PostScript

1095

Table 1 Cytokine profile during the clinical course
IL-6 (pg/ml)

Before the first LCP
After the first LCP
After the last LCP

IL-12 (pg/ml)

Serum

CSF

Serum

CSF

3.2
3.0
2.8

5660
121
6.1

18.5
,7.8
,7.8

,7.8
,7.8
,7.8

IL, interleukin; LCP, lymphocytapheresis.

In our patient, an infectious aetiology was
unlikely, and positive antinuclear antibody
suggested an autoimmune mechanism. The
specific biomarker for NMO, NMO-IgG, has
been reported in NMO and opticospinal
multiple sclerosis, suggesting an autoantibody-driven pathogenesis.1 The sensitivity of
NMO-IgG for diagnosing NMO is 73%,1 and
our patient would be classified as being in the
seronegative group. It has, however, been
reported that opticospinal multiple sclerosis,
which seems to be similar to NMO with
respect to NMO-IgG,1 is associated with the
Th1-dominant condition during both the
relapse and remission phases.4 In our patient,
a shift in the Th1:Th2 balance towards Th1
dominance was shown during both the acute
and remission phases. Th1 cells possibly play
an important part in the pathogenesis of
Devic disease. An investigation of this will
require further study with more patients.
It has been reported that a proinflammatory cytokine, IL-6-secreting cells, or a critical
cytokine for T cell-mediated autoimmunity,
IL-12-secreting cells, are increased in both
the CSF and blood from patients with NMO.5
In our patient, increased levels of CSF IL-6
and serum IL-12 were consistent with those
previously reported, but levels of serum IL-6
and CSF IL-12 were normal. Levels of CSF IL6 and serum IL-12 correlated well with
disease activity. These cytokine levels may
therefore serve as markers of therapeutic
effects as well as disease activity in Devic
disease. Levels of CSF IL-6 and serum IL-12
were reduced immediately after the first LCP,
suggesting the direct effect of LCP on
cytokines. Clinical improvement is thought
to have been induced by LCP, because of the
extremely rapid improvement within 1 day
after LCP.
In conclusion, LCP may be a therapeutic
option in fulminant Devic disease that is
unresponsive to other treatments, including
methylprednisolone treatment.
I Nozaki, T Hamaguchi, K Komai, M Yamada
Department of Neurology and Neurobiology of
Aging, Kanazawa University Graduate School of
Medical Science, Kanazawa, Ishikawa, Japan
Correspondence to: Ichiro Nozaki, Department of
Neurology and Neurobiology of Aging, Kanazawa
University Graduate School of Medical Science, 13-1
Takara-machi, Kanazawa, Ishikawa 920-8640,
Japan; ichi51@med.kanazawa-u.ac.jp

doi: 10.1136/jnnp.2005.086306
Informed consent was obtained for publication of the
patient’s details in this report.
Competing interests: None declared.

References
1 Lennon VA, Wingerchuk DM, Kryzer TJ, et al. A
serum autoantibody marker of neuromylitis
optica: distinction from multiple sclerosis. Lancet
2004;364:2106–12.

2 Aguilera AJ, Carlow TJ, Smith KJ, et al.
Lymphocytaplasmapheresis in Devic’s syndrome.
Transfusion 1985;25:54–6.
3 Kleinman S, Nichols M, Strauss F, et al. Use of
lymphoplasmapheresis or plasmapheresis in the
management of acute renal allograft rejection.
J Clin Apheresis 1982;1:14–7.
4 Ochi H, Wu XM, Osoegawa M, et al. Tc1/Tc2
and Th1/Th2 balance in Asian and Western types
of multiple sclerosis, HTLV-I-associated
myelopathy/tropical spastic paraparesis and
hyperIgEaemic myelitis. J Neuroimmunol
2001;119:297–305.
5 Correale J, Fiol M. Activation of humoral
immunity and eosinophils in neuromyelitis optica.
Neurology 2004;63:2363–70.

