and invert on walking.
On examination in August 1979, the cranial nerves and
right limbs were normal. Any slight body movement provoked the left arm to adopt a dystonic posture, with the
wrist assuming a posture of ulnar deviation and slight
flexion with the fingers extended. The outstretched left arm
also adopted a similar posture, with additional excessive
pronation. Dystonia prevented any useful action of the left
arm. At rest the left foot was inverted and plantar flexed,
with a tendency for the left big toe to adopt a spontaneous
dystonic extension. This dystonic posture of the left foot
was exaggerated when he tried to walk. Passive movement
of wrist or ankle provoked a springy and variable resistance. Muscle power was normal, the tendon reflexes were
symmetrical and normal, both plantar responses were
flexor, and there was no defect of cutaneous or deep
sensation. General examination was normal.
A CT scan in September 1979 revealed a high density
lesion with an area of low attenuation within its anterior
extent, which enhanced with contrast, in the external part
of the right lenticular nucleus centred in the region of the
putamen (fig 1). There was no displacement of the ventricular system. The following tests were reported as normal: EEG, skull and chest radiographs, copper metabolism, VDRL, liver and renal function, and blood count.
Plasma levels of glucose, urea, uric acid, cholesterol, calcium, phosphorus, and phospholipids also were normal.
The boy was treated sequentially with haloperidol (3

Torsion dystonia is a clinical syndrome characterised
by prolonged spasms of muscle contractions, which
distort the body into typical dystonic postures. The
pathological or biochemical basis of either the familial types or of most sporadic forms of idiopathic or
primary torsion dystonia is unknown.' There are
numerous reports of patients with symptomatic or
secondary dystonia but such dystonia usually has
been associated with widespread pathological
lesions in the brain.2 Cases of pure focal dystonia
due to a localised brain lesion with pathological
confirmation are rare. We report the clinical features and pathological findings in a patient with
hemi-dystonia secondary to an astrocytoma originating in the putamen.
Case report
An 8-year-old right-handed boy presented in August 1979
with a 4 month history of abnormal postures and involuntary movements of the left limbs. He was the product of a
normal pregnancy and delivery, and his early development
and general health had been excellent until this recent illness. His parents were healthy and unrelated, and there
was no history of any similar illness in any other member of
the family.
Address for reprint request: Dr JA Obeso, Neurologia, Clfnica
Universitaria, Apartado 192. Pamplona, Spain.

Received 28 November 1983. Accepted 5 January 1984

704

Hemi-dystonia secondary to localised basal ganglia tumour

705

Fig 1 Horizontal CT scans after intravenous contrast injection. (a) Below the level of the third ventricle
(no abnormality visible). (b) At level of third ventricle, showing a high density area, with a zone of low
attenuation anteriorly in right putaminal area. (c) 12 mm above b (no clear abnormality visible). (d)
Enlarged view of b. All scans were performed before radiotherapy or methotrexate treatment. The visible
lesion corresponds in location to the area of maximum necrosis seen at necropsy (see fig 2). Note also that
at the time of CT scans there was no hydrocephalus.
mg/day), benzhexol (25 mg/day), chlorpromazine (75 mg/
day), and levodopa/carbidopa (750-75/day), without benefit. Six months after the onset of symptoms, he
developed intense polyuria and polydipsia, drinking up to
three litres of fluid each day. Diabetes insipidus was
confirmed by measurement of plasma and urinary osmolality, and treatment with carbamazepine (300 mg daily)
decreased the polyuria and polydipsia. This treatment also
moderately improved the hemi-dystonia (30% over control disability score). Two months later, however, the left
arm became clumsier and walking was even more difficult.
Examination at that time showed that he had signs in the

left limbs of mild corticomotoneuron damage in addition to
the hemi-dystonia. Examination of the cerebrospinal fluid
now revealed neoplastic cells of uncertain type. At this
point it was decided to begin a course of radiotherapy.
5000 rads directed to the right striatum were given in 25
sessions over five weeks. The left sided weakness subsequently resolved and there was some slight (10%
improvement in the hemi-dystonia). There was no deterioration over the next eight months, but then, 18 months
after the onset of the illness, the hemi-dystonia worsened
and the left hemi-paresis recurred. Intrathecal methotrexate was commenced, but four days later he died of mas-

Narbona, Obeso, Tunion, Martinez-Lage, Marsden

706
ED N

5.t:

.:e.

