Beyond The Hearing Aid
DURING THE 1950s, IN THE GLORY DAYS OF THE NEW YORK Yankees, Gil McDougald was the 1951 Rookie of the Year and a five-time All-Star infielder. One day in 1955, during batting practice, he reached beyond a protective fence to pick up a ball. Right at that moment, his teammate Bob Cerv sent a line drive in his direction, striking him just above the left ear. McDougald was not severely injured, missing only a few games before returning to the lineup. But the accident had fractured his skull and damaged his inner ear.
He soon lost most of the hearing in his left ear, and the hearing in his right ear diminished as well. He stayed with the Yankees for several more seasons, helped by hearing aids, and continued to rely on the devices through the 1960s and 1970s, after his retirement from baseball. Still, his hearing continued to deteriorate. He couldn’t participate fully in family discussions. He stopped using the telephone; he ceased joining old teammates on the banquet circuit. Increasingly, he became a recluse.
Then, in 1994, he gained new hope from Stephen Epstein, a doctor and an avid Yankee fan. Epstein told him about the cochlear implant, an electronic device that promised to bypass the damaged tissues of McDougald’s inner ears and thereby restore his hearing. A surgeon and former Brooklyn Dodgers fan, Noel Cohen, installed these implants in both of McDougald’s ears a few months later.
They worked. Right at the start, he found that he could hear and repeat words without error. He renewed his old Yankee friendships, attending baseball dinners and telling stories about the manager Casey Stengel. He also launched a new career as an advocate for the deaf.
Cochlear implants are a fundamental advance over conventional hearing aids. A hearing aid amounts to a miniature microphone and amplifier that directs the amplified sound through a speaker to the eardrum. Such a device works for hearing-impaired people whose inner ears are largely intact but have lost sensitivity.
Some people, however, are so profoundly deaf that standard hearing aids offer little or no relief. Such patients can often be helped with cochlear implants. These bypass the eardrum as well as other structures of the inner ear and provide direct electrical stimulation of the nerves that lead from the ear to the brain. The brain interprets signals from these nerves as speech or other sounds. The approach is somewhat like restoring sight to the blind by connecting a miniaturized television camera directly to the optic nerve.
The association of electric currents with hearing goes back two centuries to Italy. Alessandro Volta, who invented the electric battery, placed electrodes in both his ears and reported that a flow of electricity produced a sound that resembled “the boiling of thick soup.” A similar experiment by a French researcher named Guillaume Duchenne, in 1855, gave a sound described as “the beating of a fly’s wings between a pane of glass and a curtain.”
The next important step came in 1930. Two investigators studied the electrical response in the auditory nerve of a cat and showed that these signals were similar in frequency and amplitude to the sounds the cat was hearing. This encouraged researchers to investigate the details of the mechanism of hearing and discover how sounds could give rise to such a neural response.
The key proved to lie in the cochlea, a snail-shaped organ deep within the inner ear. In humans, it is the size of a small cornflake. The cochlea holds a delicate membrane that is studded with some 16,000 tiny hair cells, each of them a microscopic valve. As a hair cell vibrates in response to sound, it opens a passage within the membrane that permits ions in an electrically conducting fluid to flow through. These ions stimulate the auditory nerve, producing the neural signals that pass into the brain.
Another investigator, Georg von Békésy of Harvard University, showed during the 19505 that the cochlea acts as a frequency analyzer. Most sounds consist of a mix of frequencies, and the cochlea separates them. Low frequencies cause the internal membrane to vibrate most strongly at the center of the cochlear coil; high ones, at the coil’s periphery. Nerve fibers, stimulated at various points along the coil, produce corresponding signals that the brain interprets as pitch.
Profound deafness can result when the hair cells of a cochlea are destroyed. Infections such as meningitis can do this; so can accidents, as with Gil McDougald. However, in such instances, the cochlea itself often remains intact, along with the auditory nerves. As early as 1957, experiments showed that direct electrical stimulation of these nerves could reproduce the sensation of hearing.
In one such test, a patient was placed under local anesthesia to have a tumor removed, thus retaining consciousness. The operation exposed his auditory nerve. When the surgeons stimulated the nerve with electrodes, he proved able to recognize differences in pitch between sounds. He also succeeded in distinguishing the words mama, papa , and hello .
The concept of a cochlear implant subsequently took form as a device that would provide a functional replacement for the entire ear. Such a device would work by inserting one or more electrodes into the cochlea, where they would substitute for the damaged hair cells and stimulate the auditory nerve directly with a flow of electric current. A microphone would convert the original sound waves into electronic form, and a processor would adapt this signal for direct nervous stimulation. The first scientist to pursue this was Professor Graeme Clark of Australia’s University of Melbourne. His father was deaf and he was determined to improve the quality of life for those who could not hear.
