Cochlear Implants: Fundamentals and Application - part 8 doc

Special Investigations 571 F IGURE 9.10. A coronal CT section of the cochlea and temporal bone demonstrating (A) the basal turn of the cochlea, (B) the tympanum, (C) the horizontal part of the facial nerve, (D) the jugular bulb, and (E) the mastoid air cells. A large emissary vein can cause brisk bleeding, and an air cell inferior to the cochlea could be mistaken for the round window. By comparing the image from the two sides it is possible to detect fluid within the air cell system or pathological erosion of the internal meatus. In meningitis, the most common cause of acquired total deafness, the inflam- matory changes may extend throughout the basal but also the middle and apical turns. There is frequently ossification from the round window for up to 6 mm along the scala tympani. These changes can usually be detected by CT scans. Labyrinthitis ossificans can extend to total obliteration of the cochlear duct by patchy or uniformly opaque bone. Some cases have fibrous material obliterating the cochlea, with the result that the CT scans appear to indicate implantable ears, whereas at surgery the scala tympani is filled with scar tissue that adheres to the bone and basilar membrane along the spiral. In cases of meningitis, examine not only the clarity of the basal turn but also the upper turns as opacification would suggest the presence of fibrous tissue. Magnetic resonance imaging (MRI) is essential in excluding fibrous tissue in the cochlea. There are a variety of changes in ears affected by otosclerosis (Damsma et al 1984; Mafee et al 1985). The otospongiosis reduces the opacity of the otic capsule, and increases the diameter of the cochlear duct. In addition, there are bony ac- cumulations at the round and oval windows. The changes seen as a result of these 572 9. Preoperative Selection processes include obliteration of the round window, roughened walls of the co- chlear spaces, mottling of the otic capsule producing a double-barrel appearance, and sometimes widening of the cochlear perilymph spaces. Once again the ap- pearance of the two sides should be compared to determine which ear is more suitable for implantation. It is important to have life-sized images to measure the width of the scala tympani and determine whether the electrode array will pass. There are many effects of trauma in producing total deafness. Most cases with loss of hearing from a head injury have no fractures involving the cochlea. In these cases care must be taken that the bony defect of a craniotomy is clear of the planned surgical site. Fracture lines may be seen in the internal auditory canal, up to the cochlea or all the way through the modiolus or vestibule. In such cases it is important to check that there is no evidence of cerebrospinal fluid within the middle ears by comparing the aeration of the two air cell systems. In the Mondini dysplasia the cochleae have one to one-and-a-half turns or the dysplasia may appear as a common cavity. In some cases a well-defined modiolus is present and in other cases it is absent. If present, it indicates that a perimodiolar array such as the Nucleus Contour can be used as the spiral ganglion cells lie centrally. But if the modiolus is absent, the nerves lie peripherally and a straight but flexible array that lies around the periphery would be preferable. With the Mondini dys- plasia there also may be a large vestibule, wide semicircular canals, and an en- larged cochlear duct. It often also may be found in sudden deafness. It is important to recognize this dysplasia as it may be associated with a perilymph gusher. MRI MRI (Fig. 9.11) is required to exclude the presence of fibrous tissue in the scala when the CT scan shows no evidence of new bone after meningitis. MRI is also valuable in assessing the thickness of the cochlear nerve in children with an inherited hearing loss and any brain abnormalities. Electrical Stimulation of the Promontory For the cochlear implant to produce hearing sensations, a sufficient number of auditory neurons must be capable of being stimulated electrically. That is, the profound hearing loss should be largely cochlear in origin. Standard audiological and medical tests to differentiate between cochlear and retrocochlear losses cannot be used with profound losses. Therefore, electrical stimulation of the promontory was required. However, as indicated above it is not used routinely, but the methods and results are discussed below. A needle electrode (such as is used in electrocochleography) is inserted through the tympanic membrane with the tip resting on the promontory in the middle ear, as close to the round window as possible. The needle is best inserted with local anesthesia by iontophoresis. The needle may be held in place by a rare earth magnet fixed to a plastic ear speculum. Small electric currents are passed between the needle electrode and a surface electrode placed on the ipsilateral cheek. Pa- Special Investigations 573 F IGURE 9.11. Magnetic resonance imaging (MRI) showing the cochlea (C). tients are asked to report any sensations elicited. If they report some sensation, they are asked whether it is a hearing or a tactile sensation, whether it is contin- uous or intermittent, what pitch it