British Medical Bulletin 63:183-193 (2002)
© 2002 The British Council
Cochlear implants and brain stem implants
Richard T Ramsden
Department of Otolaryngology, Head and Neck Surgery, Manchester Royal Infirmary, Manchester, UK
 |
Abstract
|
|---|
This chapter describes the development of two implantable prosthetic
neurostimulators which, in the last 20 years, have revolutionised
the management of severe-to-profound sensorineural deafness.
We have witnessed their rapid evolution from the realms of esoteric
laboratory abstraction, with many critics and little perceived
clinical use, to a routine treatment which is safe, effective
and, indeed, cost effective. It is one of the great triumphs
of biomedical and surgical collaboration, and is without any
doubt the greatest ever advance in the treatment of deafness.
|
Introduction
|
|---|
The anatomy and physiology of the auditory pathways have been
described in a previous special issue
1. To summarise, a sound
wave entering the ear causes the tympanic membrane and ossicular
chain to vibrate. The stapes, the smallest and innermost of
the three ossicles moves in-and-out in the oval window and is
the interface between the middle and inner ears. The mechanical
movements of the middle-ear sound-conducting apparatus are transmitted
to the inner-ear fluids and a wave, the travelling wave of von
Bekesy, passes up the cochlea to reach a maximum at a point
determined by its frequency. For the process to continue, the
physical energy of the travelling wave has to be converted into
electrical energy that can be propagated through the auditory
nerve to the brain stem and from there to the higher auditory
centres. This process of transduction occurs in the organ of
Corti. Depolarisation in the inner hair cells initiates transmission
through the first order neurones, the cell bodies of which are
in the spiral ganglion. The synapse with the second order neurone
occurs in the cochlear nucleus which is situated in the lower
pons cranial to the foramen of Luschka. The more cranial nuclear
projections in the brain stem and auditory cortex are complex
and a detailed understanding of their anatomy is not necessary
for the understanding of this essay.
Many disease processes may lead to loss of hair cells in the organ of Corti. The commonest is the natural process of ageing in which there is a progressive loss of cells typically starting in the basal turn of the cochlea and advancing apically, accompanied by a hearing loss that initially affects the high frequencies and, with time, the middle and lower frequencies. These cells are incapable of spontaneous regeneration (although there are hopes that some time in the next few decades neurotropic factors may be identified that might make this dream a possibility). Chronic noise exposure is another good example of progressive hair cell loss. In neither of these conditions does the hair cell loss become complete, so total hearing loss is unlikely and treatment with hearing aids may be effective. There are, however, a number of conditions in which total or near total loss of the organ of Corti may occur and the most common are listed in Table 1. It will be seen that these are all acquired conditions and may affect both adults and children. To these must be added a number of causes of total deafness that are present at birth or soon after birth, and, of these, recessively inherited non-syndromic deafness is the commonest. Whatever the cause, the absence of the organ of Corti prevents transduction and, although all other components of the auditory system may be intact, profound deafness results.
The cochlear implant is a device that takes over the role of
mechano-electric transduction and delivers to the auditory nerve
a processed signal that can be transmitted to the auditory cortex
and interpreted as sound. There are two recognizable components
of a typical cochlear implant system: (i) the implanted electrode
array with associated microcircuitry; and (ii) the external
component that refines the raw signal before delivering it to
the implanted electrode. In the early days, the cochlea was
regarded as inviolable, and fairly simple single channel analogue
systems were inserted on to the surface of the cochlea. The
modern electrode system is multichannel with up to 22 electrodes
and is inserted into the scala tympani of the cochlea. It takes
advantage of the highly developed tonotopicity of the cochlea
with an orderly progression from high frequencies at the basal
end of the cochlea to low frequencies at the apical, like a
piano keyboard. Digitised processing is now almost universal.
