Technical devices for hearing-impaired individuals: cochlear implants and brain stem implants - developments of the last decade.

Over the past two decades, the fascinating possibilities of cochlear implants for congenitally deaf or deafened children and adults developed tremendously and created a rapidly developing interdisciplinary research field.The main advancements of cochlear implantation in the past decade are marked by significant improvement of hearing and speech understanding in CI users. These improvements are attributed to the enhancement of speech coding strategies.The Implantation of more (and increasingly younger) children as well as the possibilities of the restoration of binaural hearing abilities with cochlear implants reflect the high standards reached by this development. Despite this progress, modern cochlear implants do not yet enable normal speech understanding, not even for the best patients. In particular speech understanding in noise remains problematic [1]. Until the mid 1990ies research concentrated on unilateral implantation. Remarkable and effective improvements have been made with bilateral implantation since 1996. Nowadays an increasing numbers of patients enjoy these benefits.

1. Introduction

One of the biggest steps forward in modern medicine is the possibility to replace a deaf ear, i.e. a sensory organ, with an implantable electronic prosthesis [2]. The development of cochlear implants (CI) has been the most important step for improving communication in congenitally deaf or deafened children, adolescents and adults in the last 200 years, if the foundation of the schools for the hearing impaired in Paris (1750) and Leipzig (1778) were taken as reference points.

Today, owing to cochlear implantation, congenitally deaf children or children with residual hearing can achieve almost normal auditory-verbal development.

In the early 1970ies cochlear implants typically allowed the auditory connection to the environment. With the technical possibilities of that time they were an enormous communicative remedy for many hearing-impaired individuals. Increasing knowledge of auditory physiological processes and the function of cochlear implants as well as their technical advancement now help many patients to reach open speech understanding which was considered impossible in earlier times [3], [4].

An important milestone significantly improving speech understanding in cochlear implant patients is certainly the CIS strategy (CIS = Continuous Interleaved Sampling) [3] published by Wilson in 1991 [5]. In clinical practice this strategy with high stimulation rates has helped many patients to improve their speech understanding [3], [6], [7], [8], [9], [10], [11], [12], [13]. Meanwhile, all cochlear implant manufacturers have implemented more or less modified and different "CIS strategies" in their implants [13].

Over the past two decades, influenced by innovative ear surgeons, the fascinating possibilities of cochlear implantation for congenitally deaf or deafened children and adults have created a rapidly developing interdisciplinary research field combining the efforts of specialized ear surgeons, engineers, physicists, technicians, audiologists, electro-physiologists, speech and language therapists, psychologists, teachers and, last but not least, CI users. This paper reviews, according to the wish of the president, Prof. Dr. Beleites, mainly recent developments of the past decade.

Progress in the area of cochlear implantation over the last decade is marked by the significant improvement of CI users' hearing and speech understanding. These high standards reached by this development are reflected by supplying more (and increasingly younger) children and the possibilities of binaural hearing with modern cochlear implants.

Improved speech understanding is attributed to advanced speech coding strategies [13], [1]. Despite this progress, modern cochlear implants do not yet enable normal speech understanding, not even in our best patients. Speech understanding particularly in noise remains more or less difficult [1].

Until the mid 1990ies research concentrated on unilateral implantation [14]. Remarkable, effective improvements have been made with bilateral implantation since 1996 [13], [1], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24].

2. Preliminary note: basic principles of cochlear implants

For the benefit of those who are not so familiar with the subject, this review begins with the basic principles of modern cochlear implants.

In most cases deafness is caused by cochlear disorders, e.g., loss or degeneration of sensory hair cells. Mechanical sound waves can no longer be converted into electrical stimulation.

A cochlear implant (CI) stimulates the auditory nerve directly electrically thus replacing the defective natural stimulus transmission of the sensory cells of the inner ear to the auditory nerve. The non-functional cochlea serves only as a place holder to keep the electrode in the most optimal position to the auditory nerve [3], [11]. The CI system takes over the functions of the middle and inner ear: Sound is picked up by the microphone, processed by the external speech processor and wirelessly transmitted via a coil to the implanted receiver. The speech processor codes the sound signal, e.g. speech or music, as a sequence of electrical pulses and relays those pulses to the stimulation electrodes in the cochlea (Figure 1 (Fig. 1)).