Clark began his work in 1967. In Los Angeles, a private venture led by William F. House had been pursuing its own program since the early 1960s and had devices ready for implantation in the early 1970s. The electronic circuitry posed few problems, for though the era of the microchip was not yet at hand, it proved feasible to package the circuits in conveniently compact containers. Still, there was a problem: biocompatibility.
Human organs do not easily accept implants made from artificial materials, and the cochlea proved especially difficult. It does have a small hole at its base, called the round window, which provides a natural entryway for a slender electrode, but the cochlea’s electrically conductive fluid resembles seawater, and this created a number of problems.
In such a fluid, two dissimilar metals can give rise to electrochemical reactions when in contact. These corrode the metals and can also release toxic heavy-metal ions into the immediate vicinity of auditory neurons, which are among the body’s most sensitive cells. Moreover, these reactions can produce electric currents, as from a battery, that are strong enough to cause some of the water in the cochlear fluid to decompose into damaging bubbles of hydrogen and oxygen. Finally, ions derived from metal atoms in an implant can produce toxic levels of acidity or alkalinity within the fluid.
The best metals, the most biocompatible, included platinum and iridium, but even these demanded considerable care. At EIC Laboratories in Norwood, Massachusetts, the chemist Barry Brummer directed a program of research that established conditions for the safe use of these materials. These guidelines involved avoiding electrochemical reactions with high voltages, replacing metals with other materials whenever possible, and using no metals with a potential for toxicity.
The first cochlear implants were single-channel models employing only one electrode. If we think of the 16,000 hair cells in the cochlea as so many piano keys, lined up by their pitch, then single-channel implants amounted to striking a lot of them at once. The devices gave little sense of pitch, which is essential in understanding speech. Indeed, the perceived sound from such implants often was hardly more than a buzz. Even though the technology had advanced far beyond Volta’s simple electrodes of the early nineteenth century, the quality of the perceived sound showed little change.
Still, Volta had enjoyed normal hearing, whereas the implant recipients were deaf. Therefore, even a modest restoration of hearing proved useful. Although these people could not decipher spoken words, they could recognize the varying intonations of speech and tell if someone was trying to talk to them. By hearing the volume of their own voices, they could learn to speak in a normal tone. They could hear common sounds, such as a door closing or a telephone ringing, and respond to sounds of danger such as an auto horn or a barking dog.
Despite their limitations, the initial successes sparked a strong expansion of the work. Within the National Institutes of Health, one of the member institutes launched the Neural Prosthesis Program, establishing cochlear implants as a topic for federally funded research. Sensing commercial possibilities, the 3M Company became involved. It supported further work by William House at his House Ear Institute in Los Angeles. 3M initiated a similar partnership with a group at the Technical University of Vienna, led by the husband-and-wife team of Erwin Hochmair and Inge Hochmair-Desoyer.
Talented newcomers entered the field. At the University of California at San Francisco, the medical resident Robert Schindler helped install some of the earliest cochlear implants. He was pleased with the results and later declared, “I caught the bug and said, ‘This is going to be what I want to do for a lifetime.’ ” His university became an important center for research in the field.
As the work proceeded, House and the Vienna group pursued single-channel devices of improved design. Such implants appeared feasible as near-term developments. For others, however, the goal was more far-reaching: to develop multichannel cochlear implants, which could give their recipients a good chance at understanding speech. This called for inserting several electrodes into the cochlea to stimulate multiple groups of nerves, each corresponding to a range of frequencies. Studies showed that at least six such channels were necessary, and this requirement brought new problems.
To install a single-electrode implant within a cochlea, it sufficed merely to push it past the round window for a very short distance. By contrast, multichannel devices called for a strip of electrodes that could reach deep within the cochlea while conforming to the tight curvature of its coil. Surgeons expected to insert these electrodes by probing through the ear canal, so the array of electrodes had to remain straight while being nudged through these passages, then somehow be made to curve.
But as the researcher Gerald E. Loeb wrote, the line of electrodes could not “simply be pushed into the cochlear spiral as if it were a plumber’s snake passing through a curved drain-pipe.” The internal shape of the spiral tended to force the strip through the delicate cochlear membrane, which protects the adjacent nerves. With the membrane punctured or torn, cochlear fluid would leak through, and the fluid is toxic to the sensitive nerve fibers.
