96

COMMUNICATIONS

Upper-Limb Dystonia Secondary to a
Midbrain Hemorrhage
To the Editor:
Focal dystonia limited to the hand or arm has been
described in patients with basal ganglia lesions (1,2).
Brain stem lesions have been associated with cervicocranial dystonia in a number of cases (3,4) and, on exceptional occasions, with contralateral hemidystonia (5,6).
Unilateral asterixis also has been observed in focal brain
lesions, but bilateral asterixis is generally a manifestation
of diffuse encephalopathy (7). We report a patient with
left upper limb dystonia of acute onset and bilateral asterixis related with a left brain stem hemorrhage.

Case Report
A 46-year-old woman, previously in good health, was
admitted to our hospital because she noted sudden dizziness and weakness of the right limbs. When she arrived at
the hospital, she was conscious, dysarthric, had mild hypophonia and referred vertical and horizontal diplopia.
Her pupils were 2-3 mm in diameter, symmetric, and
poorly reactive to light. Spontaneous gaze resulted in a
convergent strabismus with an alternant ocular skew deviation. Saccadic and pursuit upward eye movements
were absent, and voluntary downward movements were
markedly reduced but improved with oculocephalic maneuvers. She had a bilateral pseudesixth nerve palsy
with convergence spasms. Optokinetic nystagmus was
normal. There was a discrete right central facial and right
upper limb paresis. Tendon reflexes were brisk and symmetric, but Babinski sign was not present. Touch, pain,
and thermal senses were diminished on the right side of
the face, the right upper limb, and the right upper area of
the trunk. No sensory abnormalities, including proprioceptive and vibration senses, were present on the left
limbs. She had a left cerebellar syndrome with fingernose and heel-knee-shin ataxia, upper limb intention
tremor, and gait instability.
A left upper limb abnormal posture was observed at
rest, which was more evident when the patient held the
arms outstretched. In this position, there was ulnar deviation of the hand, hyperextension of the fingers with abduction of the fourth and fifth fingers, and supination of
the arm. This dystonic posture was strikingly aggravated
by some voluntary movements. The hand then showed an
unwanted posture with abduction and flexion of the
metacarpophalangeal joints of the fourth and fifth fingers,
accompanied by adduction of the thumb and flexion of
the interphalangeal joints of the second and third fingers.
Moreover, there was ulnar deviation of the hand and, at
times, hyperflexion of the wrist. Bilateral asterixis, more
prominent on the right arm, but also present in the proximal and distal muscles, was observed during the first few
days.
Routine laboratory tests, including toxicologic screening, chest radiogram, and electrocardiogram were normal. Cranial computed tomography (CT) showed a hem-

Movement Disorders, Vol. 1 1 , No. I , 1996

orrhagic lesion in the left midbrain and cranial magnetic
resonance imaging (MRI) evidenced that this lesion extended from the left pretectal region of the midbrain to the
left superior cerebellar peduncle, with ipsilateral red nuclei displacement (Fig. 1). Arteriographic study failed to
demonstrate a vascular malformation. Median nerve
evoked potentials recorded on the scalp were normal bilaterally. Electromyographic (EMG) recordings showed a
characteristic dystonic pattern with cocontraction of agonists and antagonist muscles in the left forearm and
sometimes in the left arm, and occasionally irregular periods of electrical silence occurring simultaneously in the
flexors and extensors muscles of both forearms.
The patient’s clinical evolution was relatively satisfactory. Asterixis diminished dramatically within a few days.
A week later, she was able to walk without aid and the
cerebellar syndrome improved. Five months later the left
dystonia had also improved. Ocular abnormalities were
unchanged. MRI showed only a residual deposit of hemosiderin in the midbrain (Fig. 2 ) .