Repetitive transcranial magnetic
stimulation of the motor cortex
for hemichorea
Hyperkinetic disorders such as chorea, ballism and dystonia can be observed after
lesions (usually small strokes) affecting the
thalamus, basal ganglia and related circuits.1
Neural circuit models of basal ganglia circuitry suggest that these result from disinhibition of thalamo-cortical projections because
of a reduction in inhibitory output from the
internal pallidal segment. Increased thalamocortical drive may lead to an increase in the
excitability of excitatory and inhibitory circuits of the frontal areas of the cortex,2 3
including M1, that may contribute to hyperkinesia.
Circuits of the human motor cortex can be
activated non-invasively with transcranial
magnetic stimulation (TMS) of the brain,4
and because the excitability of neural circuits
in the cerebral cortex is not static, repetitive
TMS (rTMS) can produce changes in neurotransmission that outlast the period of
stimulation.5–7
rTMS has been proposed as a therapeutic
tool in several psychiatric and neurological
disorders, and a recent review suggested that
the best effects of therapeutic rTMS are seen
when protocols that depress network excitability are used to treat disorders characterised by cortical hyperexcitability.8
A recent study showed that excitability of
the motor cortex can be reduced for 30–
60 min after application of a novel paradigm
of rTMS termed continuous h burst stimulation (cTBS).9 10 This protocol leads to a
decrease in the excitability of excitatory and
inhibitory cortical circuits that is long lasting.
Thus, after TBS, there is a decrease in the
amplitude of the corticospinal volleys evoked
by transcranial stimulation,10 and the size of
the resulting motor-evoked potentials.9 In
addition, a reduction in excitability of intracortical inhibitory circuits was shown by a
decrease in short latency intracortical inhibition, a putative marker of c aminobutyric acid
transaminase-A activity.9
The aim of this preliminary study was to
investigate whether cTBS of the motor cortex

could reduce involuntary movements in a
patient with hemichorea-ballism.
We hypothesised that the reduced excitability in excitatory and inhibitory circuits of
the motor cortex that follows cTBS9 could
normalise excitability of frontal areas and
improve hyperkinesias.
A 68-year-old woman with a history of
hypertension and glaucoma suddenly presented with severe involuntary movements of
the left arm, leg and face. An MRI of the
brain, carried out a week after the stroke,
showed a lesion with signal characteristics of
subacute haemorrhage in the midbrain and
the caudal diencephalon on the right side,
and thus conceivably affecting the subthalamic nucleus or the substantia nigra, or both
these structures (fig 1). A lesion of these
structures would remove excitatory drive to
the internal segment of the globus pallidus
and the resulting decrease in pallidal inhibition of thalamus would lead to a consequent
increase in thalamo-cortical excitation.
Treatment with levetiracetam (500 mg twice
daily) produced no benefit.
In the weeks after the stroke, there was
mild improvement in the severity of dyskinesias. Movements on the left, and particularly in the arm, however, never resolved
completely, and caused pain in the left
shoulder.
Ten weeks after the stroke, neurological
examination showed left hemichorea, particularly affecting the arm, with ballistic movements at the shoulder and distal choreoathetoid movements. Mild choreatic movements were also observed in the left hemifacial muscles and leg. Posture, voluntary
movements or tasks requiring concentration
increased intermittent dyskinesias at rest.
The rest of the neurological examination
was unremarkable apart from blindness of
the left eye secondary to glaucoma.
Treatment with increasing doses of tetrabenazine up to a dose of 25 mg thrice daily
failed to produce any benefit after 2 months.
At this time, rTMS of the motor cortex was
carried out as an add-on treatment.
The patient gave informed consent to the
study that was performed with the approval
of the appropriate Institutional ethics committee.
rTMS was applied over the left-hand motor
area by using a MagPro (Medtronic A/S,
Copenhagen, Denmark) stimulator and a
figure of eight-shaped coil. The stimulation
intensity was 80% of the active motor threshold (AMT), defined as the minimum single
pulse intensity required to produce a motor
evoked potential .200 mV on more than five
out of ten trials from the contracted contralateral first dorsal interosseous.
rTMS was carried out using the cTBS
pattern in which three pulses of stimulation
are given at 50 Hz and repeated every
200 ms, for a total of 600 pulses.9 This
protocol leads to pronounced and prolonged
suppression of cortical excitability that
reaches a maximum about 5–10 min after
the end of the protocol.9
Sham rTMS was performed by using the
same stimulator connected to the placebo
butterfly coil MCF-P-B-65, which has no
stimulating effect on the cortex but produces
auditory and tactile sensations similar to
those produced by the real coil. The site of
the stimulation and the number of stimuli
were identical to those used for the real
magnetic rTMS.
Two sessions of real rTMS (the first and the
third) and one session of sham rTMS were