One solution called for a very slender and highly flexible strip that could indeed be pushed into position. However, this tended to force the electrodes against the outer cochlear wall, far from the nerves. A better approach used a thicker line of electrodes that could be coiled in advance, and straightened, as if made of rubber, for insertion through a tube. It then would curl back into its spiral shape during insertion, hugging the inner wall close to the nerves.
The placement and design of the electrodes demanded painstaking care. Because cochlear fluid is electrically conducting, current from any electrode tended to spread throughout that organ, stimulating even distant nerves. This had not been a problem in building single-channel implants, in which the whole point was to reach as many nerves as possible. But for a multichannel device, it was important for each electrode to stimulate only the nerves in a highly localized area.
For recognizing speech, the most important frequencies lie between 500 and 3,500 hertz. Along the cochlear spiral, the corresponding receptors cover a distance of no more than 14 millimeters, roughly half an inch. To install multiple channels within this short distance, it proved useful to design each channel as a closely spaced pair of electrodes, one of them being a source of electric current and the other being a sink. Each such pair then gave only the desired localized stimulation of the nerves, even when immersed in cochlear fluid.
Clark, in Australia, was the first to devise a useful multichannel implant. His first patient, in 1978, was a 48-year-old man who had completely lost his hearing following a recent head injury. Medical examination disclosed that this deafness had resulted from concussion of the cochlea, with the auditory nerves remaining intact. Clark’s colleagues installed a nine-channel device. The surgery went well; afterward, the patient could understand his wife as she spoke to him. Here was the first clear indication that multichannel implants could advance beyond the simple buzzes of single-channel devices and restore a useful ability to understand speech.
This success won attention from an Australian corporate group called Nucleus, which specialized in advanced medical equipment. In 1981, with support from the Australian government, Nucleus and the University of Melbourne launched a joint effort aimed at developing a commercial cochlear implant. This led to the formation of a new firm, Cochlear, Ltd., with an American branch taking shape as Cochlear Corporation in Englewood, Colorado. Their product was a 22-electrode implant, with the electrodes being linked in source-sink pairs to give 11 channels.
Through the mid-1980s, work in the United States remained experimental, for cochlear implants required approval from the Food and Drug Administration before they could be made generally available. The first such implant, a single-channel model from House and 3M, won FDA approval in 1985. A second single-channel device, from the Vienna group, soon followed. However, multichannel implants quickly demonstrated a clear superiority. By 1990 single-channel versions were no longer being sold in the United States.
When inserting a strip of electrodes into a cochlea, a surgeon had to exercise considerable care, since each channel corresponded to a specific range of frequencies and hence to a particular location within the cochlear spiral. If the electrodes were in the wrong place, there would be a mismatch, with patients reporting high-pitched speech that sounded like Donald Duck.
Even with the use of source-sink pairing, there remained a need for improved methods of channel separation, to prevent individual electrodes from stimulating the wrong neurons even when they were properly positioned. At the Research Triangle Institute in North Carolina, the research director Blake Wilson decided that he could address this issue by taking a lesson from Hollywood.
When we watch a movie, we do not see the truly continuous motion of the real world. Instead, we see a rapid succession of fixed images projected onto the screen, which gives the illusion of continuous movement. In 1989 Wilson realized he could produce a similar illusion of continuous sound, with good channel separation, by firing cochlear-implant electrodes one at a time in rapid succession. In this way, he could prevent the separate channels from interfering with one another. This technique went on to become a standard feature in commercial implants.
Electronic processing of the speech signal has also yielded benefits. The normal ear can deal quite well with a range of loudness that extends from a lover’s whisper to a jackhammer. Implant systems cover a far more restricted range of volume, which means that right at the outset, it was necessary to provide automatic volume control. The simplest form of speech processing used electronic filters to divide the heard sound into frequency bands, then adjusted the volume of the resulting signals to fit the lesser range of the system and fed them to their corresponding channels.
In addition, speech carries important features that have been extracted electronically, to aid in recognizing words. Spoken words show “formants,” or frequencies that carry most of the volume. At Cochlear Corporation, designers have found it useful to identify these formants as they change from one millisecond to the next, and to use them in defining the signals presented through the individual channels. These channels are frequency bands: 500 to 750 hertz, perhaps, then 750 to 1,000, then 1,000 to 1,500, and so on. A related strategy divides the heard speech into as many as 2.0 such frequency bands, with the strongest or loudest ones driving selected channels. Graeme Clark has found that when a particular frequency is important, it can help to present it not merely to one channel but to nearby channels as well.
