557

Letters to the Editor

Some of the episodes may have been associated with fever. A third severe episode
occurred at the age of 19 two weeks after
normal vaginal delivery of a healthy child.
She developed rapid onset of spasms affecting her legs and arms which progressed so
that she had difficulty breathing and swallowing. She was also vomiting. Examination
showed a temperature of 42°C with tachycardia and tachypnoea and generalised
muscular spasm. Her creatine kinase was
raised at 740 IU/l and investigation for an
infective source including blood cultures
was negative. A chest radiograph was normal and autoantibodies were negative. She
responded well to treatment with intravenous dantrolene within one week and had
sequelae.
permanent neurological
no
Further admissions to a general medical
unit occurred two weeks after starting
phenytoin, with rash, fever and generalised
muscle stiffness, and two months later just
after starting trimethoprim for a urinary
infection, with rash and muscle stiffness
only. No further episodes have since
occurred during treatment with carbamazepine. In addition to the episodes of
fever and muscle stiffness she has several
years of twitching of the hand and calf muscles. No muscle spasm due to exertion or
cold was described. Neurological examination showed hypertrophy of both calves and
continuous twitching of the calves and forearms. Two sisters and her parents had no
neuromuscular symptoms; none had ever
received a general anaesthetic.
Sensory and motor nerve conduction
studies and EMG were performed. The
sensory (superficial peroneal and median)
and motor (posterior tibial and median) distal latencies and conduction velocities were
normal. The sensory nerve action potential
amplitudes were normal as were the compound muscle action potential (CMAP)
amplitudes recorded from the abductor pollicis brevis and abductor hallucis. The
CMAP waveforms were followed by after
discharges lasting for up to 50 seconds,
which made accurate estimation of F wave
latencies impossible. Concentric needle
EMG of tibialis anterior and medial gastrocnemius showed repetitive spontaneous
discharges of motor unit potentials at all
sites sampled. This activity occurred irregularly in rhythmical 0-5-1-5 second bursts at
high rates (50-200/s).
This patient has electrophysiologically
established spontaneous and stimulus
induced neuromyotonia without neuropathy
with episodes of hyperpyrexia consistent
with malignant hyperthermia.' Patients with
neuromyotonia may have increased sweating and feel unwell but we think that malignant hyperthermia has not previously been
described. The underlying deficit in neuromyotonia is likely to lie in the nerve cell
membrane, and some patients may have
channel
abnormalities.2
potassium
Malignant hyperthermia has been associated with muscle membrane disorders, and
abnormalities in muscle ion channels have
been described.' We suggest an anomalous
ion channel common to muscle and nerve
as a possible mechanism for the association
in this patient.

T D GRIFFITHS
Neurology Department,
Middlesbrough General Hospital
S CONNOLLY
Department of Clinical Neurophysiology,
Massachussetts General Hospital,
32 Fruit Street,
Boston, MA 02114, USA

P K NEWMAN
Neurology Department,

Middlesbrough General Hospital,
Ayresome Green Lane,
Middlesbrough,
Cleveland,
TS5 SAZ, UK
M C BATESON
Bishop Auckland General Hospital,
Cockton Hill Rd,
Bishop Auckland,
County Durham,
BL14 6AD, UK

Correspondence to Dr Newman.
1 Nelson TE, Flewellen EH. The malignant
hyperthermia syndrome. N Engl J Med 1983;
309:416-8.
2 Newsom-Davis J, Mills KR. Immunological
associations of acquired neuromyotonia.
Brain 1993;116:453-69.
3 Foster PS, Denborough MA. The effect of calcium channel antagonists and BAY K 8644
on calcium fluxes of malignant hyperthermia
susceptible muscle. Int J Biochem 1993;25:
495-504.