The external component comprises a microphone that sends an
electrical signal to the ‘brains’ of the system
– the speech processor. There are many different strategies
employed by the various manufacturers, but the aim of speech
processing is to manipulate the raw signal so that the most
important features necessary for speech recognition are preserved
and delivered to the ear. Underlying early speech processing
strategies was the recognition that there are various bands
(or formants) of spectral energy in speech. Low frequency information
corresponds to the vowel sounds and contributes to the prosodic
patterns of speech. The higher frequencies convey consonant
information which is essential for speech recognition. Modern
strategies try to convey the time-varying spectral patterns
that are characteristic of speech by analyzing the signal in
short time frames and signalling the most prominent spectral
peaks in each frame. The other factor that influences the fidelity
of the signal delivered to the ear is the rate at which speech
is sampled and digitized by the speech processor. Of the current
generation of cochlear implant systems, the maximum stimulus
rate, summed across all electrodes and using pulsatile stimulation
is approximately 96,000 s
-1. The processed signal is transmitted
through the intact skin by a process of inductive coupling.
An external coil is held magnetically over an internal coil
that is part of the implanted component of the system. A microchip
decodes the incoming wave form and directs it to the appropriate
intracochlear electrode or electrodes depending on the frequency
of the sound, and at a rate determined by the specification
of the individual processing programme.
Devices implanted in the body should be reliable, safe and last for many years – in the case of a child, that means life (Fig. 1). Experience gained from pacemaker technology has proved invaluable, particularly in providing water-tight ceramic or silastic sealing systems for the electronics and cumulative failure rates of as little as 3% at 10 years are quoted by most manufacturers. Many implant programmes round the world have been operating for 15 years and most devices have continued to function without trouble. Implants that do fail can be replaced usually without difficulty and without loss in performance. Most of the important recent advances in implant design relate to speech processing strategies rather than the design of the implantable component. This means that performance with an early implant system can be improved by replacing or modifying the external component, rather than removing and replacing the intracochlear electrode. Advances in electrode design have been somewhat less spectacular. At present, the major manufacturers are pre-occupied with developing so-called ‘modiolus-hugging’ electrodes. As their name suggests, these are designed to lie close to the spiral ganglion and to the residual neural elements within the cochlea that the implant aims to stimulate. Their alleged advantages are more precise stimulus delivery and reduced power consumption. Neither of these claims can, as yet, stand up to close scrutiny, and many surgeons have fears that the new electrodes may traumatize the cochlea.
|
Who is suitable for a cochlear implant?
|
|---|
There are two main groups of candidates for implantation: (i)
postlingually deafened adults; and (ii) prelingually or congenitally
deaf children. To understand the issues relating to these two
groups, it helps to understand the process of speech and language
development in the normally hearing child. At birth, the child
is not capable of speech, but by the age of a year babbling
commences and by 18 months recognizable words begin to appear.
From then on lexical, grammatical and semantic skills are acquired
at a staggering rate, and this capacity remains immense during
childhood. During this time, the auditory pathways are in a
state of maximum plasticity. This facility wanes as the teens
approach and is almost lost by mid-teenage. A normally hearing
child, deprived of the sound of human speech during this critical
period, and restored to it in adult life would be unable to
acquire normal speech despite having normal peripheral auditory
function. This phenomenon is similar to that observed in strabismic
children who develop amblyopia
2. Adults who lose their hearing
after the critical period already have a programmed auditory
cortex. A cochlear implant rapidly re-activates dormant neural
networks. This is clearly observable in the many adults who
can converse almost effortlessly within hours of ‘switch-on’.
Additional improvement occurs in many as a result of subsequent
cortical re-organization. For an implant to be effective in
a congenitally or prelingually deafened child, it has to be
inserted while the auditory system is still plastic or programmable.
There is convincing evidence that the earlier a child is implanted
the better the improvement in auditory performance
3. The extent
of neuronal survival is clearly an important determinant of
outcome, and it must also be recognized that electrical stimulation
(from the implant) prevents further neural degeneration. Most
cochlear implant programmes like to implant children as early
as possible usually around the age of 2 years.