is, and what loudness level. The timing of the stimulus presentation is then varied to check that the sensations reported corre- spond to the stimuli. For instance, the stimulus is presented with a rhythm (half a second on and half a second off) and the patient needs to show that the timing of the sensation corresponds to the timing of the stimulus (for example, by moving the hand in time). It is important to verify the sensations reported, as tinnitus and a desire to hear can give misleading reports. To be considered a positive result, the patient need only report consistent hearing sensations. However, further test- ing is carried out in an attempt to gain some knowledge of the neuronal population and the perceptual abilities of the patient. The pulse rate of the stimulus is varied to determine whether the patient can differentiate between pulse rates on the basis of pitch (as the pulse rate is increased from 100 to 200 Hz, the pitch sensation should increase). The minimum gap perceived is determined using two stimuli, with a break in the middle of one. The patient needs to determine which stimulus has the break in it. The breaks are progressively reduced from approximately 150 ms to the level at which the patient reports no break and is unable to determine the correct response. In many adults with acquired deafness, the minimum gap detection is less than 10 ms. The dynamic range for different frequencies may also be of value in assessing performance. An adaptation test is carried out using a constant stimulus, presented at a most comfortable level. The stimulus is pre- sented for a period of 60 s, and patients are asked to raise their finger for as long as they hear the sound. Abnormal adaptation is said to occur if the stimulus is 574 9. Preoperative Selection not heard for the full 60 seconds. If this occurs, a very careful review of the history and the overall results is done to determine the likelihood of a retrococh- lear cause for the deafness. No firm conclusions can yet be made about the relationship between the prom- ontory stimulation results and results with the cochlear implant. It does, however, seem evident that good pulse rate discrimination, minimum gap detection of ap- proximately 10 ms, and no adaptation auger well for the patient’s cochlear implant use. What is less clear is the relationship between relatively poor results on the promontory stimulation test and subsequent cochlear implant use. Many factors can contribute to the promontory test results, and not all of these are relevant to the use of a cochlear implant. In particular, stimulation of nonauditory neurons (such as the tympanic branch of the glossopharyngeal nerve) can make the pro- cedure uncomfortable and can lead to poorer results. In addition, electrical stimulation tests can provide further information to fa- cilitate a comparison between the ears. It is useful if one ear was deafened for a longer period of time or was congenitally deaf. If all else was equal (CT scans and audiological results), the ear with the better promontory stimulation result would be chosen for cochlear implant surgery. For this reason, both ears would be tested with promontory stimulation. An addition to the above tests is the use of a speech processor and the receiver- stimulator from a cochlear implant to enable speech information to be presented to the person during the promontory stimulation test. The features presented are fundamental frequency and intensity-timing cues, as there is only one electrode channel. The ability of the patient to discriminate on the basis of these cues is then assessed. This provides additional information about the patient’s perceptual processing abilities and also helps to shape the patient’s expectations more real- istically, particularly for those who have been profoundly hearing impaired for a long period of time. It is necessary to explain that this would be the minimum the patient would expect to receive. For the vast majority of profoundly hearing-impaired patients who have un- dergone the promontory stimulation test, the result has been positive, suggesting sufficient residual auditory neurons to allow successful cochlear implant use. However, several patients have obtained negative results. This occurred primarily with a stimulus that had biphasic pulses of relatively short pulse width. This led to uncomfortable tactile sensations and pain more often than with a square wave pulse train. Patients with a negative result have later successfully received a co- chlear implant. For this reason electrical stimulation is not used routinely. How- ever, one patient in the Melbourne clinic had a definite negative result (where there was no hearing sensation and also no uncomfortable tactile sensations). It was subsequently discovered that this person had a disease causing neural de- generation of the central nervous pathways. Vestibular Assessment Preoperatively, a vestibular assessment is routinely carried out, using electronys- tagmography. The stimuli for the caloric stimulation are the routine warm (44ЊC) Management 575 and cool (30ЊC) water irrigations, but also a cold (less than 5ЊC) water irrigation. This latter stimulus can differentiate between a hypoactive labyrinth with minimal function and an inactive labyrinth. The