.I
.:A.
Am-..

I

..-

)r
'lk-..%
Mr.

.4..

:-

.",Io
.*.6-

Fig 2 (a) Coronal section of the formalin fixed brain at the level of the anterior commissure. In the right putamen, in its
inferior-lateral region, there is a cystic area of necrosis. There is infiltration of tumour into both the external and internal
segments of the globus pallidus. There also is an area of haemorrhagic necrosis subependymally at the inferior angle of the
right lateral ventricle. There is hydrocephalus affecting not only the lateral ventricle, but also the anterior end of the third
ventricle. (b) Coronal section at the level ofthe mammillary bodies. The area of cystic necrosis in the infero-lateral region of
the putamen, and the surrounding tumour infiltration again is shown. In addition, there is obvious macroscopic tumour in
the right hypothalamus above the mammillary body. However, the subthalamus, thalamus, claustrum and caudate nucleus
are normal bilaterally.

showed the features of a low-grade astrocytoma, with only
moderate hypercellularity and only a few undifferentiated
cells (fig 3). The distribution of the tumour, as assessed in
serial coronal sections, is shown in fig 4. The area of cystic
necrosis was confined to the infero-lateral putamen. The
solid tumour was confined histologically within the basal
ganglia to the putamen, but extended perivascularly into
the globus pallidus (pars extema and interna). There also
was slight perivascular extension medially into the internal
capsule and dorsally into the subependymal zone of the
roof of the right lateral ventricle. Solid tumour was present

sive gastric haemorrhage.
Necropsy was done four hours after death. The brain
weighed 1530 g and looked normal in external appearance.
After formalin fixation, serial coronal slices (fig 2) revealed
slight hydrocephalus and an area of necrosis in the inferolateral segment of the right putamen. There were obvious
areas of tumour adjacent to the zone of necrosis in the
putamen, and also in the right globus pallidus and right
hypothalamus above the mammillary body. In addition,
there was a zone of subependymal haemorrhage at the
inferior angle of the right lateral ventricle. Histology

q A,.

*~~~~~~~j
4:

4

V~
\''
*-

^

4

t

z

Fig 3 Histology of the tumour. An area
t.4,_
W, 4.;~
of globus pallidus is shown to illustrate
both the cellular nature of the tumour
(astrocytoma), and its perivascular
distribution in otherwise normal brain in
this area. The scale marker is 100 ,u

Haematoxylin-eosin.

p,
#

&

t
A

Hemi-dystonia secondary to localised basal ganglia tumour

707

Fig 4 Histological distribution of the tumour shown in diagrams of sections of the brain. (a) Coronal at level of anterior
commissure, (b) Coronal at level of infundibulum, and (c) Horizontal at level of third ventricle. The solid white zone
indicates the area of cystic necrosis in the right putamen. The posterior halfof the putamen (also in white in the horizontal
diagram) was normal histologically. The solid black zone indicates the areas of brain replaced by solid tumour. The dotted
black zones indicate areas of brain containing perivascular tumour.

also in the hypothalamus, bilaterally, in the anterior commissure and in the optic chiasma. The thalamus, claustrum,
subthalamus, substantia nigra (zona compacta and

reticulata), superior colliculus, red nucleus and brain stem
were normal, as were the motor cortex and cerebellum,
both macroscopically and microscopically.