Implant candidates are assessed in some detail by the implant team (Table 2). Cochlear implant technology is expensive and the process of rehabilitation involves the skills of many professionals. As a result, the cost to the National Health Service of an implant with assessment, surgery and 2 years of rehabilitation is in the region of £30,000. An appropriate and rigorous selection process is, therefore, desirable. A number of criteria are considered.
|
Degree of deafness
|
|---|
This is always the first and most important consideration –
is the hearing loss bad enough to warrant an implant or could
as good a result be obtained with conventional hearing aids?
Criteria are changing as implant technology improves. Fifteen
years ago, the typical implantee was a postlingually deafened
adult with a pure tone threshold of 100–110 dB and no
speech discrimination (profound deafness). Now, teams are implanting
patients who are still deriving some limited benefit from their
hearing aids. Their maximum speech discrimination score in the
best-aided condition might well be in the region of 20–40%
(using material such as Bench, Kowal, Bamford sentences). In
the paediatric population, there are problems in assessing hearing
thresholds and thus candidacy. Behavioural audiometry may be
difficult or impossible. Objective audiometry, notably the Auditory
Brain Stem Response (ABR), is widely used in assessment, but
cannot as yet give accurate information about the low frequency
thresholds. Speech audiometry is clearly out of the question
in a prelingually deaf child with no lexical base. For these
reasons, assessment of the rate of development of auditory performance
over time is often necessary, particularly acquisition of language
using hearing aids during a trial period which may last several
months.
Age of the patient and duration of deafness
In postlingually deafened adults, age is relatively unimportant. Duration of deafness is, however, and the longer the period of auditory isolation the less good the outcome is likely to be4. It is impossible to set firm rules but a 55-year old, deaf for 40 years is a less favourable prospect than a 70-year-old, deaf for 5 years. The age of a congenitally deaf child is important for the reasons explained above. It is unlikely that a congenitally totally deaf child over the age of 7 years would do well with an implant, and congenitally deaf adolescents are bad candidates.
Imaging
High quality CT and MR imaging is essential to provide details of the anatomy of the cochlea and its connections, and may reveal a number of absolute or relative contra-indications to implantation. Congenital malformations (dysplasias) or acquired conditions such as cochlear obliteration, temporal bone fracture or otosclerosis should be reliably identifiable using the current generation of scanners.
General health
Common-sense rules about suitability for general anaesthesia and reasonable life expectancy apply. Of particular importance is an evaluation of central or cognitive function which may reveal potential difficulties with information processing with the implant. This is an issue after meningitis and with deaf and multiply handicapped children, more and more of whom are being referred for assessment. Each child has to be looked at on his or her merits.
Motivation, expectations and cultural issues
It is important that individuals contemplating implantation for themselves or for their children should have a realistic view of what the outcomes are likely to be. Exalted expectations are often perpetuated by the tabloid press. It is the job of the team to temper enthusiasm with realism, based on what can be predicted from knowledge of the individual subject. It is particularly important for parents to have some idea of the on-going nature of rehabilitation and that hard work is needed by them as well as the rehabilitation team and the child's teachers over a period of many years.
Many of these factors are only relative contra-indications, but taken together they allow the team to give an informed opinion to the patient or to parents of a child about likely outcome, and enlighten the discussions that precede a final decision about whether to proceed with surgery.
|
Surgery
|
|---|
The operation to insert the internal component of the cochlear
implant system, the actual implant, has to be meticulously performed,
but is well within the capability of most otologists. Through
a small postauricular incision, a cortical mastoidectomy is
performed and the middle ear is entered through the facial recess
– the so-called posterior tympanotomy. The stapes can
then usually be easily identified and 2 mm below it is the round
window niche. The scala tympani of the basal turn of the cochlea
is entered by drilling in front of the round window niche with
a 1–1.5 mm microdrill. Usually a perilymph-filled cavity
is encountered into which the electrode array can be gently
introduced. As discussed previously, the latest generation of
electrode arrays has a mechanism to carry it close to the modiolus
and the spiral ganglion. The package comprising the receiver
coil and the microchip is recessed into a bony well in the outer
table of the skull, and the overlying pericranium usually provides
sufficient stability without the need for anchoring ties. The
operation takes about 1.5 h and, at the end of the procedure,
the electrical integrity of the device can be tested by a number
of measures including neural response telemetry. Using this
technique, whole nerve action potentials are recorded from the
auditory nerve in response to stimulation of individual electrodes
within the cochlea. The information thus obtained may help the
rehabilitation team in the initial estimation of threshold values
of stimulation at the time of switch-on of the device, as well
as confirming the integrity of each channel in the array. The
incidence of postoperative complications is very low. Although
the facial nerve is in the surgical field, as indeed it is with
most tympanomastoid surgery, damage to it is very rare with
an incidence of under 0.5%.