results of the vestibular assessment are typically not used to determine suitability for the cochlear implant operation. Instead, they are helpful in making predictions about the likelihood of temporary postoperative vertigo. They can also be used as a reference in the unlikely event of a person developing balance problems at a later stage. Management In selecting adults and children for surgery it is important to ensure that their hearing status is such that they would do better with an implant than a hearing aid. The factors that predict successful outcomes are also important as parents wish to know how well their child is likely to perform after surgery. Hearing and Speech Perception On the basis of the tests described, it is possible to place the patients into one of three categories with respect to speech discrimination: Category I These are patients suitable for implantation of either ear. There is no significant auditory discrimination of open-set words or sentences, nonsignificant scores on closed-set tests of spectral discrimination, and no significant aid to speech reading in either ear. Category II These patients are suitable for implantation of the unaided ear. The aided ear (the better ear) has no open-set speech discrimination, but may have some significant closed-set speech discrimination that provides a significant aid to lipreading. Category III These patients may be suitable for implantation. The aided ear has less than 30% to 40% open-set word discrimination, and the ear to be implanted shows no significant open-set speech discrimination but obtains significant closed-set speech discrimination which aids speech reading. It is also possible, however, to consider a person for an implant on the basis of excessive recruitment (where the dynamic range is severely reduced) or where use of a hearing aid produces excessive tinnitus, dizziness, or other uncomfortable feeling. 576 9. Preoperative Selection Predictive Factors The selection of a patient for a cochlear implant requires a number of consulta- tions, special tests, and counseling as discussed below. This process is conducted by a number of clinicians, and progress should be reviewed at appropriate stages by a team headed by an experienced clinician. The factors that predict successful outcomes are important for selecting pa- tients. Patients, and in the case of children their parents, wish to know how well they are likely to perform after surgery. Knowledge of the factors predicting outcomes helps answer patients’ questions. General Predictive Factors for Adults The general predictive factors for adults are age when deafened, age at implan- tation, duration of deafness, duration of implantation, etiology, the presence of progressive hearing loss, degree of residual hearing, speech reading ability, speech-processing strategy, and medical condition. These factors are similar to those for children. The factors that correlate with speech perception results have been evaluated in a number of studies (Dowell et al 1985; Hochmair-Desoyer and Burian 1985; Nadol et al 1989; Parkin et al 1989; Dorman et al 1990; Blamey et al 1992). The analyses were undertaken on patients with the Nucleus F0/F2, F0/ F1/F2, and the Multipeak-MSP systems and the Ineraid device. There was considerable variability in the results in adults. However, in a study on 64 postlinguistically deaf adults at the University of Melbourne Clinic using the Nucleus F0/F1/F2 strategy, 43% of the variance of the CID sentence scores was accounted for by the duration of deafness, frequency discrimination, and gap detection for the promontory tests, the number of electrodes in the cochlea, and the dynamic range of the intracochlear electrodes. In a combined study by Blamey et al (1996) on 808 patients, duration of deafness, age at onset of deafness, eti- ology, and duration of implant experience accounted for 21% of the variance, of which duration of deafness was the greater part (13%). So it is still not possible to confidently tell patients how well they will perform, and more research is required before better prediction can be achieved. Age When Deafened Age when deafened is not a significant factor in postlinguistically deaf adults as it is with children who are deafened in their critical period of language devel- opment. It has little effect up to 60 years (Blamey et al 1996). Age at Implantation Age at implantation and duration of deafness were both interrelated. In the adult, they can be separated and both correlate negatively with results. Initially older people had a longer duration of deafness when presenting for implantation. Now people come with a shorter duration of deafness so age and duration do not correlate. A further analysis of adult results showed they were only poorer if the Management 577 patient was over 60 years (Blamey et al 1995). In addition, learning is more difficult for the elderly for a number of perceptual tasks. Duration of Deafness Blamey et al (1992) found that duration of deafness was the main general factor that correlated with speech perception. This was supported by the findings of Gantz et al (1993). Shipp and Nedzelski (1994) reported a strong relation between duration of deafness divided by age and a composite measure of auditory per- ception. With