Preserving Residual Acoustic Hearing with Combined Acoustic and Electric Hearing
In recent years, a unique approach in the treatment of severe high-frequency sensorineural hearing loss has been explored. This new approach modifies existing cochlear implantation techniques by preserving any remaining hearing in the implanted ear and has already been implemented in more than 100 people worldwide. Many people who receive cochlear implants (CIs) still have some residual hearing in the opposite, non-implanted ear and sometimes wear a hearing aid in this ear to provide additional beneficial sound stimulation. This new technique involves preserving low-frequency hearing in the implanted ear, with the CI filling in high-frequency sounds. The aim is to produce improved speech understanding and other hearing sensations via combined acoustic (A) and electric (E) hearing. With A+E hearing, one uses both acoustic hearing (sound waves entering the ear in typical fashion, amplifi ed by a hearing aid) plus electrical hearing (sound waves being received by a sound processor, which transmits signals to the electrode implanted in the inner ear, or cochlea). The addition of acoustic hearing, either via the opposite, or the implanted ear, can often improve a person's total hearing abilities.
Research in animals with hearing loss has shown that sensorineural hearing loss exceeding approximately 60 decibels (dB) is due to substantial inner ear hair cell loss. However, people with this level of hearing loss can still perceive speech cues, generally signaled by intensity differences in the lower frequencies of the hearing spectrum. If low-frequency speech is amplified, even a small number of functioning inner ear hair cells should be capable of transmitting these low-frequency speech sounds to the brain. However, the speech cues located in higher frequencies tend to require a fuller population of inner ear hair cells for accurate transmission. Thus, amplification via a hearing aid is often not very effective for people with severe to profound high-frequency hearing loss. These people, however, can benefit from a CI.
The underlying purpose and function of a CI is to stimulate the auditory nerve directly, via electrical impulses, replacing the inner ear hair cells that have been lost or damaged. The electrical impulses are transmitted into the nearby auditory nerve via an electrode array of varying lengths, which is inserted into the cochlea. In the case of profound or severe hearing loss across the entire frequency range, usually few functioning inner ear hair cells remain in the cochlea and very little, if any, usable speech information can be heard with the ear thus, hearing aids are not an effective treatment. In these cases, a CI with a standard-length electrode (approximately 20-25 mm) aims to replace the function of all, or nearly all, the inner ear hair cells across the entire hearing frequency range. People with this level of hearing loss, therefore, are generally not appropriate candidates for the new technique of preserving hearing in the implanted ear, although future develop-ments may include preserving residual hearing in some of these people.
The current generation of standard-electrode CIs is capable of providing very high levels of speech understanding in quiet backgrounds. Although there is a wide range of results with cochlear implantation, this is usually not because of a fault in the device. Individual characteristics play a large role, such as the age at which the person was implanted, with younger patients generally faring better. One might reasonably ask then, why not just implant every person who does not benefit from hearing aids with a standard length CI? The answer is that in many cases, the remaining low frequency acoustic hearing can actually improve overall perception of sound when combined with high-frequency hearing via electrical stimulation.
Research has shown that the information from as few as three to four electrodes in an implant is adequate for high levels of speech understanding in quiet backgrounds. However, trying to hear one specific voice in a noisy background requires significantly more electrodes, thereby increasing the ability to distinguish pitches (known as frequency resolution). When the background noise is speech or speech-like in nature, the frequency resolution requirements can be even more demanding, as better frequency resolution is needed to distinguish between various speakers. Even the top-performing implant users cannot take advantage of more than six to eight electrodes of electrical stimulation. This seems to be a result of how the electrodes interact with each other in the cochlear fluids, limiting the independence of each electrode. Therefore, traditional CI users have tremendous difficulty in recognizing speech when there are competing signals, such as hearing a person across the table from them at a crowded restaurant.
One study directly compared the behaviorally measured frequency resolution of three types of listeners: those with normal hearing, those with hearing loss and those with traditional CIs. Normal-hearing listeners had the best frequency resolution, followed by listeners with sensorineural hearing loss; the poorest resolution was observed in the traditional CI users. People with normal hearing can take advantage of various cues, such as pitch, timing and localization cues, that allow them to separate multiple sound sources and focus on the target speech. CI listeners, on the other hand, are presumably unable to perceive some of these cues and, instead, competing speech is confused with the speech they want to hear. This appears to be related to some limitations on frequency resolution and localization abilities in people with CIs; using existing residual hearing offers an opportunity to improve these cues for CI users.
Frequency resolution affects pitch perception, and limited frequency resolution means decreased enjoyment of music. Preserving acoustic hearing in an implanted ear offers a way to enhance music appreciation. One study demonstrated that, while normal hearing listeners had no difficulty in discriminating piano notes one semitone apart (i.e., adjacent piano keys), the CI listener's ability to discriminate notes was generally much poorer: The average listener needed a half-octave difference in pitch to be able to discriminate one pitch from another. Some listeners required as much as two octaves' difference between notes for discrimination. This poor pitch perception makes recognizing melody very difficult for CI users. The typically poor pitch perception of CI users is most likely a combination of limitations of current electrode technology and poor survival of remaining nerve fibers in the damaged cochlea.