|
Special surgical problems
|
|---|
Surgical difficulties may be encountered in a number of well
recognized conditions.
Inner ear dysplasia
The most extreme form of dysplasia, aplasia, is a contra-indication to implantation and would be picked up on pre-operative imaging. At the other end of the scale, the Mondini deformity and the large vestibular aqueduct syndrome are usually not difficult to implant. Between these extremes, the common cavity deformity in which there is little differentiation of the cochlea beyond the primitive otocyst stage is challenging, with the possibility of inadvertently inserting the implant into the posterior cranial fossa through the inner ear. Associated with this is a high risk of CSF fistula. Furthermore, in such cases of dysplasia, the facial nerve may be abnormally placed and be at increased risk. Aplasia of the auditory nerve is rare, but needs to be recognised (on MR imaging) as it is an absolute contra-indication to cochlear implantation.
Cochlear obliteration (osteoneogenesis)
In a number of conditions, the cochlear lumen may become obliterated, either by fibrous tissue or new bone, making insertion of the implant difficult, if not impossible. Meningitis, otosclerosis and skull-base fracture are associated with the deposition of new bone. Autoimmune ear disease may be associated with intracochlear fibrosis. Modified surgical techniques and electrodes have been developed to deal with these problems.
Chronic middle ear and mastoid disease
Insertion of a foreign object into the body is contra-indicated in the presence of active infection. Tympanomastoid disease, if present, including cholesteatoma must be eliminated at a first stage operation and the implant inserted at a subsequent date.
|
Switch-on and tuning
|
|---|
About 1 month after the operation, when the skin is well healed,
the implant is connected to the external component and tuned
up. Each electrode in the cochlea has an electrical threshold
at which the stimulus is just perceived as sound. It also has
a higher level at which the stimulus just ceases to be comfortable.
These are the T and C levels and the difference between them
is referred to as the ‘dynamic range’.
The process of establishing T and C levels for all 22 electrodes and entering them into the memory of the speech processor is called ‘mapping’ and requires the skills of the audiological and rehabilitation members of the cochlear implant team. Mapping is usually straightforward in adults, but can require considerable skill and patience with small prelingual children whose lack of language means that they have to be programmed using conditioning techniques. If one sets the levels too low, nothing will be heard; if too high, a non-auditory response such as pain or facial twitching may occur, which will upset the child and ensure the end of his or her co-operation in the whole mapping process. Mapping has to be repeated on a number of occasions in the first weeks and months since the psychophysical features of the auditory system change as a result of stimulation by the implant.
|
Outcomes
|
|---|
Nearly all implanted postlingually deafened adults achieve some
degree of open set speech understanding using the implant alone
(
i.e. without lip reading); most achieve a high level of performance
within a matter of weeks and can use the telephone with a fair
degree of proficiency. Cochlear implantation in adults has been
shown to be cost effective. Summerfield and Marshall calculated
the cost per quality adjusted life year (QALY) for cochlear
implantation and a number of other common conditions. As can
be seen in
Table 3, cochlear implantation compares very favourably
5.