deafness of long duration, adults are likely to require long periods of rehabilitation for adequate speech perception, and the results are poorer (Dow- ell et al 1997; Tomblin et al 1999). However, with a long duration of deafness the person can often perceive phonemes and have good results for place pitch discrimination, but cannot so readily integrate the information and understand speech. Duration of deafness may have its effect through a greater loss of the neurons and their connections. Etiology (Cause of Deafness) With etiology, Meniere’s disease correlated positively, and meningitis negatively with results in the adult. This may be the result of the reduced number of elec- trodes inserted due to labyrinthitis ossificans (Blamey et al 1995). Progressive Hearing Loss and Residual Hearing A progressive hearing loss is associated with better results, as is the presence of some residual hearing (Gantz et al 1993). If there has been a progressive hearing loss, the patient will have learned to use degraded auditory information, and this skill will subsequently be useful when given a cochlear prosthesis. Speech Reading Ability There is a weak correlation between speech reading ability and speech perception and this may be because it reflects good top-down processing skills (Cohen et al 1993; Gantz et al 1993). Speech-Processing Strategy The speech-processing strategy has a marked effect on results. Improvements have primarily come by presenting additional frequency information on a place- coding basis. In postlinguistically deaf adults this has been seen with the Nucleus speech-processing strategies where there has been a progressive improvement in scores from the F0/F2 to F0/F1/F2 to Multipeak to spectral maxima sound pro- cessor (SPEAK). The addition of more temporal information with the higher rate of stimulation with the ACE strategy has also helped. The presentation of infor- mation at a higher rate has also led to an improvement from a fixed-filter strategy with interleaved pulses (IPs) to continuous interleaved stimulation (CIS). This was discussed in more detail in Chapter 7. 578 9. Preoperative Selection Duration of Implantation Finally, the duration of implantation is strongly correlated with good speech per- ception. Learning is required to use the speech-processing strategy in postlin- guistically deaf adults, and the learning is less and steeper when the strategy provides more information and is more speech-like. Not only did word and sen- tence scores improve over time, but to a lesser extent so did vowel and consonant recognition. Tye-Murray et al (1992) found that a phoneme composite score in- creased by an average 8.6% in the first 9 months of implant use, and a further 4.4% in the second 9 months. There were small nonsignificant changes thereafter. This applied to both the Nucleus Multipeak (n ס 13) and Ineraid (n ס 14) patients. Medical Condition There may be medical issues that influence the results. For example, neurosyphilis and schizophrenia are conditions that affect central auditory processing or cog- nitive ability. Specific Predictive Factors for Adults The specific factors predicting speech perception scores in the adult are electrical stimulation of the promontory results, length of insertion and the number of stim- ulating electrodes, and dynamic range. Electrical Stimulation of the Promontory Results There was a positive correlation between preoperative tests of temporal process- ing via promontory stimulation of the auditory nerve (electrode placed on the medial wall of the middle ear), and speech perception results in 64 postlinguis- tically deaf adults at the Melbourne Clinic (Blamey et al 1992). Discriminating gaps smaller than 50 ms for low rates of stimulation and pitch changes for rates of 100 and 200 pulses/s suggested a good result. The ability to detect changes in rate of stimulation and gaps between stimuli appeared to be a more central func- tion, and thus important for segmenting speech and processing the slow frequency changes occurring in voicing. If duration difference limens were large, speech results would be poor and vice versa (Blamey et al 1992). Length of Insertion of the Electrode Array and Number of Stimulus Electrodes A positive relationship was seen between the length of insertion and the number of electrodes. Both correlated positively with speech perception. Two studies were carried out on adults, and showed that there was increasing benefit in having additional electrodes up to 20. A regression analysis by Blamey et al (1992) revealed there was a difference between nine and 21 electrodes (12 electrodes) that accounted for a 24% increase in score (i.e., 2% per electrode). Studies also demonstrated that 20 rather than eight banded electrodes provided improved speech processing for the Nucleus Multipeak and SPEAK in noise. These results Management 579 highlight the importance of multiple-channel stimulation for coding the spectral information in speech. In contrast, the number of stimulus channels for a fixed filter (modified channel vocoder) was shown by Dorman et al (1989) to be four, but this did not apply to noise. In addition, Shannon et al (1995) found with acoustic models of fixed-filter strategies that a four-channel processor gave hearing