Arg296 to Cys296 polymorphism in exon
6 of cytochrome P-450-2D6 (CYP2D6) is
not associated with multiple system
atrophy
An allelic association between mutant alleles
of the cytochrome P-450-2D6 (CYP2D6)
and Parkinson's disease has been shown by
several groups'-3. The analysed mutations lie
in exon 4 and 5 of the CYP2D6 gene,
homozygotes showing the poor metaboliser
phenotype. Analysis of these mutant alleles
among patients with multiple system atrophy showed no difference in the frequency
of these alleles from that in control subjects.4
A further polymorphism causing an
amino acid change from Arg296 to Cys296
at the HhaI site in exon 6 of the CYP2D6
gene has been described.56 The frequency of
this polymorphism, which is not associated
with the poor metaboliser phenotype,5 was
26% among normal white subjects,5 but only
9% in Japanese control subjects.6 Recently,
an association with the Arg296 to Cys296
polymorphism was suggested in a small
series of 10 Japanese patients with multiple
system atrophy and it was suggested that
this polymorphism may be a useful marker
for susceptibility to this disease.7
We examined the frequency of this polymorphism in a larger series of 74 white
patients with multiple system atrophy. The
diagnosis was clinical in 598 and pathologically established in a further 15, in whom
frozen brain samples were analysed. The
method for the detection of the polymorphism has been described elsewhere.6 Our
results show a similar frequency of the
mutant allele (48 of 148 total alleles) among
the patients with multiple system atrophy
to the published frequency among white
control subjects (32% v 26%; X2 = 0-97,
P = 0 32).
We conclude that, at least in white subjects, the HhaI polymorphism in exon 6 of
the CYP2D6 gene is not associated with
multiple system atrophy and is therefore not
a useful marker for susceptibility to multiple
system atrophy.
Financial

support

from

the

Deutsche

the
and
(DFG)
Forschungsgemeinschaft
Parkinson's Disease Society is gratefully acknowledged.
0 BANDMANN

G K WENNING
N P QUINN
A E HARDING

University Department of Clinical Neurology,
Institute of Neurology,
Queen Square,
London WClN 3BG, UK
1 Smith CAD, Cough AC, Leigh PN, Summers
BA, Harding AE, Maraganore DM, et al.
Debrisoquine hydroxylase gene polymorphism and susceptibility to Parkinson's disease. Lancet 1992;339:1375-7.
2 Armstrong M, Daly AK, Cholerton S,
Bateman DN, Idle JR. Mutant debrisoquine
hydroxylation genes in Parkinson's disease.
Lancet 1992;339: 1017-8.
3 Kurth MC, Kurth JH. Variant cytochrome
P450 CYP2D6 allelic frequencies in
Parkinson's disease. Am J Med Genet 1993;
48: 166-8.
4 Plante-Bordeneuve V, Bandmann 0, Wenning
G, Quinn NP, Daniel SE, Harding AE.
CYP2D6 debrisoquine hydroxylase gene
polymorphism in multiple system atrophy.
Mov Disord 1995;10:277-8.
5 Armstrong M, Idle JR, Daly AK. A polymorphic cfo I site in exon 6 of the human
cytochrome P450 CYP2D6 gene detected by
the polymerase chain reaction. Hum Genet
1993;91 :616-7.
6 Tsuneoka Y, Matsuo Y, Iwahashi K, Takeuchi
H, Ichikawa Y. A novel cytochrome
P450IID6 mutant gene associated with
Parkinson's disease. Y Biochem 1993;114:
263-6.
7 Iwahashi K, Miyatake R, Tsuneoka Y, Matsuo
Y, Ichikawa Y, Hosokawa K, et al. A novel
cytochrome P-450IID6 (CYPIID6) mutant
gene associated with multiple system atrophy. J Neurol Neurosurg Psychiatry 1995;
58:263-4.
8 Quinn NP. Multiple system atrophy. In: Fahn
S, Marsden CD, eds. Movement disorders 3.
Oxford: Butterworths International Medical
Reviews, 1994:262-81.