In summary, poor frequency resolution appears to be a major remaining hurdle in improving the listening abilities of people with CIs. It is also interesting that the frequency resolution of still existing
acoustic hearing, even when there is a severe hearing loss, is often better than that provided by a CI. For these reasons, preserving residual acoustic hearing in either or both ears of people receiving a CI has some potential advantages. However, in many patients, the remaining acoustic hearing is only usable for speech recognition in the lower frequency regions.
Preservation of Hearing Following CI Surgery
But how can acoustic hearing be preserved in a person receiving a CI? Implanting only one ear is an obvious option, if the other ear has usable hearing. Recently, however, preserving residual acoustic hearing in the implanted ear has received attention.Preservation of hearing in the implanted ear has been reported in several animal studies from the 1990s, which found that placement of a short electrode did not cause tissue damage in areas of the ear that are not adjacent to the electrode. These studies found that many hair cells not only survive but can function essentially normally.
Further studies reported that some patients examined after receiving a CI had at least some response to acoustically presented tones. The residual acoustic hearing in one human patient who had received a standard long-electrode device enabled understanding of speech at levels higher than would occur by mere chance. In the late 1990s, a group at the University of Iowa began implanting a newly-designed CI with a modified electrode that was much smaller in diameter and only 6 mm in length to intentionally preserve some low-frequency hearing in selected patients. This electrode was later changed to 10mm in length and even better results were obtained. This device became known as the Iowa/Nucleus Hybrid device. Researchers in Europe proceeded along similar paths at the same time. Based on these early reports of preserving low-frequency hearing following cochlear implantation, a number of other centers have specifically attempted to preserve residual hearing in implanted ears of their patients.
As more people have become eligible for cochlear implantation, we are seeing more individuals with various amounts of still usable hearing in one ear. These people have the chance to use acoustic hearing in one ear combined with electric hearing in the other. In a number of studies, patients were implanted with a standard-length electrode in one ear, and used a hearing aid in the other ear. When compared to listening with the CI alone, the acoustic/electric combination achieved better speech recognition in noisy backgrounds, presumably because the acoustic hearing assisted in the separation of target voices from background noise, due to its more precise pitch perception. The acoustic ear was found to also contribute to speech recognition in quiet. Similar advantages exist for using preserved acoustic hearing in the implanted ear.
If, as is often the case, the ear opposite the implant is the "better" ear, then improved speech recognition may primarily be a reflection of the better ear's status. Yet even when this factor is taken into account, the evidence is clear that combining acoustic and electric hearing either within one ear or across two ears can provide a significant advantage for many people.
The Impact of A+E Hearing in Quiet
When evaluating the success of the A+E approach, several factors need to be considered. The fi rst is the many uncontrolled variables when making comparisons across various devices, clinics or patient populations. Some of these variables include: the degree of residual hearing; very low acoustic hearing-alone scores, perhaps indicating poor nerve survival; and very high acoustic hearing alonescores, possibly limiting how much improvement the added stimulation can provide. For example, recent data from the Iowa/Nucleus Hybrid clinical trial suggests that those with more than 35 years of severe to profound hearing loss above 2,000 hertz often do poorly with the added electric stimulation. This may be because of insufficient surviving nerve cells in the base of the cochlea needed to take advantage of a short 10 mm electrode.
Second, merely showing an increase in speech recognition performance does not, by itself, validate the A+E approach. The combined score must also be better than the electric-alone score, in order to demonstrate that preserving residual hearing was beneficial. A third caution is that, although testing the implanted ear by itself is a good way of evaluating the success of the same-ear A+E approach, in real life many people listen and receive speech cues through both ears.
A 2006 study reported the preliminary results of 47 subjects enrolled in the Food and Drug Administration (FDA) multicenter clinical trial of the Iowa/Nucleus Hybrid 10 mm implant. Of the 19 subjects with nine months of experience with their device, 16 demonstrated significant listening improvement with combined A+E speech processing. The subjects who had 12 months of experience averaged 72 percent correct on a word test in the combined mode, compared to 32 percent correct prior to cochlear implantation when they were tested wearing two hearing aids. Only four of the 47 subjects did not derive benefit from the combined A+E speech processing.