www.jnnp.com

1096

carried out at intervals of about 10 days. At
each session, the patient was evaluated by a
skilled movement disorders neurologist. Both
the patient and the neurologist who scored
the dyskinesias were blinded to the rTMS
protocol (sham or real) used in each session.
Dyskinesias were measured before and
10 min after stimulation by means of a
Dyskinesias Scale (ranging from 0 to 4, see
appendix) scored in all the following conditions: at rest; while performing a distracting
mental activity; while walking; and while
performing voluntary tasks (pouring water,
drinking from a glass, putting on a lab coat,
buttoning and finger-to-nose manoeuvre).
Drinking was the action on which the worst
performance occurred and it was therefore
the task chosen to compare the severity of
dyskinesias in all the six assessments.
The evaluation of the involuntary movements was carried out live and video recordings of the first and second sessions were
obtained.
The patient could not distinguish between
real and sham stimulation.
Real rTMS produced a dramatic improvement in dyskinesias during functional tasks
requiring the use of the upper limb (for video
A and table, see http://jnnp.bmjjournals.com/
supplemental).

PostScript

Scores on the evaluation scale improved
from 4 to 1 (pouring, drinking), from 3 to 2
(buttoning), from 3 to 1 (rest, finger-to-nose)
and from 2 to 1 (walking) after the first
session and similarly after the third session
(table). The patient reported that the benefit
lasted about 24 h after both sessions, with
associated reduction of the pain at the left
shoulder. After the sham stimulation (second
session), no consistent variations were
reported by the patient or by the evaluating
neurologist. Basal scores on that day were 3
for pouring, drinking, buttoning and fingerto-nose manoeuvre and 2 for rest and
walking: a slight improvement was noticed
by the evaluating neurologist but was judged
to be insufficient to modify the rating on each
of the considered tasks (for video B and table,
see http://jnnp.bmjjournals.com/supplemental).
Studies in patients with chorea due to
Huntington’s disease3 and in a patient with
diabetes with hemiballism–hemichorea2 suggest that in these conditions increased
thalamo-cortical drive may increase excitability of excitatory and inhibitory circuits of
frontal areas of the cortex, including M1, and
contribute to the development of hyperkinesias. The present study shows that rTMS
protocols that suppress the excitability of

excitatory and inhibitory circuits of the motor
cortex may be useful in the treatment of
hyperkinetic disorders. This would be consistent with a recent study in which lowfrequency rTMS, which is another way of
decreasing cortical excitability, was used on
midline frontal cortex to reduce dyskinesias
in patients with Huntington’s disease.11
In this study we cannot determine whether
the principal effect of cTBS occurs through its
suppression of excitatory or inhibitory circuits, or both. What is more certain, however,
is that the effects are likely to have occurred
because of a change in cortical rather than
subcortical circuits. Thus, cTBS is known to
suppress excitatory circuits that generate I
waves4 within the motor cortex. Similarly,
cTBS reduces short-latency intracortical inhibition,9 which again is a cortical phenomenon.4 Currently, there is no evidence that
cTBS leads to any changes in excitability of
subcortical circuits.
Because the effects of TMS are short-lived,
cTBS is not suitable as a chronic intervention.
It can, however, turn out to be useful in
selecting patients for implantation with
cortical epidural electrodes as both TMS and
epidural stimulation appear to activate similar elements in the motor cortex.12 A similar
approach has been suggested by Lefaucheur13
to select patients with pain who may be
candidates for chronic epidural motor cortex
stimulation.

V Di Lazzaro, M Dileone, F Pilato,
M-F Contarino, G Musumeci, A R Bentivoglio
Institute of Neurology, Università Cattolica, Rome,
Italy

P A Tonali
Fondazione Don C Gnocchi, Rome

J C Rothwell
Sobell Department of Neurophysiology, Institute of
Neurology and The National Hospital for Neurology
& Neurosurgery, Queen Square, London, UK
Correspondence to: Vincenzo Di Lazzaro, Istituto di
Neurologia, Università Cattolica, L.go A. Gemelli 8,
00168 Rome, Italy; vdilazzaro@rm.unicatt.it

doi: 10.1136/jnnp.2005.082875
Informed consent was obtained for publication of the
patient’s details in this report.
Competing interests: None declared.