Ataxic hemiparesis with bilateral leg
ataxia from pontine infarct
Ataxic hemiparesis is a syndrome characterised by weakness and cerebellar-like
ataxia on the same side of the body.' A
lesion resulting in ataxic hemiparesis must
involve both the corticospinal fibres and the
afferent or efferent cerebellar fibres in locations where the tracts are in close proximity.
The afferent and efferent cerebellar fibres
form a loop extending from the cerebral cortex through the pons and middle cerebellar
peduncle to the cerebellar cortex and then
extending from the dentate nucleus through
the superior cerebellar peduncle, red
nucleus, and thalamus back to the cerebral
cortex. Ataxic hemiparesis has been associated with lesions in the corona radiata, thalamus, midbrain, and pons. Fisher and Cole
first reported that a paramedian infarct of
the basis pontis located at the junction of the
upper one third of the pons with the lower
two thirds could produce a contralateral
ataxic hemiparesis.' One of the major questions concerning pontine ataxic hemiparesis
is why the limb ataxia is seen only contralateral to the lesion and not bilaterally.' 2 The
corticopontine fibres terminate by synapsing
with the pontine nuclei and most fibres then
cross the midline to enter the contralateral
middle cerebellar peduncle. A basis pontis
infarct might thus be expected to produce
bilateral limb ataxia because it would involve
ipsilateral pontine nuclei and corticopontine
fibres as well as pontocerebellar fibres that
have crossed from the contralateral side. We
report a case of a mid-pontine paramedian
infarct with caudolateral extension resulting
in ataxic hemiparesis with bilateral leg
ataxia.
An 80 year old white man with a history
of coronary artery disease suddenly noticed
left sided weakness. On examination, he had

Letters to the Editor

558

Figure 1 Axial T2 weighted MRI of
mid-pons showing infarct in the right lateral
basis pontis extending medially.
a blood pressure of 120/72. He had a left
central facial weakness and a mild left hemiparesis affecting the arm and leg equally.
Incoordination of the left upper extremity
was present on finger to nose testing and
bilateral heel-knee-shin ataxia was present,
worse on the left. The muscle stretch
reflexes were normal and the plantar reflexes
were flexor bilaterally. Brain MRI disclosed
an area of increased signal intensity involving both the paramedian mid-pons and
extending laterally in the caudal third of the
pons (fig 1).
It is unclear why patients with ataxic
hemiparesis usually have ataxia contralateral
to the lesion rather than bilateral limb
ataxia. It has been suggested that the pon-

tine nuclei are more vulnerable to ischaemia
than the transverse pontocerebellar fibres
crossing from the contralateral side but there
is no histological evidence for this.3 Huang
and Chang2 suggested that the crossing
transverse pontocerebellar fibres take an
oblique downward course to the contralateral middle cerebellar peduncle and in
this way a more rostrally placed pontine
lesion may involve only the corticopontine
fibres and pontine nuclei ipsilaterally, but
miss the more caudally placed crossing pontocerebellar fibres. A more caudolateral
lesion would be expected to give rise to
ipsilateral ataxia (fig 2). Our case supports
this suggestion, as the lesion extended more
laterally and caudally than previously
described pontine infarcts associated with
ataxic hemiparesis. The more caudolateral
part of the infarct in our patient may have
damaged the crossed pontocerebellar fibres
and have resulted in the leg ataxia ipsilateral
to the infarct, whereas the more medial and
rostral part may have given a contralateral
ataxic hemiparesis (fig 2). This suggestion is
supported by Fisher's report of a lower lateral pontine infarct4 and Fisher and Tapia's
report of a lateral medullary infarct extending into the lower lateral pons. Both these
infarcts gave rise to ataxia only ipsilateral to
the lesion.

SHERRY WITHIAM-LEITCH
PATRICK PULLICINO
Department of Neurology, State University of
New York at Buffalo,
Buffalo General Hospital, 100 High Street,
Buffalo NY 14203, USA

Correspondence to: Dr S Withiam-Leitch.
1 Fisher CM. Ataxic hemiparesis. A pathologic
study. Arch Neurol 1978;35:126-8.

2 Huang CY, Chang KH. Pontine ataxic hemiparesis, a lateral penetrator syndrome? _7
Neurol Neurosurg Psychiatry 1984;47: 1046-7.
3 Nabatame H, Fukuyama H, Akiguchi I,

Kameyama M, Nishimura K, Torizuka K.
Pontine ataxic hemiparesis studied by a high
resolution magnetic resonance imaging system. Ann Neurol 1987;21:204-7.
4 Fisher CM. Lacunar infarct of the tegmentum of
the lower lateral pons. Arch Neurol 1989;
46:566-7.
5 Fisher CM, Tapia J. Lateral medullary infarction
extending to the lower pons. .7 Neurol
Neurosurg Psychiatry 1987;50:620-4.

