A Brief History of the Cochlear Implant and Related Treatments

Introduction

Cochlear implants (CIs) have been described as one of the great advances in medicine in the second half of the 20th century because they replace an entire sensory organ ( ). What was once regarded as miracle—the restoration of hearing for a deaf man reported in the Gospel of Mark (7: 31–37)—is now a clinical reality. With modern CIs the majority of users can converse with ease in quiet environments and most users are able to communicate routinely with telephones or cell phones, even with previously unfamiliar persons at the other end and unpredictable and changing topics.

The first CIs provided an awareness of environmental sounds and some assistance in lipreading. However, these early devices did not provide enough information for users to discriminate among sounds unless they were grossly different from each other in amplitude envelope and presented in small closed sets, e.g., a dog barking, ocean waves, or a police siren. In addition, the aid to lipreading was modest.

Despite these humble beginnings, recipients of the CIs were grateful to be “reconnected” to the hearing world. Indeed, many of those users were profoundly grateful and their reports were at least as enthusiastic as the reports from today’s users, who for the most part are able to talk on the telephone. The “set point” has changed, and judgments of benefit by users are a matter of perspective.

The pioneers in the development of the early CIs had to overcome many obstacles in producing the first devices. Relative to today’s knowledge base, little was known about the processes underlying hearing and very little was known about how to provide electrical stimulation to a sensory system. However, the greatest obstacle may have been the vociferous criticisms leveled by a chorus of world-renowned experts in otology and auditory science (e.g., as described in ). These experts pointed to the complexity of the inner ear and wondered how the intricate functions of hearing could be approximated with crude, pervasive, and highly synchronous activation of the auditory nerve—patterns that are fundamentally different from the recorded patterns in experimental animals. The question was a good one and the initial, relatively poor, results with CIs were not unlike what the experts had predicted.

What the experts missed, by focusing on the auditory periphery, was the remarkable ability of the brain to utilize a sparse and otherwise unnatural input from the periphery. In retrospect, the job of designers of CI systems was to provide just enough information in the right form, so that the “hearing brain” could take over and do the rest. As described in Chapter 100 , that information is provided in modern CI systems with multiple and perceptually separable sites of electrical stimulation in the cochlea, and good use of those sites with processing strategies aimed at representing in the clearest possible way most or all of the information that can be perceived with CIs, including temporal, intensity, and site-of-stimulation information. With those elements, the majority of modern CI patients achieve high scores on tests of speech understanding.

Steps Toward Ever-Better Hearing With Cochlear Implants and Related Treatments

In our view, five steps characterize the development of the modern CI.

• Proof-of-concept demonstrations that a variety of auditory sensations could be elicited with electrical stimulation of the auditory nerve in deaf persons.

• Development of devices that were safe and could function reliably for many years.

• Development of devices that could provide multiple sites of stimulation in the cochlea to take advantage of the tonotopic (frequency) organization of the cochlea and the ascending auditory pathways in the brain.

• Discovery and development of processing strategies that utilized the multiple sites far better than before.

• Stimulation in addition to that provided by a unilateral CI, i.e., with a second CI on the opposite side or with acoustic stimulation, the latter for persons with useful residual hearing in either or both ears.

Steps 1–3: From Proof-of-Concept Demonstrations to Multisite Implants

The early history of the CI is illustrated in Fig. 99.1 . The figure shows key events and development efforts worldwide.

The story begins with Alessandro Volta who reported that he experienced a “boom in the head” followed by a sensation of sound like that of “boiling, thick soup” when he placed a metal rod connected to one end of a battery stack into one of his ear canals and then another metal rod connected to the other end of the stack into his opposite ear canal. He pulled the rods out of his ear canals as soon as he could react and did not repeat the experiment. The voltage of the battery stack was about 50 V, a rather heroic stimulus according to today’s knowledge. Also, the auditory sensation may have been elicited via electromechanical transduction, as in “electrophonic” hearing (e.g., ), rather than by direct electrical stimulation of the auditory nerve or any of the auditory structures in the brain. With that said, however, Volta’s observations planted the idea that electrical stimulation might be an alternative way to produce auditory sensations, even for deaf persons who are insensitive to acoustic stimulation ( ).