Discussion
Our patient developed upper-limb dystonia of acute onset in association with an ipsilateral brain stem lesion. We
have not found any report in the literature indicating such
an association. No motor or sensory abnormalities were
present in the dystonic limb, and MRI showed a midbrain
hemorrhage but no other structural central nervous system lesions. In patients with secondary isolated focal dystonia of the hand and arm, Marsden et al. ( 2 ) found that
the responsible lesion causing dystonia mainly affected
the contralateral thalamus, but caudate or lentiform nucleus were involved in some cases. Reports of secondary
limb dystonia occurring in patients with brain stem lesions are exceptional. Leenders et al. (5) published a case
of ipsilateral blepharospasm, contralateral hemidystonia,
and parkinsonism in a patient with a mesencephalothalamic lesion, and Krauss et al. (6) reported a patient with
contralateral hemidystonia and torticollis secondary to a
traumatic pontomesencephalic lesion.
Dystonia in patients with lesions in the basal ganglia
may be caused by an abnormal input from the thalamus to
premotor cortex, including the supplementary motor
area, from a lesion either in the thalamus itself or in the
striatum ( 2 ) . An explanation for the pathophysiology of
the limb dystonia is difficult in our case, in which no
lesions were identified in the basal ganglia and no sensory
abnormalities were present in the dystonic limb to explain
dystonia on the basis of deafferentation. The left cerebellar syndrome ipsilateral to the dystonia suggests involvement of cerebellar pathways that could have contributed
to the development of dystonia in our patient. In previously reported cases of hemidystonia with brain stem lesions, patients had lesions situated at a level that was
higher or lower than the lesion in our case; however, the
dystonia was contralateral to the lesion. In our case, disruption by the hematoma of the left dentato-rubrothalamic pathway before the decussation of the brachium
conjunctivum at the level of the inferior colliculli could
explain why the cerebellar syndrome and dystonia were
ipsilateral to the lesion. Classically, lesions involving this

COMM UNICATIONS

97

FIG. 1. A: T2-weighted MRI showing a recent hemorrhagic lesion in the left midbrain with mass effect over the ipsilateral red nucleus.
B and C: T1-weighted MRIs showing the hematoma in the coronal and sagittal planes, respectively.

fascicle or cerebellum are not associated with dystonia.
However, as Hedreen et al. (8) suggested, it is possible
that a lesion involving a single population of cerebellar
neurons could produce dystonia. Disconnection between
the cerebellum and thalamus offers a reasonable explanation for the pathophysiology of the dystonia in our case,
but other possibilities may be considered. It has been
proposed that limb dystonia in the brain stem (6) or spinal
cord lesions (9) may be produced by lesions involving the
descending inhibitory supraspinal pathways from reticulospinal tracts. Such a mechanism also could have played
a role in our case. Although reticulospinal tracts originate
in the central pontine and dorsal and central pars of the
medula (medial reticulospinal tract) and ventrolateral
pontine tegmentum (lateral reticulospinal tract) (lo), always ventral and caudal to the hemorrhagic area in our
patient, we cannot rule out that these nuclei were not
receiving an abnormal input from other structures such as
the basal ganglia, thalamus, or cerebellum due to the
hemorrhage.
In experimental animals, brain stem lesions involving

red nucleus, brachium conjunctivum, portions of the medial reticular formation, and medial longitudinal fasciculus can produce spasmodic torticollis and other facial dyskinesias (4,l I ) . Other investigators (12) have found that
glutamic acid decarboxylase activity is elevated in the
deep cerebellar nuclei of the rat mutant dystonic, supporting the view that cerebellar dysfunction may be a
primary component of the dystonia in these animals.
There is some evidence suggesting that the cerebellum or
its projections could be involved in human dystonic disorders. Electrical stimulation in the region of the interpositus and dentate nuclei of the cerebellum resulted in
abnormal tonic postures, whereas lesions made in the medial dentate region improved athetosis, tremor, or spasticity in patients (13). Several cases of blepharospasm
associated with brain stem lesions had clinical or radiological evidence of involvement of the cerebellum or its
pathways (3,14); however, these structures have not been
considered in the pathophysiology of blepharospasm.
Pathological studies in patients presenting with craniocervical dystonia have been contradictory. Some investiga-

FIG. 2. T1-weighted MRIs in the coronal (A) and axial (B) planes 5
months after stroke showing reabsorption of the hematoma with a residual area of hemosidrrin.