It has even proved possible to restore hearing to patients who lack a working set of auditory nerves. Several thousand Americans suffer from neurofibromatosis, a life-threatening genetic condition in which tumors can grow on the spinal cord. Sometimes they also grow on the auditory nerves. The only treatment involves surgery, and removal of the tumors severs these nerves. To restore hearing, it then becomes necessary to bypass not only the ear but the cochlea and auditory nerve as well. Instead of installing a cochlear implant, a patient needs nothing less than an implant within the brain.
The House Ear Institute has developed such an implant, which Cochlear Corporation is manufacturing. The pertinent region of the brain is its stem, which is about as soft as Jell-O. Conventional electrodes would sink into it, but the House researchers addressed this problem by designing electrodes with a backing of Dacron mesh. Fibrous tissues from the brain stem grew into the mesh, forming a capsule that held the electrodes in place. Brain-implanted patients tend not to hear as well as people with intact nerves and multichannel cochlear implants. Still, the House research director Robert Shannon says that a brain implant can restore enough hearing to diminish feelings of being cut off from the outside world.
How good are the cochlear devices? Among adult recipients, as many as one-fifth have obtained little or no benefit. Three-fifths use the implants to assist them in lip reading, and the remaining one-fifth have regained a large measure of their ability to hear effectively. Gil McDougald has reason to count himself fortunate, for he is in the last of these groups.
The less than total success of these implants, despite their technical virtuosity, raises particularly strong issues when using them with children. In 1990 the FDA approved the 22-electrode Cochlear Corporation device for use with infants as young as two years. The results have been dramatic.
One girl, deaf from birth, was given a cochlear implant at three. Within three weeks, she began responding to sound. A month later, she started using her own voice. A year later, she tested on a par with two-and-a-half-year-olds in her use of language. In another instance, the mother of an implanted four-year-old boy made a good deal of noise while emptying the dishwasher. “He told me to be quiet,” she said. “It was wonderful.”
Still, the availability of these implants can discourage parents of deaf children from having them learn a full range of methods to cope with their disability. Coupled with implants, these techniques, which include lip reading and sign language, can give a child a particularly broad ability to get along in a world of people with normal hearing. But parents of children with implants face the temptation to rely on the devices alone, ignoring the low-tech approaches.
Roslyn Rosen, a deaf woman who has served as president of the National Association of the Deaf (NAD), said in 1993: “Most of the cochlear implant children are still deaf children. And at $2.0,000 to $40,000 for the procedure, the real losers are the child, the family, and the insurance companies.” Like many activists, Rosen wanted deaf children to grow up as part of the deaf community instead of being taught to shun it or to consider deafness a sign of inferiority. That same year, Harlan Lane, a psychology professor with normal hearing at Northeastern University, who has chaired a NAD study group on cochlear implants, noted a danger that “the child might end up in a no-man’s land,” unable to communicate easily with either the deaf or the hearing. In recent years the NAD’s opposition to implants for young children has softened. The organization now stresses the need for parents to inform themselves about all options before making a choice, and the importance of teaching implanted children sign language and other aspects of deaf culture.
There are perhaps some 300,000 profoundly deaf who can benefit from cochlear implants. More than 30,000 of them now use devices built by Cochlear, the market leader. Still, hearing aids remain the mainstream technology for most of the deaf or hearing-impaired. More than 4,000,000 people use them in the United States, and the devices continue to improve. The hearing aid has long been a focus for advanced electronic technology, for it demands compactness, while providing high-value services. Thus, in 1953, hearing aids became the first commercial products to introduce transistors in place of vacuum tubes. They have continued to lead the way for other electronic applications requiring miniaturization and signal-processing capacity.
With hearing aids providing an increasingly wide range of cost and performance, the cochlear implant is not likely to make them obsolete. The high cost of an implant and its associated surgery will limit the number of recipients. Still, every successful implant carries its own story of an individual life made fuller and richer.
A woman named Joanne Syrja had never been able to hear high frequencies, and her hearing deteriorated through the years until she was left in profound deafness. Then, in 1991, when she was 44 years old, she received a cochlear implant. It gave her the ability to hear people with ease, even when talking on the telephone, and it did much more.
“It was last March,” she said about a year later. “I was going into my office, and there are some trees right in front of the doorway. I heard a noise, and I saw this bird sitting there. Every time its mouth opened, I heard this noise. I never heard birds before in my life. It was incredible. I just stood there with tears running down my face. I knew what I heard was the bird chirping, and it was beautiful.”