Figure 1

Function of a cochlear implant system:

Sound is picked up by a microphone (1), processed by the external speech processor (2) and wirelessly transmitted by a transmitter (3) to an implantable receiver (4). The electrical signal, coding e.g. speech or music as a defined sequence of electrical pulses, is then relayed to an electrode (5). The electrode is inserted through the mastoid, the facial recess, past the facial nerve through the middle ear into the non-functional cochlea (6). The electrode carrier stimulates the auditory nerve. The brain decodes the relayed neural impulse and the original signal is recognised as speech or music (schematic taken from product information).

The electrode is inserted through the mastoid and the facial recess, passing the facial nerve, into the non-functioning cochlea. The electrode carrier has several electrode contacts which are positioned at different places in the cochlea to enable tonotopic stimulation of the auditory nerve. Fast series of pulses via several channels allow effective use of the temporal principle beside the tonotopic principle. This leads to natural sound and intelligible perception of the coded speech signals. Fast speech coding strategies, like the CIS strategy (CIS = Continuous Interleaved Sampling), are based on B.S. Wilson's [5] design. Important for the effectiveness of the implanted system is the pulse rate which should be at least 1500 Hz per channel. Another decisive factor seems to be that all pulses are newly calculated from the speech signal so that each pulse conveys new information. Compared to constantly newly calculated pulses, several repetitions of an identical pulse cause information loss [3].

Speech understanding depends on the way the auditory nerve is stimulated along the axis of the bony cochlea (modiolus). Since patients perceive usable and desired auditory sensations also with electrodes placed in the very tip of the cochlea, it seems plausible to assume that the entire length of the cochlea should be used if possible [11], [25]. Detailed investigations show that four to eight stimulation electrodes should be distributed as widely as possible along the cochlea to achieve good and stable speech results [3], [26], [27].

3. Development of pre-operative diagnosis and indications

The CI system codes information of incoming sound signals based on the speech coding strategy and converts those signals into bioelectrical pulses which electrically stimulate the auditory nerve. A CI is thus suitable for patients with insufficiently functioning cochleae. The auditory nerve itself must, however, not only be intact anatomically but also excitable to convey the coded information.

3.1 Preoperative diagnosis

Radiological diagnosis before implantation should include high-resolution petrosal bone computer tomography (CT) and, especially in children, magnetic resonance imaging (MRI). Information obtained with those imaging techniques is complementary [28], [29].

Functional magnetic resonance imaging (fMRI) can be used to display electrically evoked auditory sensations [30]. This method takes advantage of the different magnetic characteristics of oxygenated and deoxygenated blood (BOLD effect). Cortical metabolism increases during electrical stimulation, the activated area responds with a regionally increased blood flow. This causes a shift in the relation of oxygenated and deoxygenated haemoglobin. Images taken at two different points of time (in resting and stimulated state) are compared with statistical methods and the differences (depending on the stimulated areas) spatially matched.

Clinical application of this helpful tool is currently limited to individual cases due to long calculation times [30].

A promontory test before cochlear implantation is not used in general. Its prognostic value is uncertain. However, it has a high demonstrative value for patients and parents. When using an ear channel electrode instead of a promontory needle, it is non-invasive and not too much strain for the patient. It is an easy and fast method, particularly when combined with loudness scaling in analogy to the loudness scaling "Würzburger Hörfeld" [31].

3.2 Extending implantation criteria: cochlear implantation not only in patients with complete deafness but also in patients with residual hearing

Considering the positive results of cochlear implantation, a cochlear implant is not only indicated in cases of complete deafness. New implantation criteria also include "insufficient speech understanding" with correctly and optimally fitted hearing aids. Insufficient speech understanding is defined by the speech understanding levels achieved with modern cochlear implant systems [4], [32].