Movement Disorders, Vol. 11, No. I, 1996

98

COMMUNICATIONS

tors (15,16) found cell loss in the dentate and other nuclei
of the brain stem, but similar findings have not been observed in other cases (17). Finally, the association of cerebellar ataxia and dystonia has been described in neurodegenerative disorders such as dentato-rubro-pallidoluysian atrophy (I@, Machado-Joseph disease (19), and
olivopontocerebellar atrophy (20), but in these cases the
basal ganglia could be involved and dystonia has been
considered a minor feature. However, Fletcher et al. (21)
reported eight cases of isolated cerebellar ataxia and focal
dystonia; three patients had torticollis, and six patients
had writer’s cramp, but dystonia was also present in the
outstretched hand in some of these. The cerebellar syndrome was generally more prominent on the side
of the dystonic limb. Although the investigators suggest
that basal ganglia involvement is the responsible mechanism of the dystonic disorder, these cases demonstrate
focal dystonia associated to degeneration of the cerebellum, brain stem, and spinal cord.
Our patient also had bilateral asterixis. Unilateral asterixis has been reported in one patient with a mesencephalic syndrome, suggesting disturbance in and around
the midbrain periaqueductal and pretectal regions (22).
Although bilateral asterixis has been associated with metabolic encephalopathy, we have found one case in the
literature that describes bilateral asterixis secondary to a
structural midbrain lesion (23). The investigators in this
study suggested that damage to the mesencephalic reticular formation was the responsible mechanism of the asterixis. Such a mechanism could explain the asterixis in
our case because bilateral involvement of the periaqueductal reticular formation due to the hemorrhage could be
possible in our patient. Others have described a case of
ipsilateral asterixis due to a cerebellar hemorrhage (24)
and have concluded that asterixis in their patient may
have been due to a reduction of afferent information
through the dentato-thalamo-cortical input. Bilateral involvement of the dentato-thalamic pathways is another
possible explanation of the bilateral asterixis in our patient because MRI shows that the hemorrhage could be
involving either the left dentato-rubro-thalamic fascicle
before decussation and perhaps also, partially, the right
dentato-rubro-thalamic fascicle above the red nucleus.
We conclude that symptomatic limb dystonia can be
produced by a focal brain stem lesion and that cerebellar
dysfunction could play a role in its pathophysiology. Furthermore, we believe that the presence of a brain stem
lesion may explain bilateral asterixis in the absence of
diffuse encephalopathy.

Legends to Videotape
Segment 1. This segment demonstrates the patient’s left
upper limb dysmetria and intention tremor and left leg
heel-knee-shin ataxia.
Segment 2. This segment demonstrates the left arm and
hand dystonia (a) when the patient holds her arms outstretched and (b) when she tries to execute some movements with the limb.
Segment 3. Five months after stroke, the patient’s dystonia is still present but has improved.

Movement Disorders, Vol. 11, No. 1 , 1996

J. Esteban Mufioz
Eduardo Tolosa
Albert Saiz
Nicolas Vila
Maria J. Marti
Rafael Blesa
Parkinson’s Disease and Movement Disorder Unit
Servicio de Neurologia
Hospital Clinic i Provincial
Barcelona, Spain

