J Neurol Neurosurg Psychiatry 2001;71:278–283

280

with late-onset familial and sporadic
Alzheimer’s disease. Neurology 1993;43:1467–
72.
4 Grimaldi LME, Casadei VM, Ferri C, et al.
Association of early-onset Alzheimer’s disease
with an interleukin-1á gene polymorphism.
Ann Neurol 2000;47:361–5.
5 Combarros O, Sanchez-Guerra M, Llorca J, et
al. Polymorphism at codon 129 of the prion
protien gene is not associated with sporadic
AD. Neurology;55:593–5.

Complex musical hallucinosis in a
professional musician with a left
subcortical haemorrhage
Auditory hallucinosis consists of abnormal
acoustic perceptions that occur in the absence of a corresponding acoustic stimulus
while the patient is aware of their non-real
nature.1
Musical hallucinosis represents a particular
type of acoustic hallucinosis, in which the
acoustic perception is formed by music,
sounds, or songs. It is frequent in psychiatric
diseases and is sometimes reported in sensory
neural deafness, but rarely after stroke.1 2 We
describe a case of musical hallucinosis in a
professional musician with a left subcortical
haemorrhagic lesion, presumably caused by a
cavernous angioma.
A 35 year old, right handed man was
referred to our inpatient department in July
1999 7 days after the onset of a slight
clumsiness of his right hand followed by
complex acoustic perceptions. The patient
had attended a symphonic concert where an
orchestral transcription of Wagner’s “Siegfried” was played: the patient is a connoisseur
of music and a composer. When he returned
home, about 1 hour later, his musical hallucinosis started. Auditory perceptions were
described by the patient as a symphonic piece
of music performed by an orchestra with
numerous kettledrums and percussion instruments. It was a rather familiar music,
unknown to him, but similar to what he had
heard during the concert. The theme was
played in a minor tonality with frequent use
of drums and other percussion instruments
interspersed with string instruments. A chorus played by string instruments accompanied the theme. The patient said that the
music resembled a piece by the late German
romantic authors (for example, Mahler,
Bruckner, and Wagner’s latest works). The
music was initially low in intensity but
progressively increased; it was perceived in
the middle of his head as if he was listening
with headphones on. Conflicting emotions
occurred: he felt that it was the most
frightening and terrifying music he had ever
heard and strongly desired to push it out of
his mind but, on the other hand, he was
deeply fascinated and said that he would like
to compose such an exciting piece.
The patient said that during his musical
hallucinosis he was able to speak, watch, and
understand television programs and to go
about his normal activities. He reported that
during the phenomenon his hearing was normal and he could hear everything going on
around him, such as the noise outside the
house (for example, from the road) and all the
usual noise going on in his own house.
The musical hallucinosis lasted about 90
minutes and afterwards the patient fell asleep;
he did not have musical hallucinosis during
the next day on awakening and it did not
recur during the next 20 months.
Seven days after the episode the patient
was admitted to our department. On admission a neurological examination evidenced

Figure 1 (A) T1 weighted brain MRI and
(B) anatomical drawing of the coronal MRI
images. An area of altered signal (hyperintense
in centre and hypointense at periphery),
consistent with a haemorrhagic lesion, involves
the left putamen and the external capsula and
just touches the acoustic radiation. Comparison
between the images outlines the strict relation
between the haemorrhagic lesion and the
acoustic radiation (double arrows) that runs
from the medial geniculate body (single arrow)
to the acoustic cortex in the superior temporal
gyrus.
only a slight motor impairment of the right
hand. His hearing sensation was normal on
clinical and instrumental examination. No
signs of drug or alcohol misuse were evident.
Moreover there was no history of psychiatric
disorders.
A cranial CT showed a small hyperdense
lesion on the left temporal lobe at subcortical
level. Brain MRI (fig 1A) evidenced a haemorrhagic lesion involving the left putamen
and the external capsula near the insula. The
lesion was located next to the acoustic radiation (fig 1 A and B). Cerebral angiography
was normal. Three EEG recordings (performed on days 1, 3, and 5 after admission)
highlighted only a mild abnormal slow activity at the temporal level, without epileptiform
grapho-elements. Audiograms and brain
stem auditory evoked potentials were normal.
Transient musical hallucinosis has been
described in several situations, such as
psychiatric disorders, alcoholism, drug and
chemical intoxication, ear and acoustic nerve
diseases and, rarely, brain stem lesions mainly
involving the tegmentum.1 2 Even if musical
hallucinosis has been reported in hemispheric
lesions2 a clear relation with the central
acoustic pathway has never been described.
In our patient the prolonged duration of
the episode, the preservation of consciousness and memory, and the absence of epileptiform abnormalities on EEG rule out an epileptic genesis of musical hallucinosis. In
patients with sensory-neural deafness musical
hallucinosis may be determined by an
increased cortical excitability due to a
deaVerentation phenomena1 or by a spontaneous activation of cerebral areas involved in
musical perception.3
In the present case, it might be directly
related to the impairment of the acoustic
radiation, containing both ascending (excitatory) and descending (inhibitory) fibres. The