The first implant of a device for direct electrical stimulation of the auditory nerve in a totally deaf person was performed by the physiologist/physician André Djourno and the surgeon Charles Eyriès in an operative procedure in Paris in 1957, 167 years after Volta’s report. The patient had had bilateral cholesteatomas that had been removed in prior operations, and the removals took the cochleas and most of the auditory nerves with them. In addition, the facial nerves were sectioned, which left the patient with bilateral facial paralysis as well as total deafness. The purpose of the operation performed by Eyriès was to bridge the gap in the facial nerve on one side with a nerve graft, to restore some control of the facial muscles on that side. The operation was successful and also allowed the implantation of an induction coil, with the end of the wire from one end of the coil placed within the “shredded” stump of the auditory nerve (and probably within and certainly immediately adjacent to the cochlear nucleus in the brain) and the end of the wire from the other end of the coil placed within the temporalis muscle. Currents could be induced in the induction coil with another coil placed outside the body and overlying the implanted coil after the operation. Stimuli were presented during the operation, after the stimulating coil had been placed, and stimuli were presented after the operation when the external coil was used to excite the implanted coil. The patient reported auditory percepts in response to the stimuli during and after the operation. He could discriminate small changes in intensity but only large changes in the frequency of stimulation. Above 1000 Hz he could not discriminate any changes in frequency up to very high frequencies. In his daily life following the operation he could sense the presence of environmental sounds but could not understand speech or discriminate among speakers or many sounds. He could identify speech sounds within small closed sets, e.g., with three words in each set, most likely on the basis of rhythmic cues. His device failed after several months of use (a broken solder joint was the problem) and had to be repaired in another operative procedure. The device failed again in the same way soon thereafter and was summarily explanted by Eyriès, and at that point he withdrew from the project. Djourno carried on with the surgeon Roger Maspétiol, who implanted a similar device in a patient who became deaf due to streptomycin poisoning of the cochlea. In this case the “active” electrode was placed on the promontory of the cochlea rather than in the auditory nerve or cochlear nucleus. The patient was not enthusiastic about her implant and was lost to follow-up.

The early work in Paris is described in far greater detail in and . In his recounting of the work, Eisen posits that the cochlear nucleus rather than the auditory nerve may have been stimulated in the first patient; thus the implant may have been the first auditory brainstem implant rather than the first CI. With that said, however, the work demonstrated for the first time that a variety of auditory percepts could be elicited with direct electrical stimulation of auditory neurons in either the auditory nerve or the cochlear nucleus.

Results from the first patient were reported (also in 1957) in the French journals Comptes Rendus des Seances de la Societé de Biologie and Presse Medicale . The results were not widely known outside of France until by serendipity a patient showed her ear doctor in Los Angeles a short newspaper article very briefly describing the experiments by Djourno and Eyriès. That ear doctor was William House, who later invented many of the procedures used in modern otology and is acknowledged as the “father of neuro-otology.” House was inspired by the article and set out to develop an auditory prosthesis for the deaf based on electrical stimulation and that could be used safely and for many years in the daily lives of the patients. He and his coworkers, most notably an electrical engineer, Jack Urban, are largely responsible for achieving step 2 in the list presented in the preceding section in this chapter. The devices that were widely applied provided a single site of stimulation along the length of a short biocompatible wire inserted into the basal part of the scala tympani (ST) in the cochlea. These devices ultimately were manufactured and distributed by the 3M Company in St. Paul, MN, USA, and were the first CIs approved for clinical use by the United States Food and Drug Administration (FDA), in 1984.

The 3M devices and their single-site predecessors enabled an awareness of environmental sounds and provided a useful adjunct to lipreading. However, speech understanding with hearing alone still was not generally possible with the devices, although some understanding by a tiny fraction of the patients was reported.

House performed his first two implant operations in 1961, very soon after he was made aware of the French results. One of the patients was implanted with a single-site device and the other with a multisite device; the latter device included five wires of different lengths inserted into the ST. The devices were removed a short time later due to infection in one of the patients and suspected infection in the other. Unfortunately, one member of the collaborating teams shared details of the research with the local press, and the resulting newspaper articles described a possible “cure” for the deaf and otherwise distorted the findings. Following the articles, the implant team was deluged with requests from deaf persons and their families for help, and House was enraged by the articles. He thus decided for these and other reasons ( ) to discontinue his work on the CI, at least for a while. Later, in the late 1960s, he resumed the work with Jack Urban and others following the development of the heart pacemaker and advances in electronics such as the transistor, which House regarded as “game changers” for the development of a useful auditory prosthesis for the deaf.

In 1969 three deaf patients were implanted with multisite devices of a somewhat different design that included a percutaneous connector for direct electrical access from outside the body to the implanted electrodes ( ). One of the patients was tested extensively and for many years. He heard different pitches with stimulation of the electrodes one at a time. Various ways of representing speech information with the five intracochlear electrodes were also tested, including presenting the same information to all of the electrodes simultaneously. None of the tested ways enabled speech recognition, but useful auditory percepts were produced and stimulation of all of the electrodes together seemed to be at least as effective as the tested alternatives. House and Urban thus decided to develop a single-site device that could be used for years by each recipient and that would be based on findings from the initial studies. In addition, House and Urban thought that a single-site device would be easier to manufacture, more reliable, and less traumatic to the cochlea than the more complex multisite devices.

A wearable take-home processor for the single-site device was produced in 1972. The device also included a transcutaneous transmission link, with an external transmitting coil and an implanted receiving coil, instead of a percutaneous connector. House and Urban reasoned that the transcutaneous link would greatly minimize the chance of infection.