The Impact of A+E Hearing in Noise
In a 2004 study, researchers demonstrated that, compared to electric stimulation alone, A+E hearing had the potential to provide a significant advantage for understanding speech in background noise, particularly when the competing sounds were other talkers, using simulated Hybrid processing of speech and normal-hearing subjects. In that same study, three Hybrid patients were also tested and showed an advantage for understanding speech in noise when compared to top-performing long-electrode (electric alone) patients.A 2008 study showed that, when compared to traditional CIs and electric hearing alone, the Hybrid group had a significant advantage of four to five decibels, on average. However, it should be noted that some individuals were able to perform the task when sound was 15 to 20 decibels lower than the best performance of traditional CI subjects, thus showing the impressive potential benefits of preserving residual acoustic hearing.
The Impact of A+E Hearing with Music
While CI recipients generally are quite good at perceiving rhythmic cues in music, their recognition of melodies is usually much poorer than normal, especially when rhythmic or lyrical cues are not available. The residual low-frequency acoustic hearing of an A+E approach can provide pitch information needed to enjoy music to a greater degree. In 2006, one researcher tested typical-hearing, traditional long-electrode, and Hybrid short-electrode patients on melody and instrument recognition. It was discovered that Hybrid patients were nearly as accurate as normal-hearing patients for melody recognition, whereas the long-electrode patients performed very poorly. This is a result of the poor frequency resolution provided by electrical stimulation, as described earlier. Hybrid patients did show a deficit compared to those with typical hearing when it came to recognizing instruments, but this was primarily for instruments in the higher-frequency ranges, where the information was received via the CI rather than by the acoustic hearing.How Much Hearing is Worth Preserving?
Also of practical interest is a consideration of how much residual low-frequency hearing is required for the short-electrode A+E strategy. Research fi ndings suggest that, on average, the advantage of preserving residual hearing exists unless hearing loss approaches profound levels. A 2005 study of traditional CI users who used a hearing aid in the non-implanted ear showed that residual hearing in the non-implanted ear could assist CI users in understanding speech in background noise, even when that ear was not capable of speech recognition by itself. The benefi t of acoustic hearing remains as long as the hearing loss is not profound (using a hearing aid, if appropriate, along with the CI). However, in cases of profound or near-total hearing loss there may be little benefi t to aided hearing; these people are most likely candidates for a traditional, long-electrode CI.Summary and Future Questions
The preservation of residual hearing has been shown to be a practical and effective solution for severe, high-frequency hearing loss. It can overcome some of the inherent disadvantages of traditional, electric-only, long-electrode cochlear implantation. These advantages of the A+E approach are primarily a result of the better frequency resolution provided by the residual acoustic hearing as compared to electric stimulation. Thus the advantages of the A+E approach are most evident in situations where frequency resolution is important, such as music perception and recognizing speech amid background noise.The clinical trials of the A+E approach are still in their early years and it will be some years before FDA approval is secured; therefore, the issue of long-term success rates deserves continued attention. How stable is residual acoustic hearing over longer periods of time? A 2006 study retrospectively looked at the changes over time in the hearing of non-implanted ears that had audiograms that at one time fi t the criteria for the short electrode. These were people who were not subsequently implanted, since, at the time, they did not meet the criteria for standard-length cochlear implantation. They found that in adults, low-frequency hearing remained relatively stable and consistent. However, for children the rate of hearing loss was generally greater and much more variable from child to child. Research looking at the long-term stability of low-frequency hearing in people who are implanted with electrodes designed to preserve hearing will become available in future years as the current patients have their devices for longer periods.
The optimal length of the electrode for A+E hearing is also a matter of debate. Although it would seem to be a logical assumption that longer electrodes present more risk to residual hearing, this has not been conclusively demonstrated. Hearing preservation must be balanced against the possibility that a shorter electrode may or may not provide as much information to the auditory system as a longer electrode, particularly in the few unfortunate cases where residual acoustic hearing is not preserved. It is possible, however, to successfully re-implant with a standard electrode those few who do not benefit from a short electrode. In addition, future developments in electrode design or surgery may serve to reduce the risk of damage for any electrode insertion.
In a somewhat related issue, the ability of the auditory system to adapt to cochlear implantation may also infl uence the choice of electrode length. Further research in this area is certainly needed. A+E hearing shows promise in preserving and capitalizing on residual low-frequency hearing, resulting in better hearing for implant recipients. This may take the form of merely encouraging people with a traditional long-electrode CI in one ear to use a hearing aid in the other ear. It may now also be accomplished by preserving residual low-frequency hearing in the implanted ear of people with severe high-frequency hearing loss. This would provide a rehabilitative solution for a group of people for whom there were previously no attractive treatment options.
Christopher W. Turner, Ph.D., is a professor of speech pathology
and audiology, otolaryngology at the University of Iowa in Iowa
City, Iowa.
Lina A. J. Reiss, Ph.D., is an assistant professor, School of
Medicine-Otolaryngology & Head & Neck Surgery Department, at
Oregon Health & Science University in Portland, Ore.
Bruce J. Gantz, M.D., FACS, is a professor and head of the
Department of Otolaryngology & Head and Neck Surgery at the
University of Iowa Hospitals and Clinics in Iowa City, Iowa.