www.jnnp.com

inhibitory fibres run from thr auditory cortex
to lower structures of the central acoustic
pathway (medial geniculate nucleus and inferior colliculus) and presumably modulate
acoustic perception. The comparison of brain
MRI (fig 1 A) and of a corresponding
anatomical drawing (fig 1 B) suggests that the
lesion just touches the acoustic radiation
between the left medial geniculate body and
the auditory cortex. Another explanation of
such peculiar findings in our patient may
derive from a recent hypothesis regarding
musical hallucinosis in acquired deafness3:
the subcortical lesion may have caused either
a disconnection between the primary auditory and the association cortices or an
impairment of the “neural networks for the
perceptions and imagery of sounds, including
the auditory association and the frontal
cortex”.3 Indeed, the closeness of the lesion to
the superior temporal gyrus may interfere
with the associative fibres connecting the
auditory cortex to the other cerebral areas
involved in musical perceptions.
Compared with previously reported
cases,1–3 our patient presents several peculiarities. Firstly, the duration of the musical hallucinosis was shorter and the auditory perceptions were heard bilaterally and not
lateralised in the opposite ear. Secondly, it
occurred in the absence of sensory-neural
deafness and might be related to a lesion
involving the central acoustic pathway, even
at a hemispheric level. This is not in
agreement with the notion that complex
acoustic hallucinosis is invariably related to
damage to the peripheral acoustic pathway or
to combined central and peripheral dysfunction.1 2 Thirdly, our report greatly supports
the role of the dominant hemisphere in musical processing, by contrast with the accepted
notion that musical perception is a specific
function of the non-dominant hemisphere.1 4
We can speculate that the musical training of
the patient might have determined the shift of
musical representation from the nondominant to the dominant hemisphere.5
Finally, several features of musical hallucinosis in our patient are fascinating. The similarity between the acoustic perceptions and
the symphonic music that he had previously
heard leads to the hypothesis of an involvement of acoustic memory circuits. The
professional experience and the personal sensibility towards symphonic music might both
have contributed in the determination of
musical hallucinosis influencing the processing of musical sensations.
P CERRATO
D IMPERIALE
M GIRAUDO
C BAIMA
M GRASSO
L LOPIANO
B BERGAMASCO
First Division of Neurology, Department of
Neuroscience, University of Torino, Via Cherasco 15,
10126 Torino. Italy
Correspondence to: Dr P Cerrato

1 Berrios GE. Musical hallucinations: a historical
and clinical study. Br J Psychiatry
1990;156:188–94
2 Paquier P, et al. Transient musical hallucinosis
of central origin: a review and clinical study. J
Neurol Neurosurg Psychiatry 1992;55:1069–73
3 TD GriYths. Musical hallucinosis in acquired
deafness. Phenomenology and brain substrate.
Brain 2000;123:2065–76
4 Kasai K, Asada T, Yumoto M, et al. Evidence for
functional abnormality in the right auditory
cortex during musical hallucinations. Lancet
1999;354:1703–4

J Neurol Neurosurg Psychiatry 2001;71:278–283
5 Bever T, Chiarello RJ. Cerebral dominance in
musicians and non-musicians. Science 1974;
185:537–9.