Discussion

Pathologically well-documented cases of focal brain
lesions causing focal or hemi-dystonia have been
described but rarely. Urechia et a13 published the
case of a 14-year-old boy with a tumour in the

708

putamen, caudate nucleus and globus pallidus.
Chromatolysis and fatty degeneration of the subthalamic nucleus also was observed. Another localised lesion producing focal dystonia was reported by
Lopez Aydillo and Sanz Ibafiez,4 who studied a
56-year-old female diabetic patient with dystonia of
the left arm of vascular origin secondary to complete
necrosis of the right putamen and head of the caudate nucleus, as well as demyelination of the globus
pallidus. Other cases of focal or hemi-dystonia
associated with local brain lesions have had more
widespread or complicated pathology. One of Sciarra and Sprofkin's5 patients (case No 7), a 15year-old girl, presented with dystonia of the left
upper limb and jerky movements of the toes. A
large glioblastoma invading the right basal ganglia
from the frontal lobe was found at necropsy. Gordin6 described a 23-year-old man who developed
torticollis with the chin deviated to the right, twisting of the trunk to the right with excessive lordosis
and hemi-dystonia of the left limbs. At necropsy,
extensive changes were found in the right caudate
nucleus, putamen and, to a lesser extent, the external segment of the globus pallidus. These consisted
of neuronal degeneration and glial proliferation,
with perivascular gliosis, and were attributed to "a
slight encephalitic affection". This diagnosis raises a
question of more diffuse pathology, although none
was described. In all of the remaining published
cases of symptomatic focal dystonia studied
pathologically that we have had the opportunity to
review in detail, the dystonia had emerged as a late
sequela of acute vascular hemiplegia,7- or was
accompanied by additional neurological signs from
its onset,5 10-12 indicating initial extensive brain

damage.
A number of cases of focal dystonia of one limb,
or of hemidystonia, have been described since the
advent of the CT scan, with lesions in the region of
the contralateral basal ganglia, either vascular'3-'5
or traumatic'6 in origin. However, the exact localisation and extent of such lesions is difficult to establish
by radiological techniques alone. Bilateral putaminal lesions also may be associated with dystonia. For
instance, such pathology is found in Wilson' s disease
and also occurs in Leigh's disease, in both of which
occasional patients may present with typical dystonia. However, the pathology in such cases
undoubtedly is more widespread, so that exact
clinico-pathological correlation is difficult.
The localisation of the tumour in our patient
allows accurate clinico-pathological correlation.
Typical dystonia on action and at rest, with mobile
spasms showing co-contraction of antagonist muscles and overflow of muscle activity, was present in
the left limbs for several months as the only

Narbona, Obeso, Tunon, Martinez-Lage, Marsden
neurological sign. Five months after the onset of the
illness, a CT scan revealed a focal lesion in the
external part of the lentiform nucleus, corresponding to the putamen. Even at the end of the illness, 18
months after the dystonia first began, the maximum
tumour necrosis was in the infero-lateral putamen.
By then, however, there had been tumour spread
into the globus pallidus, hypothalamus (with diabetes insipidus), optic chiasm, and internal capsule.
However, the lesion was centered in the inferolateral putamen, which, from all the evidence, seems
to have been its site of origin.
That a lesion of the putamen may cause contralateral hemidystonia is of considerable interest in the
light of present knowledge on the distribution of
cortical inputs into the caudate-putamen. The
sensori-motor cortex projects, in primates, to the
putamen, with some degree of somatotopic organisation.'7'-9 Kunzle,'7 for example, has identified
projections from motor cortex to putamen in the
monkey, with the leg area anterodorsally, the face
area postero-inferiorly, and the arm area intermediate between the two. There is some resemblance
between this organisation and the site of dystonia in
our patient and the site of the lesion at necropsy.
In the light of this case, and the few published
reports reviewed above, focal limb dystonia may be
regarded as secondary to a lesion of the lenticular
nucleus and probably the putamen in particular.
This is not to suggest that some other brain structures may not also be implicated in other cases of
dystonia in man. In all these cases of symptomatic
dystonia, dystonia may be regarded as a release
phenomenon due to disinhibition of some other nucleus or area within the basal ganglia or elsewhere.
The major outputs of the putamen are to both segments of the globus pallidus (which were damaged
in our patient) and to substantia nigra (zona
reticulata) which was spared. The subthalamic nucleus, thalamus, superior colliculus and brain stem,
all of which receive projections from the globus pallidus or substantia nigra zona reticulata, also were
intact. Which of these areas are involved by disinhibition as a result of the putamen lesion to cause the
dystonia cannot be decided. Whether or not a similar mechanism operates at the biochemical level to
account for the motor abnormalities present in
patients with idiopathic or primary torsion dystonia
is not known. However, cases such as those
reviewed here suggest that biochemical examination
of brains from patients with idiopathic dystonia
should include careful dissection of discrete areas of
the putamen and globus pallidus, and of their sites of

