Every day in America, parents confront fear when they are told their child has hearing loss, and up to 25,000 children are given this diagnosis every year. Most present to the physician's office within the early years of life. Fortunately, most of these children have a conductive loss due to chronic serous otitis media and fluid within the middle ear space. In this case, there is usually a history of repeated ear infections, and multiple antibiotic treatments, and the hearing loss is generally mild to moderate. When chronic, this problem may be dealt with by draining the middle ear fluid and placing tympanostomy tubes.
Children with sensorineural or inner ear hearing loss require more intensive investigation and long-term evaluation and treatment, but their conditions may be improved. Children with a sensorineural hearing loss may have varying degrees of disability, ranging from mild to moderate to severe or profound hearing loss. Children with a mild to severe cochlear hearing loss usually benefit from amplification and are fitted with hearing aids accordingly. Children with a severe to profound cochlear hearing loss usually do not benefit from hearing aids and are candidates for cochlear implantation. These children are the subject of this article.
There are many causes of severe to profound sensorineural hearing loss in children. The most common is a congenital defect. These are usually within the sensory epithelium of the inner ear and not visible with standard x-rays. Occasionally, computerized tomography scans will identify a more severe anomaly within the cochlea as a bony defect of the inner ear. Early hearing loss may also be acquired as a result of meningitis. All children suffering from meningitis should undergo extensive auditory evaluation to exclude hearing loss. Infants born prematurely or with a severe medical disorder should also be evaluated for hearing loss.
The goal of assessing auditory function is to determine the degree of hearing loss and to rule out other causes of deafness. Clinical evaluation usually includes a complete history and a physical examination; a microscopic examination of the ear by an otologist; an audiologic evaluation, including audiometry, auditory brainstem evaluation, and otoacoustic emissions; and imaging studies when indicated.
Testing may first be available in neonates, and universal infant hearing screening provides testing of all infants prior to discharge from the nursery. This is gaining national momentum as the National Institutes of Health has come out in its support. Seven states now mandate this and many others are considering similar legislation.
Current screening of infants is painless, quick, and inexpensive. This may involve two tests: an auditory brainstem response and otoacoustic emissions. These complement one another. The auditory brainstem response assesses the hearing by neurologic response. It can be done without sedation while the child is sleeping. A click tone is presented to the ear while electrodes are placed on the scalp to record the brain wave activities. This test evaluates high-frequency hearing and can usually identify the level of a loss up to approximately 80 dB, which is in the severe range.
Otoacoustic emissions evaluation assesses the function of the inner ear hair cells. The response is either present or absent; it will be absent if hearing thresholds are above 35 dB, which is in the mild range. Presently, otoacoustic emissions do not rule out a mild hearing loss, but indicate the presence of normal hearing or a mild hearing loss versus one above 35 dB. If otoacoustic emissions are present, the child typically has enough hearing to develop normal speech and language skills. Absence of otoacoustic emissions means that the hearing loss is at least moderate and may be severe to profound. If the otoacoustic emissions test is absent, further testing with auditory brainstem audiometry must be performed to define the level of loss. Both tests are easily performed, as neither requires a behavioral response from the infant. Together, there tests allow determination of hearing level.
DEFINITION OF PREUNGUAL OR POSTLINGUAL HEARING LOSS AND COCHLEAR IMPLANTS
Most investigators agree to classify deafness that occurs before 2 years of age, including deafness that is congenital, as prelingual and that ocairring after the age of 5 as postlingual. These terms relate onset to the development of oral language skills. There is less agreement regarding the classification of children whose deafness occurs between 2 and 5 years of age. The incidence of acquired deafness is highest during the first 2 or 3 years of life, whereas deafness acquired after age 3 is relatively uncommon. The etiology of acquired deafness (whether prelingual or postlingual) is most often meningitis, with late-onset inherited types, ototoxic medications, and other infections accounting for relatively few cases.
Age at cochlear implantation appears to correlate with results of implant performance when the onset of hearing loss is postlingual. However, the majority of children who are implant candidates sustain childhood deafness before age 2 or 3 years. There appears to be negligible difference in the postlingual performance of children with congenital or early acquired deafness. Age at implantation is emerging as an important variable, but more research is needed in that area.1
CURRENT INDICATIONS FOR COCHLEAR IMPLANTS IN CHILDREN
Original criteria for candidate selection required a bilateral profound hearing loss and a lack of benefit from conventional amplification.2 Other studies support implantation of the poorer ear in a child who has had suitable training with well-fitted hearing aids and a plateau in soundonly speech recognition below the point of understanding what is spoken in everyday life.3 More recently, profoundly deaf children who have worn well-fitted hearing aids and received intensive auditory training are also appropriate candidates, if they can only detect sound or recognize closed-set words based on intensity or duration cues (ie, categories 1 and 2).4
Thus, as the performance of children with implants has become more impressive, there is a trend toward the inclusion of children with some residual hearing in the selection criteria for implantation.5,6 Profoundly deaf children with severe mental retardation, with schizophrenia, or with no oral or aural communication skills by approximately 8 years of age do not appear to be appropriate implant candidates. Audiologic criteria to determine which ear to implant are based on aided speech recognition data, electrical stimulability, and length of auditory deprivation.
