Despite the recent progress in speech coding strategies, the sound information included in the fundamental frequency remains a serious limitation of CI. As described previously, modern electrode arrays are minimally traumatic. Because low-frequency sounds are arranged in the apical portion of the cochlea, residual hearing can be preserved and can be utilized to detect low-frequency sounds by using a short and low-volume electrode array (hybrid or electric acoustic stimulation (EAS)). In EAS, the sounds captured by the microphone are divided into low-frequency and high-frequency sounds. The low-frequency sounds are acoustically amplified and conducted to the external auditory canal using an earphone. The high-frequency sounds are processed and electrically transmitted to the spiral ganglion neuron using the intracochlear electrodes (Fig. 15.2).

EAS utilizes specially developed electrode arrays. In FLEX series electrodes (MED-EL), the basal end contains seven paired electrodes, and the apical end contains five single electrodes, instead of 12 pairs of electrode channels (24 contacts). This design makes the tip of the electrode array soft and thin. The Slim-Straight Electrode and the Hybrid L24 Electrode (Cochlear Ltd) are also designed to preserve residual hearing. The Slim-Straight Electrode is thin and smooth. The Hybrid L24 Electrode is short to prevent damage to the superior turn of the cochlea. Using these electrode arrays, more than 90 % of patients sustained measurable hearing, and the average decrease in the pure-tone average at low frequencies was less than 15 dB [14, 15].

Recently, increasing numbers of patients receive CIs in both ears. Using two CIs improves sound source localization and may benefit language development. In condition with noise from the first CI side, the second CI may improve speech understanding [16]. In normal-hearing subjects, the sound information received by the two ears differs in intensity (the interaural level difference), arrival time (the interaural time difference), and spectrum. These differences are detected by the superior olivary complex and the upper central nervous system, and they are used as major cues for sound source localization and speech extraction from background noise. Among these 3 cues, those with bilateral CI mainly utilize the interaural level difference, which is primarily derived from head shadowing effects. The interaural time difference and the spectral difference are barely perceptible because the electrode activation pattern does not have sufficient temporal and spectral resolution.

The Digisonic SP Binaural implant (Neurelec) may represent a solution for this problem. The Digisonic SP Binaural implant contains two intracochlear electrode arrays. One electrode array is inserted into the cochlea on the same side as the implant, and the other array is inserted into the other cochlea via a subcutaneous tunnel. The recipient wears two microphones in each ear, and the sound recorded in the two ears is transmitted to one speech processor. With this system, the temporal and spectral differences are digitally calculated and can be included in the speech coding strategy. The results of binaural CIs are still comparable to those of bilateral CIs [17]; however, future progress in microelectronics and speech coding strategies may improve the performance of this type of CI.

Despite the recent progress described above, the pace of improvements in the performance of CIs has slowed, perhaps because the capacity for CIs to improve is approaching a limit. One problem with modern CI is the limited number of physical electrodes that are available. As described above, speech recognition does not improve significantly with increases in the electrode number beyond 6 [4]. The main reason for this limited electrode number is thought to be the current spread. One solution for this limitation is the induction of neurites from the spiral ganglion neurons to an electrode array made of biocompatible materials. This development may improve the insulation between the electrodes and reduce the current spread. Completely new devices may solve the current spread problems. Laser stimulation is also a candidate for stimulating a small area of auditory nerves [18]. Penetrating cochlear nerve implants have also been reported to provide high spectral resolution [19]. Piezoelectric materials attached to the basilar membrane can precisely stimulate the spiral ganglion neurons [20]. Inner ear implants composed of piezoelectric materials, a totally new concept in auditory devices, are discussed in the following chapters.

The other problem is that CI surgery inevitably causes inner ear damage. Improvements in the electrode array may reduce the insertion damage to the inner ear. Due to recent advances in the field of regenerative medicine, many doctors regard CI surgery as a chance to deliver therapeutic agents directly to the inner ear [21]. The application of regenerative medicine to CIs is discussed in the following chapters.