Cochlear implantation in a profoundly
deaf patient with MELAS syndrome
Cochlear implantation is now an established
technology for restoring hearing in profoundly deaf patients. Adults who have lost all
useful hearing in both ears are suitable for
cochlear implantation if they are profoundly
deaf (generally this implies hearing thresholds of 100 dB nHL or worse, across the frequency range 125 to 8000 Hz), with aided
hearing thresholds worse than 60 dBA for the
frequencies 250 to 4000 Hz and scoring less
than 30% in a test of sentence discrimination,
using their hearing aids and without lip reading. We describe a patient with MELAS syndrome (mitochondrial myopathy, encephalopathy, lactic acidosis and stroke-like
episodes) who became profoundly deaf and
who has successfully undergone cochlear
implantation and rehabilitation.
A right handed secretary with MELAS
syndrome, and a confirmed A to G mutation
at nucleotide 3243 in the mitochondrial
genome, was referred to the cochlear implant
programme of The Royal National Throat,
Nose, and Ear Hospital. She had insulin
dependent diabetes, congenital cataracts,
short stature, leg weakness, fatigue, and hearing loss. She had never had encephalopathy
or strokes. Her mother is also diabetic, has
glaucoma, and has a lesser degree of deafness,
and her sister has been profoundly deaf from
adolescence in addition to having severe
mental retardation. The patient had begun to
experience bilateral hearing loss at the age of
22, with slow deterioration up to the age of
29, by which time she was profoundly deaf in
the right ear. By the age of 30 she was also
profoundly deaf in the left ear and had developed tinnitus. She had no spontaneous
vertigo, but sudden movements could leave
her temporarily unsteady. At the age of 31 she
was referred for assessment for cochlear
implantation. Her ability to communicate
with her family was severely restricted
because of her deafness. She had developed a
modest lip reading ability and was able to lip
read her husband to a limited extent, but
relied greatly on finger spelling and written
information. Her own voice quality had
begun to deteriorate. She was found to have
no measurable hearing thresholds except for
a 250 Hz tone at 105 dB nHL in the right ear.
No tests of speech discrimination were possible, as she had virtually no hearing in either
ear. Middle ear impedence was normal, and
she had normal bilateral vestibular function
on caloric testing. Auditory brain stem
responses and electrocochleography showed
no peaks in response to wide band clicks presented to either ear at 100 dB nHL,
consistent with profound sensorineural deafness. Electrical stimulation of the cochlea,
using sinusoidal stimuli presented through a
transtympanic needle electrode placed
through the tympanic membrane onto the
promontory of the middle ear,1 2 gave rise to a
subjective sensation of hearing in both ears,
with better performance on gap detection
and temporal diVerence limen tests on the
right. A CT scan of the temporal bones was
normal.
The findings listed above showed her to be
within the criteria for cochlear implantation,
and she was implanted in the right ear at the
age of 32 with a Nucleus 22 multichannel

281

implant. All 22 electrodes were inserted into
the cochlea and there were no surgical
complications.
Subsequent switch on and rehabilitation
went well, and the patient has made good
progress. She is able to discriminate environmental sounds well, including diVerent birdsongs, and participate in conversation. Verbal
communication with her family has improved. There have been no specific problems
with the implant and she is able to converse
on the telephone using the implant. The
patient has resumed full time work in an
oYce. The tinnitus has remained stable, and
there have been no vestibular problems. At
the 2 year assessment she scored 97% correct
on CUNY/UCL sentences (a British adaptation of a sentence discrimination test developed at City University, New York), using her
implant and lip reading and 92% correct on
BKB (Bamford, Kowal and Bench) sentences
(another speech discrimination test) using
her implant but without lip reading. Speech
production was within normal limits, although the narrow pitch range reflected her
slightly flat pattern of intonation.
MELAS syndrome was first described in
1984 and is one of a group of mitochondrial
cytopathies, associated with point genetic
mutations. In the brain the characteristic
abnormalities are basal ganglia calcification
and focal lesions of cerebellar and cerebral
atrophy, resulting from cellular rather than
vascular dysfunction.3 Although it does not
feature in the acronym, hearing loss is a common finding in MELAS. Reports of large kindreds and patient series have shown that at
least 50% of patients have a moderate or severe
sensorineural hearing loss: 21 of 28 patients
with MELAS in an Australian series were
deaf,4 as were eight of 14 patients in a British
series.5 The phenotypic expression of the
mutation is subject to at least three constraints;
the percentage of mutant mitochondrial DNA
in the target tissue (which has at most a loose
correlation with clinical lesions),5 the oxidative
stress to which diVerent organs or cell populations are exposed, and as yet unidentified collaborating somatic mutations which enhance
selective aspects of the syndrome.
The cochlea is an organ exquisitely vulnerable to oxidative stress. The outer hair cells
have a precarious, indirect metabolic support
from Deiter cells, and the stria vascularis is
both metabolically very active and nonmitotic, hence further subject to mutation
accumulation. Recently detailed audiological
findings have been reported in 18 patients
with MELAS, and the authors argued that
the hearing loss in their patients was entirely
due to cochlear lesions.4 There were excellent
speech discrimination scores in six of 12
patients with mild to moderate deafness, and
excluding severe and profoundly deaf patients with absent responses, there were normal and symmetric brain stem evoked
responses in 18 of 20 latencies recorded from
10 patients. Promontory stimulation testing
in two patients was normal, and CT and MRI
were reported as showing no lesions which
could contribute to hearing loss.
Central auditory lesions have been reported as a cause of hearing loss in MELAS.
Imaging studies using both CT and MRI
have shown that the occipital and parietal
lobes and cerebellum are the brain regions
most likely to show focal lesions,3 and a perfusion study using 123I-IMP SPECT, and
acetazolamide challenge, showed that patients with MELAS typically have hypoperfusion of the occipital and parietal lobes, with