TIMING OF IMPLANTATION IN CHILDREN
With increased familiarity with cochlear implantation, a trend toward earlier implantation has emerged to ameliorate the devastating effects of early auditory deprivation. Because the development of speech perception and production skills normally begins at an early age, performing implantation in congenitally or neonatally deaf children younger than 2 years has substantial advantages over implantation at a later age. To assess the growth pattern of the temporal bone, Eby and Nadol63 studied temporal bone histology and skull x-rays. Their studies showed no histologic evidence of postnatal growth of the inner ear and minimal growth of the middle ear. The radiologic studies indicated growth of the mastoid bone in all directions. They concluded that at least a 25-mm lead wire expansion should be allowed to accommodate skull growth.
Other considerations of importance in the timing of cochlear implants in young, deaf children relate to the reliability of hearing tests in infants and the control of middle ear infections.7
DIFFERENT TYPES OF IMPLANTS
Presently, two devices are approved by the Food and Drug Administration (FDA) for use in children. These are the Clarion (Advanced Bionics, Sylmar, CA) system and the Nucleus (Cochlear Corp., Englewood, CO) system. The Medel system is currently under investigation by the FDA. All use multichannel electrodes. These newer generation devices are superior to and have replaced the single-channel systems that were available earlier. Current cochlear implant systems consist of two major components. There is an internal implanted receiver electrode system and an external speech processor component. The speech processor has a microphone that is worn similarly to a behind-the-ear hearing aid. The microphone picks up sound and transmits it to the speech-processing computer. This device is housed in a small box similar to a portable radio /cassette player that may be worn on a belt. Children sometimes put it in a small backpack.
The speech processor breaks up the sound into components and determines how the sound source should stimulate the cochlear nerve.
These systems use a transcutaneous electrode transmission system. An external coil creates an electrical field that transmits an electrical signal across the skin to the internal compartment, which transmits an electrical current by multiple electrodes into the cochlea. Its electrical wire fits into the scala tympani of the cochlea and rests against the cochlear nerve. Electrical stimulation of the cochlear nerve then occurs via this electrode. The external device can be changed as new software programs are made available to break up speech and present it in better ways to the cochlear nerve. Patient performance with multichannel devices is significantly better than with single-channel devices.
The designs of the three devices mentioned above vary. The Nucleus and Medel systems have straight electrodes that follow the outer wall of the cochlea when introduced through the round window. The Clarion electrode is pre-curved to assume a more intermediate position within the scala tympani of the cochlea. A newer design includes a positioner that is inserted alongside the electrode to bring it closer to the modiolus where the spiral ganglion cells and neural elements reside. These are the structures that are stimulated by the cochlear implant.8 The Nucleus and Clarion electrodes are 25 mm long. The Medel electrode is longer (32 mm), thus reaching deeper into the scala tympani of the cochlea. The Nucleus system has 22 active electrodes and 10 inactive stiffening rings. The Clarion system uses 8 bipolar electrodes. The Medel system has 12 electrodes.
The processing strategies used with these three systems are also different. Cochlear implant professionals are constantly working on newer and more efficient processing schemes to improve the performance of cochlear implants. For the same implant, performance can significantly improve with the application of new and more effective processing strategies. For the Nucleus device, a significant improvement in open-set speech recognition occurred when the patients were switched from the early FOF1F2, a feature extraction strategy, to Mpeak, a filter bank plus feature extraction strategy. This tendency was further substantiated recently when the feature extraction processing strategy for the Nucleus device was replaced by SPEAK, a nonsimultaneous filter bank spectral peak coding system with more rapid processing capabilities. According to the Cochlear Corporation (unpublished data, March 1995) and others,9'10 40% of the Mpeak users had significant improvements on tests of open-set, sound-only word and sentence recognition, particularly when there was background noise. In summary, the results of many studies, including a large veterans' Administration project, clearly demonstrated the capability to improve speech recognition by changing the speech-processing strategy without disturbing the implanted electrode array.
SURGICAL PROCEDURE FOR COCHLEAR IMPLANTS
The surgical steps involved in the placement of a cochlear implant are the same regardless of the type of electrode used. A postauricular incision is made a few millimeters behind the postauricular fold and carried above the pinna for 3 to 4 cm before it is slightly angled posteriorly. This incision is carried to the level of the temporalis fascia. An incision is then made through the temporalis muscle and the musculoperiosteal layer overlying the mastoid bone. The resulting flaps are retracted so the spine of Henle and skin of the external auditory canal are identified.