Beside many prognostic factors influencing the result of cochlear implantation, duration of deafness and residual hearing [33] are considered to be important. Patients with residual hearing may even be able to talk to strangers on the telephone after cochlear implantation and reach, on average, higher postoperative results (80% in the Freiburg Monosyllable Test); their benefit from a cochlear implant is higher-than-average. Particularly speech understanding tests in noise, which resemble everyday situations more closely, show more effective improvements for patients compared to earlier hearing aid use [7], [33], [34], [35], [36], [37]. Patients should be informed about the possibilities with modern cochlear implants as early as possible and should receive a cochlear implant before they are completely deaf.

Based on the good hearing results, cochlear implantation in adults is now even considered for individuals who no longer understand 30 - 40% of the words in the Freiburg Monosyllable Test at 70 dBSPL [3], [33], [34], [35], [36], [38].

Derived from these criteria, clinical practice indicates that cochlear implantation should be considered if a patient can no longer talk to strangers on the telephone even with optimally fitted hearing aids. Individual assessment of optimal supply with hearing aids requires a suitable trial period and hearing training if necessary.

3.3 Cochlear implantation in children

After cochlear implantation had been proven beneficial for adults and the surgical procedures had been found save, the next logical step was to implant (at first older) children. Before deciding on implanting a child it was and still is obligatory to use adequate hearing aids.

In view of the biologically important but temporarily limited time frames of auditory-verbal development, it became necessary to implant children at increasingly younger ages.

3.3.1 Age at implantation

From a physiological point of view very early implantation seems to be desirable [39], [40]: Particularly during the second babbling phase, starting at approx. 6 months of age, until about the end of the second year the synapses in the left temporal lobe multiply explosively allowing us to learn our mother tongue. Development of this speech competence naturally requires (possibly long) learning processes. Hearing experiences from the first day of life on are crucial for these learning processes [41].

Many publications have meanwhile investigated the results after cochlear implantation in children [3], [39], [42], [43], [44], [45], [46], [47], [48], [49], [50], [51], [52], [53], [54], [55], [56], [57], [58], [59], [60], [61], [62], [63], [64], [65], [66], [67], [68], [69], [70], [71], [72], [73]. The chosen age limits of the groups are rather incidental due to the number of patients of each age group implanted in the course of time and the resulting data for analysis. Independent of the chosen age limits these investigations have proven the benefits of early implantation with regard to postoperative hearing abilities and auditory-verbal development of the tested children [42], [43], [45], [46], [49], [53], [58], [59]. The children's auditory and verbal performance was the better, the earlier they were supplied with CI [45]]. Early implanted children, i.e. children implanted before 24 months, not only scored higher in the tests on auditory-verbal development, they developed these abilities faster compared to children implanted later [45], [73]. Fortunately, all children benefit from cochlear implantation independent of their age at implantation compared to the use of hearing aids. To reach optimal results it is necessary, however, to aim at implantation before 24 months or, even better, before 12 months of age. Even though implantation before the age of 4 years seems to prevent irreversible alterations and deficits in the maturation of the auditory pathways, children of this age group do not reach the level of auditory-verbal development reached by earlier, i.e. younger, implanted children [73], [74].

Children should, therefore, be implanted and (re)habiliated at a very young age. An important step toward identifying hearing impairment or deafness in very young children to allow for early implantation has been made with the establishment of newborn hearing screening.

This way, parents can and must be informed about adequate therapy programmes within the first 24 months of their child's life. These first 24 months are crucial for the child's next seven or eight decades of further personal, professional and social development. Opportunities missed during this time cannot be compensated later [75].

While cochlear implantation in very young children was only rarely practiced in the USA [76], [77], physicians in Europe and particularly in Germany were more willing to offer cochlear implantation for younger children [9], [78], [42].

So far, only large centres in Europe and particularly in Germany have worked with children between four and twelve months of age, if, and only if, the diagnosis was certain or the reason for deafness was meningitis. Compared to the total number of implanted children, the percentage of so early implanted children is not as high as desirable for various reasons. According to Offeciers only 4.1% of all children implanted in Germany received their CI before the age of 18 months [79].