projection.
The authors are grateful to Dr JL Zubieta

Hemi-dystonia secondary to localised basal ganglia tumour
709
(Neuroradiology Unit, Clinica Universitaria) for his 10 Garcin R. Syndrome cerebello-thalamique par lesion
expert comments and Mrs M Obeso for skilful assislocalisee du thalamus. Rev Neurol (Paris)
tance.
1955;93: 143-9.
" Austregesilio A, Borges-Forte A. Sur un cas

References

Zeman W. Pathology of the torsion dystonias (dystonia
musculorum deformans). Neurology, 20 (part 2)
1970: 79-88.
2 Zeman W, Whitlock CC. Symptomatic dystonias. In
Vinken PJ, Bruyn GW eds.), Handbook ofNeurology
Amsterdam: Elsevier, 1968, Vol. 6. 545-63.
3 Urechia CI, Dragomir L, Usinievici G. Spasme de torsion unilateral cause par une tumeur cerebrale. Conf
Neurologica 1942-3;5:271-80.
4 L6pez Aydillo NR, Sanz Ibafiez J. A prop6sito de un
caso de distonia de torsi6n (variante miostatica o paralIftica) en una diabetica glucosurica. Trabajos Inst
Cajal Inv Biol 1956;48:81-108.
5 Sciarra F, Sprofkin B. Symptoms and signs referable to
the basal ganglia in brain tumour. Arch Neurol

Psychiat 1953;69:450-61.

6 Gordin R. A case of unilateral torsion dystonia: A

clinico-histological study. J Nerv Ment Dis
1939;90:344-57.
Denny-Brown D. The Basal Ganglia. London: Oxford
University Press, 1962:55-6.
Oppenheimer DR. A case of striatal hemiplegia. J
Neurol Neurosurg Psychiatry 1967;30:134-9.
Dooling EC, Adams RD. The pathological anatomy of
post-hemiplegic athetosis. Brain 1975;98:29-48.

d'hemichoree avec lesion du moyau caude. Rev
Neurol (Paris) 1937;67:477-88.
12 Chorobski J. Involuntary movements in patients with
intracranial tumours. Arch Neurol 1962;6: 27-56.
3 Demierre B, Rondot P. Dystonia caused by putaminocapsulo-caudate vascular lesions. J Neurol Neurosurg

Psychiatry 1983;46:404-9.

4 Russo LS. Focal dystonia and lacunar infarction of the

basal ganglia. Arch Neurol 1983;40:61-2.
'5 Traub M, Ridley A. Focal dystonia in association with
cerebral infarction. J Neurol Neurosurg Psychiatry
1982;45: 1073-4.
16 Brett EM, Hoare RD, Sheehy MP, Marsden CD. Progressive hemi-dystonia due to focal basal ganglia
lesion after mild head trauma. J Neurol Neurosurg
Psychiatry 1981;44:460.
71 Kunzle H. Bilateral projections from precentral motor
cortex to the putamen and other parts of the basal
ganglia. An autoradiographic study in Macaca
Fasicularis. Brain Res 1975;88: 195-209.
18 Kunzle H. Projections from the primary somatosensory
cortex to basal ganglia and thalamus in the monkey.
Exp Brain Res 1977;30:481-92.
'9 Jones EG, Coulter JD, Buxton H et al. Cells of origin
and terminal distribution of cortico-striatal fibres arising in the sensory-motor cortex of monkeys. J Comp
Neurol 1977;173:53-80.