View this table:
[in this window]
[in a new window]
|
Table 3 The cost per quality adjusted life year (QALY) for cochlear implantation and a number of other common conditions
|
|
It is becoming increasingly clear that the majority of congenitally
or prelingually deaf children implanted by the age of 2 years
are able to take their place in main stream schools after 3
years of implant use and training, albeit with some degree of
support. Speech perception and production continue to improve
with time and most develop the accents and modulations of their
geographical regions and peer groups.
|
The auditory brain stem implant (ABI)
|
|---|
This development from cochlear implant technology is indicated
for totally deaf individuals who have no auditory nerves and
who are thus not candidates for a cochlear implant. In practice,
this patient group comprises almost exclusively sufferers from
neurofibromatosis type 2 (NF2) who have been deafened as a result
of bilateral vestibular schwannomas (acoustic neuromas) or from
the surgery to remove them. Other possible indications may emerge
such as cochlear nerve agenesis or the unimplantable cochlea
from excessive ossification. The ABI has an electrode carrier
with 20 small disc electrodes (
Fig. 2) and is inserted on to
the surface of the cochlear nucleus in the lateral recess of
the fourth ventricle, accessed through the foramen of Luschka.
Technically, this is not easy, as surgical landmarks are not
always obvious. The correct position of the implant is verified
by eliciting the electrically evoked auditory brain stem response
(EABR). Adjacent cranial nerves (facial, glossopharyngeal, accessory
and trigeminal) are monitored to minimise the risk of non-auditory
stimulation. From the point of view of neuro-anatomy, there
is a major problem with the frequency maps, or tonotopicity,
of the cochlear nucleus compared with the cochlea. A surface
electrode will function most effectively if the frequency map
is distributed across the surface of the nucleus. In the cochlear
nucleus, the map is disposed obliquely through the depths of
the nucleus, and to take advantage of this arrangement a penetrating
electrode has been developed but as yet it has not been used
in clinical trials.
Multicentre trials with the surface electrode array have been
carried out in North America and in Europe
6,7. They indicate
that outcomes with the ABI are not as good as typical cochlear
implant results. Nevertheless, most patients gained an awareness
of environmental sounds and found the ABI enhanced their lip
reading scores. A small, but important, number obtained reasonable-to-good
open set speech perception using the implant alone and some
obtain limited telephone use. Unwanted non-auditory side-effects
such as facial twitching, pain in the throat, face or body may
occur with some electrodes, but these can be programmed-out
at mapping sessions.
|
Future developments
|
|---|
The near future should see the development of totally implantable
devices that use either an intrinsic power source or implanted
batteries that can be recharged remotely through intact skin.
Similarly, remote re-programming would be possible. Research
is also well advanced into mechanisms for delivering drugs or
neurotrophic factors to the cochlea and auditory system through
the intracochlear electrode
8. Of course, if hair cells could
be encouraged to regenerate, which again is the subject of much
research, cochlear implantation would become a curiosity of
medical history.
|
Footnotes
|
|---|
Correspondence to: Prof. Richard T Ramsden, Professor of Otolaryngology,
Department of Otolaryngology, Head and Neck Surgery, Manchester
Royal Infirmary, Oxford Road, Manchester M13 9WL, UK
|
References
|
|---|
- Haggard MP, Evans EF. Hearing. Br Med Bull 1987; 43
- Ryugo DK, Limb CJ, Redd EE. Brain plasticity. In: Niparko J. (ed) Cochlear Implants: Principles and Practices. Philadelphia, PA: Lippincott Williams and Wilkins, 2000
- O'Neill CO, O'Donoghue GM, Archbold SM, Nikolopoulos TP, Sach T. Variations in gains in auditory performance from paediatric cochlear implantation. Otol Neurotol 2002; 23: 44–8[ISI][Medline]
- Summerfield AQ, Marshall DH. Cochlear Implantation in the UK 1990–1994. London: HMSO, 1995; 144
- Summerfield AQ, Marshall DH. Cochlear Implantation in the UK 1990–1994. London: HMSO, 1995; 227–8
- Otto SR, Shannon RV, Brackmann DE et al. The multichannel auditory brainstem implant: performance in 20 patients. Otolaryngol Head Neck Surg 1998; 118: 291–303[Medline]
- Nevison B, Laszig R, Sollmann W-P et al. Results from a European clinical investigation of the Nucleus multichannel auditory brainstem implant. Ear Hear 2002; In press
- Clark GM. Cochlear implants in the third millennium. Am J Otol 1999; 20: 4–8[ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
Top
Abstract
Introduction