Our current observations of 48 children (15%) who received their CI under the age of 18 months (21 of those children were younger than 12 months when implanted in Würzburg) show that the efforts required of all specialists involved in CI supply to enable early implantation seem justified and desirable. These children's development, particularly in case of bilateral implantation, is measured against the development of normal hearing children [80], [24], [17]. Long-term observations of larger collectives are still missing because these children, who were between six and twelve months old, have only recently been implanted. Investigations of speech production of very early implanted children indicate the expected, early developing positive effects [73].

These observations raise the question of the optimal time or optimal time frame of cochlear implantation in children. Such considerations are not intended to initiate a competition among surgeons for the "youngest implanted child". The objective is rather to evaluate from a biological point of view when surgery can be performed to reach the best results while being of minimal risk.

3.3.2 Surgical aspects of CI surgery in children

Indication shall be based on each child's individual developmental status: The children should be able to control their heads to avoid blows to their heads. This reduces the risk of damaging the external and internal components of the CI.

From a surgical point of view it seems desirable to have a defined tension of the M. sternocleido-mastoideus at the tip of the mastoid which is associated with increasing head control. This tension is required for the development of the mastoid process [81]. This way the experienced surgeon finds enough space for an access to the cochlea even in a child's small mastoid to identify and preserve the delicate anatomic structures, especially the facial nerve (Figures 2 (Fig. 2) and 3 (Fig. 3)).

Figure 2

Intra-operative site during cochlear implantation in an 8-month-old infant (own image)

Opening of the facial recess, facial nerve covered by bone, blood-forming bone marrow around the facial nerve channel. 1.2 mm suction tool on the left.

Figure 3

Dimensions of mastoid of 6-month-old infant during cochlear implantation compared to sterile 1-Eurocent coin.

Surgery itself is planned by an interdisciplinary team including all the specialists involved in the support of the child, e.g. anaesthetists and paediatricians to guarantee optimal care for the anesthetized child before, during and after surgery. Intravenous access for example, can be challenging in infants. This shows that it is important to consider even obviously minor details carefully and in an interdisciplinary team when working with infants. Beside the special anatomic situation in children, their small respiratory pathways and relatively low blood volume have to be considered. An average-size, 6-month old infant weighing 8 kg has an average blood volume of only 640 ml (80 ml blood/kg body weight).

4. Surgical technique

The surgical technique of cochlear implantation is considered safe and without major complications [3], [67], [82], [83], [84].

Conservative surgery and atraumatic insertion of the electrode are self-evident and correct [32]. Lehnhardt coined the term "soft surgery" [84] for his surgical technique. This principle does not have any drawbacks, but Probst finds one disadvantage in the probably erroneous assumption that "soft surgery" will definitely preserve residual hearing and that this residual hearing may be used later [32].

4.1 Surgical risks

The literature describes up to 2.1% facial pareses, 0.6% of which are permanent [82], [85]. Fortunately, this risk has not been found in the more than 750 Würzburg patients implanted since 1994 (age at implantation 4 months to 86 years).

The fundamentals of safe cochlear implantation are reliable surgical techniques based on the principles and established rules of middle ear surgery [83]. The surgical technique varies only slightly in different clinics [3].

Already at primary implantation it has to be considered that cochlear implants are technical devices with limited lifetime which will eventually have to be replaced despite their high quality [86]. Based on observations with (rare) revision surgeries, we regard it as reasonable to sufficiently open the mastoid during initial surgery and to identify the facial nerve. A widely open facial recess allows the necessary view into the tympanum and the promontory with the cochleostomy.

4.2 Alternative surgical techniques

The detailed discussions of surgical techniques are complemented among others by the pros and cons of shaving, skin incision or cochlear implantation without mastoidectomy [87], [88], [89], [90], [91], [92].

The various surgical techniques available can be adapted to meet individual situations. Selection and application are at the surgeon's discretion. Individual techniques should not be seen too dogmatically because they are not too relevant for the overall results as they may seem to be when following the extensive discussions in the literature [3].

4.3 Cochlear implantation under local anaesthesia

The good results of cochlear implantation seem to justify surgery under local anaesthesia in patients with cardio-pulmonary problems [93].