www.jnnp.com

a significant defect in perfusion reserve.6 A
case report of a patient who died after having
had severe seizures and stroke-like events,
and who had had multiple imaging studies,
showed mild temporal lobe atrophy at
necropsy with associated spongy degeneration of the cortex.7 All cortical regions were
demonstrated radiologically and pathologically to be abnormal in this patient, with the
occipital lobe showing the most marked
hypoperfusion. She had become deaf 2 years
before her marked clinical deterioration.
The patient we report has had no seizures or
stroke-like episodes. Her presenting complaint
was hearing loss, which progressed over 8 years
to profound deafness. Her selection as a
candidate for cochlear implantation was
straightforward, and she has been successful in
adapting to the device and has gained a significant benefit from it. The performance of the
patient in the BKB word tests places her in the
top 5% of adult performers in our patient
series. Another patient with profound deafness
and MELAS, who had had seizures and
strokes, has recently been reported incidentally
in a large series to have been implanted with a
successful outcome, but unfortunately details
were not provided.4
The fact that this patient has gained
considerable benefit from her cochlear implant raises the possibility that other patients
with MELAS syndrome and profound sensorineural deafness could benefit from this
procedure.
D HILL
S WINTERSGILL
L STOTT
B CADGE
J GRAHAM
Cochlear Implant Unit, Royal National Throat, Nose
and Ear Hospital, 330–2 Grays Inn Road, London
WC1X 8DA, UK
Correspondence to: Mr J Graham
1 House WF, Brackmann DE. Electrical promontory testing in diVerential diagnosis of sensorineural hearing impairment. Laryngoscope
1974;84:2163–71.
2 Graham JM, Hazell JWP. Electrical stimulation
of the human cochlea using a transtympanic
electrode. Br J Audiol 1977;11:59–62
3 Sue CM, Crimmins DS, Soo YS, et al. Neuroradiological features of six kindreds with MELAS
tRNA-Leu A3243G point mutation: implication for pathogenesis. J Neurol Neurosurg
Psychiatry 1998;65:233–40.
4 Sue CM, Lipsett LJ, Crimmins DS, et al. Cochlear origin of hearing loss in MELAS syndrome. Ann Neurol 1998;43:350–9.
5 Morgan-Hughes JA, Sweeney MG, Cooper JM,
et al. Mitochondrial DNA (mtDNA) diseases:
correlation of genotype to phenotype. Biochim
Biophys Acta 1995;1271:135–140.
6 Watanabe Y, Hashikawa K, Moriwaki H, et al.
SPELT findings in mitochondrial encephalomyopathy. J Nucl Med 1997;39:961–4.
7 Clark JM, Marks MP, Adalsteinsson E, et al.
MELAS: clinical and pathologic correlations
with MRI, xenon/CT, and MR spectroscopy.
Neurology 1996;46:223–7.

CORRESPONDENCE
Lead poisoning from complementary
and alternative medicine in multiple
sclerosis
In response to the article Lead poisoning from
complementary and alternative medicine in multiple sclerosis,1 we are very concerned that this