listeners near-normal speech recognition in quiet listening conditions. So although there is some evidence that fixed filter strategies may not require more than at most seven electrodes, there is a definite correlation between the number of electrodes with the Multipeak and SPEAK strategy and speech per- ception in noise. Dynamic Range Postoperatively there is a positive correlation between the dynamic range and speech score (Blamey et al 1992). The greater the dynamic range between the threshold and maximum comfortable level, the more steps in loudness for pre- senting speech. A greater dynamic range is due to a higher density of spiral ganglion cells as seen in the studies on the human temporal bones from patients in whom psychophysics data were available (Kawano et al 1995, 1996, 1998). General Predictive Factors for Children In Melbourne, after establishing the benefits of the multiple-channel cochlear implant with F0/F2 and F0/F1/F2 speech-processing strategies in adults, the first three children were implanted in 1985 and 1986. This was the start of an inter- national trial for the FDA to determine whether the multiple-electrode cochlear implant and F0/F1/F2 strategy would benefit children. The FDA approved the device as safe and effective for children 2 years of age and older in 1990. There are similar general predictive factors for the child as for the adult: age when deafened, age at implantation, duration of deafness, duration of implanta- tion, etiology, the presence of a progressive hearing loss, degree of residual hear- ing, speech reading ability, speech-processing strategy, and medical condition. These factors have been evaluated in a number of studies (Quittner and Steck 1991; Blamey et al 1992; Gantz et al 1993; Battmer et al 1995; Blamey et al 1996). Quittner and Steck (1991) also had parents rate implant usage and found that a positive rating correlated with communication mode, time using the implant device, and performance on two subsets from the Wechsler Intelligence Scale for Children–Revised (WISC-R). In a study by Dowell et al (1995) the ability of children (n ס 100) with the Nucleus 22 system was categorized and analyzed with speech perception as the dependent variable and preoperative and postoperative parameters as the inde- pendent variable. This showed that the duration of the hearing loss correlated negatively with perception, and a progressive loss, useful preimplant hearing, experience with the implant, and an auditory-oral education all correlated posi- tively. (An auditory-oral education focuses on developing spoken language through hearing and lipreading.) These five variables accounted for 37% of the overall variance. 580 9. Preoperative Selection In a larger study on 167 children from the Universities of Melbourne and Sydney (Sarant et al 2001) the relation between speech perception and possible contributory factors was assessed using analysis of covariance with the general linear model of Minitab Version 12 (Ryan and Joiner 1994). There were five factors that had a significant effect on phoneme scores for the PBK (Haskins 1949, 1964) and CNC (Peterson and Lehiste 1962) word tests, and accounted for 51% of the variance. These factors were duration of deafness, implant experience, communication mode, clinic, and speech processor. If the effect of the clinic was excluded (as they could have different selection criteria and training method), the remaining four factors accounted for 34% of the variance. As with adults, further research is required to determine which children are most likely to do best. As discussed below, the differences between the Melbourne and Sydney clinics were that the children in Sydney had a higher incidence of residual hearing and audi- tory-oral education. The factors producing variance in speech perception and production, as well as spoken and written language for the Nucleus SPEAK strat- egy, were determined on 136 children by Geers et al (2002). The data were classified as independent variables (communication mode, classroom, therapy), and intervening variables (characteristics of the child: age, age at onset, age at implantation, IQ; family: size, parent’s education; and implant). The child and family characteristics (primarily the nonverbal IQ) contributed approximately 20% of the variance. An additional 24% was due to implant characteristics, and 12% for educational factors. In a group of older children (n ס 25) between 8 and 18 years of age, the main factors correlating with speech perception were duration of the profound loss, preoperative sentence score, and equivalent language age. These factors ac- counted for 66% of the variance. These children had a mean sentence score of 47% that was statistically the same as the overall group in Melbourne (Dowell et al 2002b). Age When Deafened If the hearing loss occurred after 4 to 6 years of age, the person is postlinguisti- cally deaf, and could expect the results normally obtained by people who have lost hearing after developing language. If the hearing loss occurred before 4 to 6 years, the person is prelinguistically deaf, and the results depend on a number of factors, which include age and language skills. In the study with the Nucleus 22 F0/F1/F2 strategy for the FDA, the results were categorized into the children’s best perception levels for detection, pattern, closed set, and open set, and it was discovered that the closed- and open-set results were significantly better postoperatively for both the pre- and postlinguistic groups (Staller 1990; Staller et al 1991a,b). The open-set scores, however, were better for the postlinguistic group as illustrated in Figure 9.12. The trial showed that 60% of children born deaf were able to understand some open-set speech. Age at the onset of deafness was shown to be a significant predictor of speech [...]