5. Development of the electrode carrier

Another increasingly investigated branch concerns the electrode carrier of cochlear implant systems. Beside insertion trauma, design and possible insertion depth are subject of ongoing scientific discussion. Some aspects of this issue have not yet been completely solved [94], [95], [96], [97], [98], [99], [100], [101], [102], [103], [104], [105], [106], [107], [108].

The electrode is an important part of the cochlear implant system: It provides the stimulation pulses and stimulates the auditory nerve. It has not yet been clarified which neural structures are stimulated inside the cochlea.

Depending on system design the electrode carriers have 8 to 24 single electrodes. Their arrangement on the electrode carrier varies for individual CI systems with regard to the length of the electrode section used for stimulation and the distance between single electrodes. A compromise between desired stimulation effects and undesirable artefacts caused, e.g., by interfering electrical fields of closely adjacent stimulation electrodes, is necessary. Given modern technical possibilities it seems to be impossible to arbitrarily increase the number of electrodes. Systems with e.g. 24 electrodes use only about 8 to 10 electrodes per stimulation cycle.

Modern technology seems to confirm an optimal number of eight to twelve electrodes [3], [26], [27], [109]. Good speech understanding in noise needs more channels compared to a situation in quiet, but even in noise 12 channels are considered more than adequate. Using e.g. 16 channels does not improve hearing [26], [27].

5.1 Deep electrode insertion

Probably based on the assumption that the bodies of the spiral ganglia are only distributed over 16 - 18 mm along the basal turn of the cochlea, an electrode insertion depth of 16 - 20 mm was generally considered sufficient. Helms showed already in 1994 that the electrodes can be routinely inserted about 30 mm up to the second turn of the cochlea (Figure 4 (Fig. 4), Figure 5 (Fig. 5)), however, the question of the benefit of this procedure, later called "deep electrode insertion" by Gstöttner [99], remained open for a long time. Patients with "deep insertion" had very good speech understanding [8].

Figure 4

Examples of complete electrode insertion for different CI systems (from left to right: Clarion implant, Nucleus implant, MED-EL implant). The right image shows a deep electrode insertion.

Figure 5

CT sequence before implantation of contralateral side: deep electrode insertion. Electrode carrier placed in second cochlear turn.

This gives rise to the question if the many patients with deep insertion actually benefit from electrically stimulating the second turn of the cochlea. First studies [25], [110], [111] indicate that deep electrode insertion, i.e. more than 30 mm, and the associated electrical stimulation of the apical section of the cochlea contribute substantially to speech understanding in quiet and in noise. Less deep insertion showed no differences in speech understanding when comparing 22 mm to 25 mm insertion depth [112], which may, however, be due to the fact that the active area of the electrode used for stimulation was the same, i.e. 17 mm in both patient groups. Simulations of Dorman et al. [111] confirm the benefits of deep electrode insertion.

5.2 Perimodiolar electrode carrier

Another possible method to improve speech understanding is the placement of the electrode carrier closer to the modiolus. Certain neurone populations are expected to be stimulated more specifically, assuming that the closeness to the stimulated tissue enables more differentiated stimulation [113]. Preformed electrode carriers are used to reach this goal. They are inserted into the cochlea with special instruments and take their intended shape inside the cochlea after removing the insertion tools. Since the same preformed electrodes are used for all patients, individual characteristics in the anatomy of the cochlea cannot be accommodated (Figure 6 (Fig. 6)). Postoperative studies show that electrode placement is not always satisfactory. Displacement of the electrode carrier from the Scala tympani into the Scala vestibuli happens more frequently than expected. The hoped for improvement of speech understanding could not be demonstrated, although perimodiolar electrodes allow lower stimulation thresholds [104].

Figure 6

Different diameters and slopes of cochlear turns to illustrate individual variations of the human cochlea

(with permission of Helge Rask-Andersen, Uppsala, Sweden).