Autoimmune chronic active hepatitis
and polymyositis in a patient with
myasthenia gravis and thymoma

Figure 2 Diagram of coronal section through
brainstem showing corticopontine fibres (1),
corticospinal tract (2), and pontocerebellarfibres
(3). The site of a pontine lesion causing ataxic
hemiparesis as described by Fisher' is shown in
bold dots, the site of a lesion causing ipsilateral
ataxia4 is shown in open dots, and the site of the
lesion in our patient is shown in cross hatching.

Myasthenia gravis is known to occur with
other autoimmune disease. It is rarely associated with polymyositis,l and very exceptionally with autoimmune chronic active
hepatitis.2 We describe a patient with generalised myasthenia gravis, cortical thymoma,
polymyositis, and autoimmune chronic
active hepatitis.
A 25 year old Chinese woman presented
with intermittent weakness of the limbs for
one month with no diplopia, speech disturbance, or bulbar symptoms. She had mild
ptosis of her left eye which increased progressively with maintained upward gaze, and
proximal limb muscle weakness. The rest of
the examination was normal. She had a positive edrophonium test, and serum antiacetylcholine receptor and antistriated
muscle antibody titres. Chest radiography
disclosed a mediastinal mass which was subsequently confirmed by CT to be a thymoma measuring 4-2 cm (anteroposterior)
x 3-1 cm (width) in the left lobe of the thy-

mus. Serial serum muscle enzymes were
raised. A repetitive stimulation test on three
muscles showed a more than 15% decrement with postexercise exhaustion. An
EMG showed abundant fibrillations, positive sharp waves and insertional irritability,
and short, small polyphasic motor units.
Biopsy of her deltoid muscle showed
perivascular infiltration of lymphocytes with
phagocytosis and variable numbers of angulated atrophic fibres.
Incidentally she was found to have hepatitis. During the initial stay in hospital, a liver
function test gave albumin 45 g/l, globulin
37 g/l, alkaline phosphatase 106 U/l, bilirubin 21 umol/l, alanine aminotransferase
1158 U/l, and aspartate aminotransferase
943 U/l. Viral markers of hepatitis A, B, and
C were all negative, whereas antinuclear factor and antismooth muscle antibody were
positive. Her HLA antigens were A33, B17,
and DR3. Liver biopsy showed a picture of
active chronic hepatitis with enlarged portal
tracts and a pronounced mononuclear cell
infiltrate which spilt across the limiting
plate. There were foci of piecemeal necroses
and the hepatocytes often displayed ballooning degeneration. Staining for hepatitis B
was negative.
Her muscle weakness responded promptly
but not completely to pyridostigmine
(60 mg four times daily), and after the liver
biopsy, she was treated with prednisolone
(55 mg daily) and azathioprine (50 mg)
daily. Three weeks after, she developed
ophthalmic herpes, which was treated with

acyclovir.
She continued to improve both in muscle
power and liver function. Two months after
treatment with prednisolone and azathioprine, the liver enzymes were considerably
decreased alanine aminotransferase 39 U/l
and aspartate aminotransferase 38 U/l.
Thymectomy was then performed and the
postoperative course was uneventful.
Histology confirmed that the thymoma was
encapsulated, non-invasive, and of cortical
type.
This is the first report of the combination
of generalised myasthenia gravis with thymoma, polymyositis, and autoimmune
chronic active hepatitis. Fifty per cent of
patients with chronic active hepatitis will die
of liver failure within five years if no treatment is given. Although the prognosis can
be considerably improved with steroid and
azathioprine treatment, most patients
develop cirrhosis.'
The susceptibility of the patient to generalised myasthenia gravis, autoimmune
chronic active hepatitis, and polymyositis
seems to be related to HLA DR3.4 There is
probably an interplay of genetic and environmental factors in the occurrence of these
diseases. Interestingly, the association of
polymyositis and HLA DR3 is claimed to be
related to the coexistent presence of antibodies to histidyl tRNA synthetase (Jo-1).5
In this regard the appearance of such
autoantibodies in polymyositis is reported to
be most uncommon4 and was absent in the
patient reported herein. Despite the coexistence of the diseases, a good response was
obtained with steroid combined with azathioprine and thymectomy.

KWAI FU KO
TINNY HO
Department of Medicine
KWOK WAH CHAN
Department of Pathology,
University of Hong Konig,
Queen Mary Hospital,
Pokfulan Road, Hong Kong