An additional table and
videos 1 and 2 are available
at http://jnnp.bmjjournals.
com/supplemental

References

Figure 1 MRI of the brain of the patient. (A) Axial SE T1-weighted image (TR 582/TE 15). (B)
Axial TSE image (TR 2200/TE 120). (C) Axial GE T2-weighted image (TR 600/TE 23). (D) Axial
T2-FLAIR image (TR 6000/TE 100/TI 2000). The images show a subacute haemorrhagic lesion in
the midbrain of the right side.

www.jnnp.com

1 Lee MS, Marsden CD. Movement disorders
following lesions of the thalamus or subthalamic
region. Mov Disord 1994;9:493–507.
2 Ziemann U, Koc J, Reimers CD, et al. Exploration
of motor cortex excitability in a diabetic patient
with hemiballism-hemichorea. Mov Disord
2000;15:1000–5.
3 Modugno N, Curra A, Giovannelli M, et al. The
prolonged cortical silent period in patients with
Huntington’s disease. Clin Neurophysiol
2001;112:1470–4.
4 Di Lazzaro V, Oliviero A, Pilato F, et al. The
physiological basis of transcranial motor cortex
stimulation in conscious humans. Clin
Neurophysiol 2004;115:255–66.
5 Berardelli A, Inghilleri M, Rothwell JC, et al.
Facilitation of muscle evoked responses after

PostScript

1097

repetitive cortical stimulation in man. Exp Brain
Res 1998;122:79–84.
6 Chen R, Classen J, Gerloff C, et al. Depression of
motor cortex excitability by low-frequency
transcranial magnetic stimulation. Neurology
1997;48:1398–403.
7 Pascual Leone A, Valls-Solè J, Wassermann EM,
et al. Responses to rapid-rate transcranial
magnetic stimulation of the human motor cortex.
Brain 1994;117:847–58.
8 Wassermann EM, Lisanby SH. Therapeutic
application of repetitive transcranial magnetic
stimulation: a review. Clin Neurophysiol
2001;112:1367–77.
9 Huang YZ, Edwards MJ, Rounis E, et al. Theta
burst stimulation of the human motor cortex.
Neuron 2005;45:201–6.
10 Di Lazzaro V, Pilato F, Saturno E, et al. Thetaburst repetitive transcranial magnetic stimulation
suppresses specific excitatory circuits in the
human motor cortex. J Physiol 2005;565:945–5.
11 Brusa L, Versace V, Koch G, et al. Improvement of
choreic movements by 1 Hz repetitive
transcranial magnetic stimulation in Huntington’s
disease patients. Ann Neurol 2005;58:655–6.
12 Di Lazzaro V, Oliviero A, Pilato F, et al.
Comparison of descending volleys evoked by
transcranial and epidural motor cortex
stimulation in a conscious patient with bulbar
pain. Clin Neurophysiol 2004;115:834–8.
13 Lefaucheur JP. Transcranial magnetic stimulation
in the management of pain. Suppl Clin
Neurophysiol 2004;57:737–48.

We present the case of a patient with
leptomeningeal carcinomatosis from a primary breast adenocarcinoma. On admission
to hospital, she developed neurogenic pulmonary oedema due to increased intracranial
pressure (ICP), as confirmed by lumbar
puncture. Tapping of the CSF resulted in
abrupt relief of the clinical sequelae.
A woman in her late 40s, with a medical
history of breast adenocarcinoma, for which
she underwent a modified radical mastectomy with axillar lymph node resection, was
admitted to our hospital for intrathecal
chemotherapy. Just 1 month earlier, she
underwent craniotomy for a solitary metastatic intracerebral lesion in the left frontal
lobe. After this uneventful procedure, she
continuously complained of drowsiness, nausea and vomiting. A CT scan showed
enhancement of cerebral and cerebellar sulci,
typical of leptomeningeal metastasis.
This diagnosis prompted admittance to our
hospital, a tertiary care oncology centre, for