6. Cochlear implant and meningitis

Frequent application of a treatment, increasing patient numbers and associated long-term experiences principally increase the risk of observing undesirable side-effects or complications. Expanding indications and increasingly implanting young children have shifted the therapeutic focus to a patient group with frequent middle ear problems. Serous otitis media and acute otitis media are often observed during childhood. Complications like mastoiditis and otogenic meningitis may develop from an otitis media and are still life-threatening conditions. Even children without CI may develop meningitis: In Germany for instance as many as 250 children may contract pneumococci meningitis each year.

Unfortunately, several cases of meningitis after cochlear implantation have been observed, some leading to the patient's death. Fortunately, this risk has been seen in only very few patients.

The risk to contract otogenic meningitis after cochlear implantation is very low but meningitis may manifest itself also in children with cochlear implants. In these patients, meningitis can be causally related to the implant or occur independent of implantation.

Analysis of this problem [114], [115], [116], [117], [118], [119], [120], [121] showed that malformations of the inner ear and deafness caused by meningitis increase the risk of meningitis [115]. Some cochlear implant systems are more often connected with cases of meningitis than others [114]. Perimodiolar electrode carriers consisting of two components have been identified as possible reasons for increased implant-associated risks. The additional place holder is intended to press the electrodes as closely as possible to the modiolus. The risk of meningitis is higher for these electrodes than for other electrode carriers. As shown by histological investigations and in petrosal bone models, perimodiolar electrode carriers may cause intracochlear lesions [105], [106], [107].

A differentiated discussion of the problems of (otogenic) meningitis after cochlear implantation is given in Arnold et al.: Mastoiditis or meningitis after implantation must be treated surgically like any other inflammatory ear disease [115].

Particular attention must be paid to avoiding life-threatening complications. The risk of otogenic meningitis must be minimized before, during and after surgery to optimally prevent these dreaded complications. In addition to the smallest possible cochleostomy accommodating the electrode diameter, it is recommended to seal the electrode at the site of entry at the promontory with lint-free connective tissue similar to a stapes prosthesis. A thicker "stopper" on the electrode to plug the cochleostomy is also helpful [83].

Another important preventive measure is the inoculation against meningitis after consultation with the paediatrician. It offers a certain degree of protection, but not 100% [75].

Special care must be taken to consequently and early treat serous otitis media, acute otitis media or mastoiditis. In cases of dubious ear aches or fever patients should immediately be referred to an ENT specialist or, if the patient is a child, also to a paediatrician.

7. Modern speech coding strategies

The introduction of speech coding strategies with fast stimulation rates has significantly improved speech understanding in cochlear implant patients [13], [1], [122] and has meanwhile proven its value in clinical practice [8]. Initial objections against fast stimulation derived from animal experiments did not hold out against further verification [123], [124].

Meanwhile, all cochlear implant manufacturers have implemented a more or less fast CIS strategy or modification thereof in their systems [13]. The results also depend on the technical realization of the CIS strategy which has been solved by various manufacturers in different ways. Stimulation and actualisation rates differ for each manufacturer. Differences in the transmitted information content are due to the fact that not all systems calculate each stimulation pulse anew [3].

The mentioned differences of various designs [122] are not always easy to see. The companies' marketing strategies tend to show those differences in a distorting light and even the expert finds it difficult to detect them. Most physicians are overwhelmed with the task as are most patients. The patient's "selection" of a certain implant type is quite questionable and patients should not be left alone with this task unless they insist. Patient-targeted marketing strategies of CI manufacturers are quite dubious; however, as stated by Probst, tendencies into that direction cannot be overlooked [32].

Speech results of unselected patient collectives with established tests, such as the Freiburg Monosyllable Test, which are sufficiently difficult to differentiate patients, should be published. The distribution function at defined time intervals after implantation could show a patient's chances of reaching the best possible speech understanding with a certain speech coding strategy.

8. Speech processor fitting

Extending implantation criteria to include increasingly younger children leads to another aspect of cochlear implantation, i.e. speech processor fitting which is considered particularly difficult for congenitally deaf children and requires much experience [75].