References
I . Obeso JA, GimCnez-Roldan S. Clinicopathological correlation in symptomatic dystonia. Adv Neurol 1988;50:113-122.
2. Marsden CD, Obeso JA, Zarranz J J , Lang AE. The anatomical basis of symptomatic hemidystonia. Brain 1985;l08:463483.
3 . Jankovic J, Patel SC. Blepharospasm associated with brainstem lesions. Neurofogy 1983;33:1237-1240.
4. Plant GT, Kermode AG, du Boulay EPGH, McDonald WI.
Spasmodic torticollis due to a midbrain lesion in a case of
multiple sclerosis. Mov Disord 1989;4:359-362.
5. Leenders KL, Frackowiak RSJ, Quinn N, Brooks D, Sumner D, Marsden CD. Ipsilateral blepharospasm and contralateral hemidystonia and parkinsonism in a patient with a
unilateral rostra1 brainstem-thalamic lesion: structural and
functional abnormalities studied with CT, MRI, and PET
scanning. Mov Disord 1986;1:51-58.
6. Krauss JK, Mohadjer M, Braus DF, Wakhloo AK, Nobbe F,
Mundinger F. Dystonia following head trauma: a report of
nine patients and review of the literature. Mov Disord 1992;
7:263-272.
7. Santamaria J, Grdus F, Martinez J, Rubio F, Arbizu T, Peres
J . Asterixis en lesiones focales del sistema nervioso central.
Rev Clin ESP 1983;168:37-39.
8. Hedreen JC, Zweig RM, DeLong MR, Whitehouse PJ, Price
DL. Primary dystonias: a review of the pathology and suggestions form new directions of study. Adv Neurol 1988;50:
123-130.
9. Berardelli A, Thompson PD, Day BL, Rothwell JC, O’Brien
MD, Marsden CD. Dystonia of the legs induced by walking
or passive movement of the big toe in a patient with cerebellar ectopia and syringomyelia. Neurology 1986;36:4044.
10. Schoenen J , Grant G. Spinal cord: connections. In: Paxinos
G, ed. Human nervous system. San Diego, CA: Academic,
1990:77-92.
1 1 . Zweig RM, Hedreen JC. Brainstem pathology in cranial dystonia. Adv Neurol 1988;49:395407.
12. Beales M, Lorden JF, Walz E, Oltmans GA. Quantitative
autoradiography reveals selective changes in cerebellar
GABA receptors of the rat mutant dystonic. J Neurosci
1990;10:18761885.
13. Nashold BS Jr, Slaughter G. Effects of stimulating or destroying the deep cerebellar regions in man. J Neurosurg
1969;31:172-186.
14. Jankovic J . Blepharospasm associated with palatal myoclonus and communicating hydrocephalus. Reply from the author. Neurology 1984;34:1522-1523.
IS. Kulisevsky J, Marti MJ, Ferrer I, Tolosa E. Meige syndrome: neuropathology of a case. Mov Disord 1988;3:170175.
16. Zweig RM, Jankel WR, Whitehouse PJ, Casanova MF, Hedreen JC, Price DL. Brainstem pathology in dystonia. Neurology 1986;36(suppl 1):7475.
17. Gibb WRG, Lees AJ, Marsden CD. Pathological report of
four patients presenting with cranial dystonias. Mov Disord
1988;3:211-221.

99

COM M UNICA TI 0NS
18. lizuka R, Hirayama K, Maehara K. Dentato-rubro-pallidoluysian atrophy: a clinico-pathological study. J Neurol Neurosurg Psychiafry 1984:47:1288-1298.
19. Barbeau A, Roy M, Cunha L, et al. The natural history of
Machado-Joseph disease: an analysis of 138 personally examined cases. Can J Neurol Sci 1984;11:510-525.
20. Duvoisin RC. The olivopontocerebellar atrophies. In: Marsden CD, Fahn S , eds. Movement disorders 2. London: Buttenvorths, 1987:249-269.
21. Fletcher N A , Stell R, Harding AE, Marsden CD. Degenerative cerebellar ataxia and focal dystonia. Mov Disord 1988;
3:336-342.
22. Tarsy D, Lieberman B, Chirico-Post J , Benson F. Unilateral
; asterixis associated with a mesencephalic syndrome. Arch
Neurol 1977;34:44&447.
23. Bril V, Sharpe JA, -4shby P. Midbrain asterixis. Ann Neurol
1979;6:362-364
24. Pullicino P, Xuereb M, Farrugia B. Hemorragie cerebelleuse
et asterixis unilateral. Rev Neurol (Paris) 1990;146:697-698.