The device was the prototype for the system that was ultimately manufactured and made available by the 3M Company. More than 1000 individuals received the “3M/House” CI before the discontinuation of the product by 3M in 1987, when multisite stimulation systems had largely supplanted single-site systems.

In 1964, soon after the initial implants by House and coworkers and before the resumption of work to develop the CI by House and Urban, F. Blair Simmons implanted a deaf and legally blind volunteer at the Stanford University Medical Center in Palo Alto, CA, USA. In the operation a bundle of six wires of slightly different lengths was inserted into the trunk of the auditory nerve rather than into the ST. The cut ends of the wires each stimulated at least somewhat different populations of neurons within the nerve, and stimulation of each electrode in isolation elicited a distinct pitch percept that was different from the pitches produced with stimulation of any of the other electrodes. In addition, different rates of pulsatile stimulation for each individual electrode elicited a range of distinct pitches up to a rate of about 300/s, which later was called the “pitch saturation limit” for electrical stimulation of the auditory nerve, either by electrodes in the ST or in the nerve trunk. These observations affirmed both the “place” and “volley” theories of pitch perception, which were controversial at the time, and also provided the foundation for processing strategies that were developed subsequently to represent frequency information with both temporal and site-of-stimulation cues.

Like House’s patients, Simmons’s patient could not understand speech with hearing alone. Speech sounded a bit like speech to the patient but “maybe was in a foreign language” that he could not understand.

Teams other than those led by House also developed single-site devices, including (but not limited to) a team in San Francisco led by Robin Michelson and a team in Vienna, Austria, led by Kurt Burian, Erwin Hochmair, and Ingeborg Hochmair-Desoyer. Among these single-site devices, the Vienna device provided surprisingly good performance for some patients, who could understand useful amounts of speech with hearing alone. To our knowledge, that was the first independently verified instance of “open set” speech recognition by otherwise deaf persons using a CI only and without any visual cues ( ). It was a landmark achievement.

In addition to developing single-site CIs, the Vienna team was the first to produce a multisite device that included a sophisticated transcutaneous transmission link using microelectronics technology that had just become available. The Hochmairs are electrical engineers and were pioneers in applications of the technology for CIs. The first multisite device produced by the Vienna team was implanted in a patient by Burian in December 1977. Another multisite device also using microelectronics technology was developed by a team in Melbourne, Australia, led by Graeme Clark. The device was implanted in a patient by Clark in Melbourne in August 1978. These devices became the “blueprints” for many future devices, all of which used the same basic designs for the implanted components and the transmission link.

As indicated in Fig. 99.1 , the Vienna and Melbourne teams were hardly alone in the quest to develop a highly effective multisite device. The other teams included (but again were not limited to) those led by Michelson and Michael Merzenich at the University of California at San Francisco (UCSF); by Donald Eddington and his team at the University of Utah in Salt Lake City, UT, USA; by Claude-Henri Chouard and his teams in Paris (a renewed effort there following and inspired by the earlier work in Paris by Djourno and Eyriès); and by Stefaan Peeters and Erwin Offeciers at the University of Antwerp in Belgium. In addition, a renewed effort at Stanford was initiated in the mid-1970s, with both Blair Simmons and Robert White as its codirectors.

Many of these efforts were contemporaneous, and some produced electrode arrays and transcutaneous transmission links that in retrospect are the progenitors of the arrays and links used in today’s devices. The processing strategies have changed since the early 1990s, as described later in this chapter, but the basic designs of the arrays and links were established before then.

Additional information about the early period, during which steps 1–3 were completed or at least largely achieved, is presented in , and . Some of these publications review many efforts and how they related to each other, and the remaining publications highlight a particular effort, e.g., the effort at the University of Utah led by Eddington is described in detail in the review by Dorman and Parkin.

Among the pioneers in the early development of the CI, House and Clark deserve special commendation for their contributions. House and his coworkers developed the first practical CI device, and he was criticized roundly by the experts of the period. (House most certainly was not the only recipient of the criticisms, but he may well have received more criticism than anyone else attempting to develop a useful hearing prosthesis for the deaf.) However, he persevered and provided a key part of the foundation for what followed. Without his efforts, and without his dedication and resolute determination, the development of the CI would have been delayed by many years, if developed at all. We as a field, and the hundreds of thousands of persons who have received a CI since the early days, owe the greatest possible debt of gratitude to Bill House.

And Graeme Clark and his coworkers developed devices that were manufactured first by Nucleus Ltd. and later by Cochlear Ltd. in Australia. These devices are by far the most widely used worldwide, and the companies and Clark were unparalleled in their successful efforts to move CIs into the mainstream of clinical practice. Indeed, the first multisite implant approved by the FDA was the Nucleus device in 1985, soon after the approval of the 3M/House single-site device in 1984. The Australian devices have been the leaders in market share since the mid-1980s, with more than 60% of the share throughout the period. In addition, Clark and his teams conducted more basic science studies and published far more papers than any of the other groups working to develop the CI.