It is not surprising that we look for objective methods in this area [32]. Beside stapedius reflex measurements we hope for additional objective intracochlear measurements of nerve potentials using the cochlear implant electrodes. Compound action potentials are considered an objective response of the neurones to the stimulation pulses emitted by the individual electrodes on the electrode carrier. Such intraoperative measurements should not be accepted without critical interpretation [32]. Mathematical correction factors are used to obtain information about the expected stimulation threshold during first fitting through intraoperative measurements. However, speech processor fittings obtained with objective methods are not yet equal to those determined with behavioural audiometry [125], [126], [127], [128].

9. Bilateral cochlear implantation

Having two ears which enable us to hear binaurally are the prerequisite to cope with many everyday hearing situations [129]. Directional hearing, spatial hearing and signal source separation are only possible with two ears. Binaural hearing supports speech understanding in noise and supplies adequate auditory quality.

First results with bilaterally implanted patients in the early 1990ies in Australia were not very encouraging [130]. A fusion of the auditory impression could be reached but speech understanding was not relevantly improved. In some tests speech understanding was even worse with two implants [130]. After these results the Australian workgroup did not pursue the approach of improving hearing with binaural cochlear implantation.

In 1996 the Würzburg workgroup under Univ.-Prof. Dr. Dr. J. Helms was the first worldwide to succeed in significantly improving speech understanding in quiet and in noise and restoring directional hearing in a bilaterally implanted patient 4 weeks after first fitting [14].

After verifying those first results in adults, children have consequentially been bilaterally implanted since 1998. The convincing results were brought to the attention of a broad scientific public in an announced discussion in 1999, hoping to make this technology available to more children [19].

Animal experiments support the neural-protective effect of electrical stimulation for the auditory nerve [131] and maturation of the auditory pathways. Although it was not yet clear when implanting the first children bilaterally if and how these children would benefit from the second implant, it seemed logical to assume binaural hearing to be good for the children based on the encouraging results in adults and in analogy to bilateral hearing aid supply. Bilaterally implanted patients benefit from sustained improvements of speech understanding in quiet and in noise (20% better speech understanding in quiet in the Freiburg Monosyllable Test, 30% better speech understanding in noise) [19], [20], [21], [24], [132], [133]. The effects and improvements of speech understanding in quiet and in noise seen in adults are also observed in children [24].

Major features of binaural hearing are spatial hearing, directional hearing and separation of signal sources. It was exciting to see if binaurally implanted children would develop spatial and directional hearing skills. In her dissertation C. Edelmann observed the first 13 bilaterally implanted children for a period of 3 years. She found directional hearing skills in children to develop at different rates. It took an average of 1.5 years for the children to develop directional hearing. Over the 3-year period not only directional hearing developed but the failure rate of judging direction decreased with increasing use of both ears. The results could be interpreted to indicate that the children not only learn to localize a sound source in space but also to more accurately localize this sound source.

Individuals with normal hearing achieve directional hearing using interaural time and level differences, deflection phenomena and sound reflections at the pinna. Bilateral CI users benefit - like normal hearing individuals - from interaural time and level differences [134]. As in normal hearing individuals, head shadow effect, squelch effect and binaural summation facilitate binaural hearing in CI users and contribute to their binaural performance [132], [133].

It is, without doubt, necessary to act immediately in cases of possible obliteration of both cochleae after meningitis, as it may be impossible to place the electrode inside the cochlea at a later time. Bilateral, simultaneous implantation should be considered unless medically contraindicated.

From a medical point of view it is not acceptable to refuse congenitally deaf children substantial improvement of their hearing situation for economical reasons as claimed by health insurance companies. Based on the low number of cochlear implantations performed in Germany it is ethically not justifiable to refuse affected individuals substantial improvement of their hearing abilities. The total costs of bilateral cochlear implantation are relatively low compared to, e.g., the high self-administration costs of health insurance companies. The gained economic benefit succeeds implantation costs by far, when children are given the opportunity to hear better thus having the chance to receive better vocational training and be integrated into the industry. From an economic point of view it is certainly much more reasonable to support bilateral implantation to enable working people to continue working (thus keeping them as tax and health insurance payers), rather than sending them into retirement at the expense of social insurance [11].