Walking-Induced Parkinsonism Due to Presumed
Idiopathic Normal Pressure Hydrocephalus
To the Editor:
Normal pressure hydrocephalus (NPH) is a clinical
syndrome without a known pathological substrate other
than hydrocephalus ( 1 ) . It exists in two forms: idiopathic
and secondary (1). The secondary form occurs after inflammatory substances enter the cerebrospinal fluid
(CSF), presumably causing obstruction of CSF flow over
the convexities of the brain. Thus secondary, or “symptomatic,” NPH may occur after subarachnoid hemorrhage, meningitis, or head trauma.
The clinical syndrome of NPH is usually thought of as
a triad of dementia, gait failure, and urinary incontinence
(2), but it is clear from clinical reports that not all elements of the triad need be present (3). Because there is no
pathological abnormality to define the condition, the diagnosis rests on a response to shunting. Unfortunately,
the response to ventricular shunting for the idiopathic
form of NPH is highly variable, but overall is disappointing (4). Diagnostic tests such as cisternography, neuropsychological testing, withdrawal of large volumes of
CSF, and magnetic resonance imaging (MRI) are unreliable due to poor or unknown sensitivity, poor specificity,
or both ( 4 3 ) . In the idiopathic form, the occurrence of
gait dysfunction preceding dementia seems to improve
the prognosis for a shunt response, and the presence of
dementia appears to reduce the benefits of shunting
(3,4,6-8), suggesting that idiopathic NPH (INPH), if defined by shunt responsiveness, is primarily a gait disorder. Presumably most demented patients with ventriculomegaly have diagnoses such as Alzheimer’s disease or
multi-infarct dementia with hydrocephalus ex vacuo, concomitant, unrelated asymptomatic hydrocephalus, or
have had the condition so long that plasticity has been
lost.
The gait abnormalities of INPH have been described in
widely discrepant terms. Sudarsky and Simon (9), who

carefully studied six patients with INPH defined by positive cisternography and beneficial response to CSF removal, basically described the gait as parkinsonism from
the waist down. That is, there was shuffling, with reduced
stride, increased simultaneous contractions of flexors and
extensors of the legs, a flat footed strike, increased time
with both feet touching the floor, and a flexed posture but
intact arm swing. A d a m et al. (lo), in the first published
report of NPH, described the gait as “difficult to categorize” but unsteady in a way that was not clearly cerebellar in origin. Fisher ( 2 ) stated that the majority of the
cases were ataxic, with “poor balance, off balance, unsteady, wobbling, drunken.” Some patients had spells of
weakness. Estanol(11) reported that the gait was apractic
because the patients could perform complicated maneuvers with their feet while seated or supported but could
not move their feet to walk.
This confusion in describing gait either reflects a wide
heterogeneity of presentations of INPH, a large collection
of different disorders associated with ventriculomegaly,
or a large collection of different disorders in patients who
independently had large ventricles unrelated to their neurodegenerative disorder.
The following case describes what I believe is a previously unreported phenomenon, walking-induced parkinsonism, which occurred in an individual with ventriculomegaly and was cured by shunting.

Case Report
The patient was a 68-year-old man referred for evaluation of possible Parkinson’s disease. For the 3 years
preceding referral he had a slowly progressive gait disorder that he attributed to a feeling of fatigue and weakness
in his legs that increased with walking. After resting he
stated that he walked normally but only for a few steps,
after which the problem began and increased with effort.
He also noted a feeling of wanting to run forward when he
became tired. He denied any other symptoms of parkinsonism such as slowness, decreased dexterity, tremor, or
changes in penmanship, voice, or balance. There were
also symptoms of urinary urgency, without incontinence,
and nocturia, which had never been evaluated. There was
no family history of any neurologic disorder, and the patient had no history of neurologic or psychiatric dysfunction, toxin exposure, or use of dopamine-blocking drugs.
He had a history of gout, for which he took colchicine and
allopurinol, and angina, for which he took nifedipine.
Evaluations by previous physicians had included a spinal MRI scan showing spondylotic changes in the lumbar
spine and two electromyograms; the results of one were
normal and the other showed mild polyradiculopathies in
both legs. He scored 29 of 30 on a mini-mental status
examination 1 year before my evaluation. The neurologic
examination was normal except for gait, diminished amplitude on a rapid heel-tapping maneuver bilaterally,
slightly decreased rapid alternating movements of the left
hand, and absent right ankle jerk. The patient initially
walked normally but after a brief period exhibited diminished arm swing on the right and then began to drag his
right leg. With prolonged walking, he developed mild
flexion of the knees and decreased stride bilaterally. The

Movement Disorders, Vul. ! I , No. 1 , 1996