... round window favored by House and Urban (1973), Clark, Patrick et al (1979), Eddington et al (19 78) , Parkin et al (1 985 ), Burian et al (1 986 ), Portmann (1 986 ), Chouard and MacLeod (1976), Lacombe et al (1 989 ), Eddington et al (19 78) , and Lacombe et al (1 984 ) A trans-external canal approach was advocated by Simmons (1966), Michelson and Schindler (1 981 ), and Banfai et al (1 984 ) In the latter cases problems... sealing procedures and permeability studies Acta Oto-Laryngologica (suppl 410): 5–15 Clark, G M., Y C Tong, L F Martin and P A Busby 1 981 A multiple-channel cochlear implant An evaluation using an open-set word test Acta Oto-Laryngologica 91: 173– 175 Cohen, N L., S B Waltzman and S G Fisher 1993 A prospective, randomised study of cochlear implants New England Journal of Medicine 3 28: 233– 282 Consensus 1995... Martin and B C Pyman 1 985 Selection of patients for multiple-channel cochlear implant patient In: Schindler, R A and M M Merzenich, eds Cochlear implants New York, Raven Press: 403–406 Busby, P A and G M Clark 2000 Electrode discrimination by early-deafened subjects using the Cochlear Limited multiple-electrode cochlear implant Ear and Hearing 21: 291–304 Busby, P A., S A Roberts, Y C Tong and G M... cochlear implants Archives of Otolaryngology 1 08: 4 78 483 Owens, E and C C Telleen 1 981 Speech perception with hearing aids and cochlear implants Archives of Otolaryngology 107: 160–163 Parkin, J L., B E Stewart, K Dankowski and L J Haas 1 989 Prognosticating speech performance in multichannel cochlear implant patients Otolaryngology–Head and Neck Surgery 101: 314–9 Pass, R F., S Stagno, G J Myers and. .. Rickards, Y C Tong and A J Williams 1977 The clinical assessment of cochlear implant patients Journal of Laryngology and Otology 91: 697–7 08 Clark, G M and B C Pyman 1997 Preoperative medical evaluation In: Clark, G., R Cowan and R Dowell, eds Cochlear implantation for infants and children Advances San Diego, Singular: 71 82 Clark, G M., R K Shepherd, B K.-H G Franz and D Bloom 1 984 Intracochlear electrode... Hearing 10: 288 –291 Dorman, M F., L Smith, G McCandless, G Dunnavant, J Parkin and K Dankowski References 589 1990 Pitch scaling and speech understanding by patients who use the Ineraid cochlear implant Ear and Hearing 11: 310–315 Dowell, R C 1997 Preoperative audiological, speech, and language evaluation In: Advances Clark, G M., R S C Cowan and R C Dowell, eds Cochlear implantation for infants and children... American Journal of Otology 20: 4 8 Clark, G M., P J Blamey, P A Busby, et al 1 987 A multiple-electrode intracochlear implant for children Archives of Otolaryngology 113: 82 5 82 8 Clark, G M., P A Busby, S A Roberts, et al 1 987 Preliminary results for the Cochlear Corporation multi-electrode intracochlear implants on six prelingually deaf patients American Journal of Otology 8: 234–239 Clark, G M., B J O’Loughlin,... cochlear implant users Ear and Hearing 16: 551–561 Dawson, P W., P J Blamey, L C Rowland, et al 1995b A clinical report on receptive vocabulary skills in cochlear implant users Ear and Hearing 16: 287 –294 Dorman, M F., K Dankowski and G McCandless 1 989 Consonant recognition as a function of the number of channels of stimulation by patients who use the Symbion cochlear implant Ear and Hearing 10: 288 –291... Barker and G M Clark 2002a Speech perception in children using cochlear implants: prediction of long-term outcomes Cochlear Implants International 3(1): 1– 18 Dowell, R C., S J Dettman, K Hill, E Winton, E J Barker and G M Clark 2002b Speech perception outcomes in older children who use multichannel cochlear implants: older is not always poorer Annals of Otology, Rhinology and Laryngology 111(suppl 189 ):... multichannel cochlear implant Journal of Speech and Hearing Research 35: 913–920 Fryauf-Bertschy, H., R S Tyler, D M Kelsay, B J Gantz and G G Woodworth 1997 Cochlear implant use by prelingually deafened children: the influences of age at implant use and length of device use Journal of Speech and Hearing Research 40: 183 –199 Gantz, B J., R S Tyler, G Woodworth, N Tye-Murray and H Fryauf-Bertschy 1994 . capacity of hearing-impaired children using hearing aids and cochlear implants: issues of efficacy and assessment. Scandinavian Audiology Sup- plementum 46: 17–25. Boothroyd, A. 19 98. Evaluating the. Journal of Otol- ogy 20: 4 8. Clark, G. M., P. J. Blamey, P. A. Busby, et al. 1 987 . A multiple-electrode intracochlear implant for children. Archives of Otolaryngology 113: 82 5 82 8. Clark, G. M.,. detection, pattern, closed set, and open set, and it was discovered that the closed- and open-set results were significantly better postoperatively for both the pre- and postlinguistic groups (Staller