The question of bilateral implantation is probably most difficult to answer in the case of elderly patients with increased surgical and anaesthesia risk factors. Patients willing to take the risk and wanting bilateral implantation to improve their situation cannot be denied. Fears of health insurance companies and their medical services of a possible "implantation wave" at the cost of health insurance seem unfounded considering that only 15% of the adults implanted in Würzburg wanted a second implantat although being aware of the possible benefits.

With budgets remaining the same, budgetary constraints caused by specified contingents could lead to the problem that only half the future CI patients could be implanted if bilateral implantation became standard procedure. This scenario caused Laszig to ask if bilateral cochlear implantation was ethically acceptable [75]. Unfortunately, nobody asks the question if the specified budgets are ethically acceptable. The question of ethical acceptability of medical treatments does not only address the problem if, based on a certain budget, a certain number of patients should be implanted unilaterally or if half that number should be implanted bilaterally, but also points to another issue: Is it acceptable to refuse, e.g., children the development of the second auditory pathway and the benefits of binaural hearing which are necessary to cope with many everyday hearing situations [129], although we have scientific proof of the benefits [2], [3], [14], [20], [21], [22], [132], [133], [134], [135]?

We must also ask the question if it is acceptable to refuse patients who contributed to the social insurance system and who could stay in their jobs with two implants (thus continuing to pay their taxes and social insurance) bilateral implantation for budgetary reasons. Costs should be seen in an overall economic and social setting. The problem is, however, that implantation costs and subsequent relief affect different cost units [9], [74].

After initially very controversial discussions on bilateral cochlear implantation, bilateral cochlear implantation has meanwhile been accepted as standard procedure in Switzerland. A clear statement for bilateral cochlear implantation was offered at the "2nd Consensus Conference on Cochlear Implants" [136], [137].

10. Auditory Brainstem Implant (ABI)

Further development of cochlear implants and their universal acceptance also influenced the development of auditory brainstem implants. In 1979 the first one-channel brainstem implant was implanted via a translabyrinthine access at the House Ear Institute. In 1992 the first patients in Germany received brainstem implants. Laszig also used the translabyrinthine access for implantation [138], [139], [140].

Meanwhile, more than 200 patients have received an auditory brainstem implant (ABI) worldwide. The translabyrinthine access was favoured in the beginning, ABIs are now also implanted via a suboccipital access [139], [141].

Commercially available brainstem implants use the speech processor technology of cochlear implants. Clinical results remain behind auditory performance with cochlear implants, despite sophisticated speech processor technology. Identification of familiar sounds like a honking car or the door bell and improved lip-reading significantly improve the patients' situation compared to complete deafness [139]. The brainstem implant enhances the average patients' ability to lip-read so far that they can follow a conversation well without having to ask for repetitions. In sentence tests these patients reach a 60 - 80% discrimination score. Some patients achieve free word understanding and are even able to use the telephone [141]. Although most patients will not reach free speech understanding, their quality of life improves so much that they use their ABI daily [138], [141].

Auditory brainstem implants can be implanted without significant additional risks during tumour removal in patients with neurofibromatosis type 2 with bilateral acoustic neurinoma [141]. Side-effects caused by erroneous sensory stimulation when activating the electrodes are rare and disappear when deactivating these electrodes.

Increasing experience with auditory brainstem implants and associated safe and in large centres almost routine application has led to discussing auditory brainstem implants for use in children with auditory nerve aplasia and patients with cochlear ossification or traumatic auditory nerve deafness [142].

Postoperative fitting of the speech processor is much more complex than fitting a cochlear implant. Use of fast speech coding strategies in analogy to cochlear implants has prevailed in auditory brainstem implants as well. Specific speech coding strategies for stimulation via the cochlear nucleus have not yet been developed. Penetrating electrodes are expected to improve the situation. Although first, preliminary results were rather disappointing and the level of the still used surface electrodes has not been reached, modern brainstem implants effectively improve communication. Patients consider these implants a great help in their daily lives and they would not want to miss them. Several of our patients even achieved open speech understanding which indicates the potential developments possible with this method. Like cochlear implant users, patients with auditory brainstem implants will benefit from technological progress.