The next step involves creating a small mastoidectomy opening. The bony edges of the mastoidectomy are kept in place because they will help anchor and stabilize the electrode in the mastoid. With the use of the incus bone as a landmark, the facial recess is opened and widened. For the Clarion electrode, the facial recess needs to be opened widely. Through the facial recess, the round window of the cochlea is identified. The upper lip of the round window is drilled away and the round window membrane is identified. An opening is then created in the anterior and inferior part of the window. The electrode array is placed through this opening into the scala tympani of the cochlea. The electrode surrounds the modiolus or cochlear nerve, thus allowing electrical current to stimulate the nerve. A well is then created in the bone over the mastoid for the internal receiver. The internal receiver is then fixed into position using silk sutures.
The round window and the facial recess are sealed with soft tissue and the electrode loops are placed under the edge of the bone. The wound is then closed in layers and a mastoid pressure dressing is applied for 24 hours. The wound heals readily and the external speech processor is programmed 6 weeks after surgery.
OUTCOME OF COCHLEAR IMPLANTS
Pediatric use of the Nucleus, SPEAK, and CIS strategies in the Clarion and Medel Combi-40 started in 1994. There is evidence suggesting that children achieve higher levels of performance with these strategies than with those used previously. The performance of implant subjects was significantly higher than that of the silver hearing aid users on all measures, and similar to that of the gold hearing aid users on vowel recognition and auditory-plus-visual sentence recognition. Silver hearing-aid users demonstrate unaided thresholds between 101 and 110 dB at two of the three (500, 1,000, and 2,000 Hz) frequencies. Gold hearing-aid users demonstrate thresholds at 500, 1,000, and 2,000 Hz between 90 and 100 dB HL, with no threshold greater than 105 dB HL at the three frequencies.
These data indicate that children with hearing levels greater than 100 dB HL could benefit more from a multichannel cochlear implant than from continued use of hearing aids only. Pediatric implant users derive substantial benefit from multichannel cochlear implants, but these benefits develop over a long time. Children's performance is continuously improving with new processing strategies that provide a better representation of the speech waveform.11
The purpose of providing a prelingually deaf child with a cochlear implant is to make it easier for the child to learn to talk and to communicate. Cochlear implants complemented with a strong auditory oral education program make it possible for children to acquire competent spoken language.
Advances in surgical technique, the development of smaller devices, the early diagnosis of hearing loss, the early use of appropriate amplification, and the demonstrated absence of an increased rate of complications make it seem likely that the prudent age limit for cochlear implantation for children can safely be lowered to younger than 2 years.
1. Osberger MJ, Todd SL, Berry SW, Robbins AM, Miyamoto RT. Effect of age at onset of deafness on children's speech perception abilities with a cochlear implant. Ann Otol Rhinol Laryngol. 1991;100:883-888.
2. Chute PM. Residual hearing in children. Presented at Cochlear Implants in Adults and Children, 100th NIH Consensus Development Conference; Bethesda, Maryland; May 15-17, 1995.
3. Skinner M, Clark GM, Whitford LA, et al. Evaluation of a new spectral peak coding strategy for the Nucleus 22 Channel Cochlear Implant System. Am ] Otol. 1994; 15(suppl 2):15-27.
4. Staller SJ, Dowell RC, Beiter AL, Brimacombe JA. Perceptual abilities of children with the Nucleus 22 Channel Cochlear Implant. Ear Hear. 1991; 12(suppl):34S-47S.
5. Zimrnerrnan-Phillips SR, Heiber S, Zwolan X Kileny P, Moeggenberg C, Telian S. Changing audiologic criteria in pediatric cochlear implant recipients. Presented at the Second European Pediatric Cochlear Implant Conference; Montpellier, France; May 1994.
6. Gantz BJ, Tyler RS, Woodworth GG, Tye-Murray N, Fryauf-Bertschy H. Results of multichannel cochlear implants in congenital and acquired prelingual deafness in children: five-year follow-up. Am J Otol. 1994;15(suppl 2):1-7.
6a. Eby TL, Nadol JB. Postnatal growth of the human temporal bone: implications for cochlear implants in children. Ann Otol Rhinol Laryngol. 1986;95:356-364.
7. Miyamoto, RT. Timing of implantation in children. Presented at Cochlear Implants in Adults and Children, 100th NIH Consensus Development Conference; Bethesda, Maryland; May 15-17, 1995.
8. Fayad J, Linthicum FH Jr, Otto SR, Galey FR, House WF. Cochlear implants: histopathologic findings related to performance in 16 human temporal bones. Ann Otol Rhinol Laryngol. 1991,100:807-811.
9. Waltzman SB, Cohen NL, Gomolin RH, Shapiro WH, Ozdamar SR, Hoffman RA. Long-term results of early cochlear implantation in congenitally and prelingually deafened children. Am J Otol. 1994;15(suppl):9-14.
10. Osberger MJ, Robbins AM, Miyamoto RT, et al. Speech perception abilities of children with cochlear implants, tactile aids, and hearing aids. Am J Otol. 1991; 12(suppl):105-115.
11. Miyamoto RT, Kirk KI, Todd SL, Robbins AM, Osberger MJ. Speech perception skills of children with multichannel cochlear implants or hearing aids. Presented at the International Cochlear Implant, Speech and Hearing Symposium; Melbourne, Australia; October 24-28, 1994.