further treatment. On admittance, the patient
had increasing dyspnoea. Pulmonary and
cardiac examination showed basilar crepitations on both lungs, tachypnoea (.24/min),
tachycardia (145 beats/min) and a blood
pressure of 80/40 mm Hg. No cardiac murmurs were heard and the electrocardiograph
showed no signs of myocardial infarction.
The peripheral oxygen saturation at that time
was 80% with no oxygen support. Blood gas
analysis showed hypoxia: pH 7.46, partial
pressure of carbon dioxide (pCO2) 4.7, partial
pressure of oxygen (pO2) 5.4, HCO32 24.7,
(BE) 1.3 and arterial oxygen saturation
(SaO2) 0.79. A chest radiograph showed a
normal heart with evidence of bilateral
alveolar filling (fig 1B; fig 1A shows the
chest radiograph taken 1 day earlier, which
was without abnormalities).
The patient was admitted to the intensive
care unit and respiration was supported by
mechanical ventilation through a facemask
and administration of oxygen. Nevertheless,
her neurological condition deteriorated to a
Glasgow Coma Scale (GCS) score of 6. On
reconsidering the CT scan and the diagnosis
of leptomeningeal carcinomatosis, we
hypothesised that increased ICP, due to a
CSF resorption disturbance, caused neurogenic pulmonary oedema. Therefore, we
carried out a lumbar puncture, showing an
increased CSF pressure of 60 cm H2O. We
drained 30 ml of the CSF, during which the
patient regained consciousness acutely (GCS
13). Cytology showed the presence of malignant cells. The tachypnoea and tachycardia
resolved, although the basilar crepitations
could still be recognised on pulmonary
auscultation. Blood gas analysis showed
recovery: pH 7.42, pCO2 5.2, pO2 8.0, HCO32
24.8, BE 1.0 and SaO2 0.93.
The patient was readmitted to the neurosurgical department for a CSF shunt procedure. She was, however, not fit for surgery,
and attempts to alleviate the increased ICP
with an external lumbar drain failed.
Treatment was discontinued when her condition deteriorated again 5 days later, after
which she died.
As far as we know, leptomeningeal metastasis has not been described as the cause of
neurogenic pulmonary oedema. Although the
association of an increased ICP and neurogenic pulmonary oedema has been described
previously, the pathophysiological mechanisms remain obscured.1–3 The medulla oblongata is believed to be the critical anatomical
structure responsible for the pathogenesis of

A

B

Neurogenic pulmonary oedema
in a patient with leptomeningeal
carcinomatosis

LI

L

neurogenic pulmonary oedema, probably
acting through the sympathetic component
of the autonomic nervous system.1 2
The rapid improvement after the normalisation of the increased CSF pressure is more
consistent with neurogenic lung oedema
than with other causes of respiratory distress.
As reviewed by Fontes et al,4 the clinical
features as described in our patient—that is,
tachypnoea, dyspnoea, tachycardia and hypotension—often occur in acute neurogenic
pulmonary oedema. As Fontes et al conclude,
we believe that doctors should always be
aware of this clinical condition after neurological events, especially when no other
causes of pulmonary oedema can be established. Supportive measures are the treatment of choice. In our patient, relieving the
increased ICP corrected the respiratory failure.

R Dammers, M J van den Bent
Department of Neuro-Oncology, Dr Daniël den Hoed
Cancer Clinic, Rotterdam, The Netherlands

R Dammers
Department of Neurosurgery, Erasmus Medical
Centre, Rotterdam
Correspondence to: Dr Martin J van den Bent,
Department of Neuro-Oncology, Dr Daniël den Hoed
Cancer Clinic, PO Box 5201, 3008 AE Rotterdam, The
Netherlands; m.vandenbent@erasmusmc.nl

doi: 10.1136/jnnp.2005.079178
Informed consent was obtained for publication of the
patient’s details in this report.
Competing interests: None declared.

References
1 Brambrink AM, Dick WF. Neurogenic pulmonary
edema. Pathogenesis, clinical picture and
therapy. Anaesthesist 1997;46:953–63.
2 Malik AB. Pulmonary vascular response to
increase in intracranial pressure: role of
sympathetic mechanisms. J Appl Physiol
1977;42:335–43.
3 McClellan MD, Dauber IM, Weil JV. Elevated
intracranial pressure increases pulmonary
vascular permeability to protein. J Appl Physiol
1989;67:1185–91.
4 Fontes RB, Aguiar PH, Zanetti MV, et al. Acute
neurogenic pulmonary edema: case reports and
literature review. J Neurosurg Anesthesiol
2003;15:144–50.

Onset of cluster headache
triggered by emotional effect: a
case report
Cluster headache is a strictly unilateral headache that occurs in association with autonomic symptoms. Stress is a recognised
precipitant of migraines, but not of cluster
attacks. We describe the case of a patient
having migraine for years, in whom extreme
emotional stress triggered cluster headache
attacks.

Background

Figure 1 Chest radiograph: (A) taken 1 day earlier; (B) with evidence of bilateral alveolar filling.

Neuroimaging has contributed considerably
to the knowledge about the neurobiology of
cluster headache. Activation in the region of
the posterior hypothalamus is observed,1
coinciding with subtle structural abnormalities in the same region.2
Genetic studies indicate an autosomal
dominant inheritance with low penetrance
and an increase in risk of 4–18 times for

www.jnnp.com