Over the past 30 years, hearing care clinicians have increasingly relied on cochlear implants to restore auditory sensitivity in selected patients with advanced sensorineural hearing loss. This article examines the impact of intervention with cochlear implantation in children and adults. The authors report a range of clinic-based results and patient-based outcomes reflected in the reported literature on cochlear implants. The authors describe the basic assessment of the physiologic response to auditory nerve stimulation; measures of receptive and productive benefit; and surveys of life effects as reflected measures of quality of life, educational attainment, and economic impact.
- •
The primary goal of cochlear implantation in children is to facilitate comprehension and expression through the use of spoken language.
- •
Early educational intervention is associated with improvements in language development after cochlear implantation.
- •
Recent analyses show that in adults, the age at implantation carries minimal or even statistically insignificant predictive power on postimplantation outcomes. The duration of deafness and preoperative speech-perception scores have the highest predictive power on postimplantation outcomes across the adult population.
- •
Age-related degeneration of the spiral ganglion and progressive central auditory dysfunction raise potential concerns about the efficacy of cochlear prostheses in the elderly, but comparable gains in speech understanding have been reported for both elderly and younger groups of implant recipients.
Introduction
Over the past 30 years, hearing care clinicians have increasingly relied on cochlear implants to restore auditory sensitivity in selected patients with advanced sensorineural hearing loss (SNHL). This article examines the impact of intervention with cochlear implantation in children and adults. The authors report a range of clinic-based results and patient-based outcomes reflected in the reported literature on cochlear implants. The authors describe the basic assessment of the physiologic response to auditory nerve stimulation; measures of receptive and productive benefit; and surveys of life effects as reflected measures of quality of life, educational attainment, and economic impact.
Auditory outcomes
Auditory performance is measured at preimplant and postimplant intervals, allowing the assessment of candidacy criteria and longitudinal tracking of the patients’ progress. Measurement variables associated with auditory testing should be standardized as much as possible. Clinicians can choose between closed-set tests (eg, forced choice of 1 answer from a list of 4) and open-set tests (auditory alone without context) of words and/or sentences. Closed-set tests and sentence tests typically produce substantially higher correct percentages than do open-set tests and tests of single words. This difference reflects the amount of contextual information available when word and sentence material are presented. Voice presentation can also affect speech-perception scores, with live presentations typically producing higher rates of correct responses than taped presentations.
Trends toward higher rates of open-set speech recognition with newer implant technology and longer implant experience have prompted calls for more stringent assessments of receptive capability. Increasing the difficulty of a speech-perception test has the effect of limiting the ceiling effect that results from testing with simple, everyday phrases. For the purpose of generating more meaningful comparative data, increasing the test difficulty tends to normalize distributions across populations, thereby enabling more powerful statistical analyses of differences between groups.
Auditory outcomes
Auditory performance is measured at preimplant and postimplant intervals, allowing the assessment of candidacy criteria and longitudinal tracking of the patients’ progress. Measurement variables associated with auditory testing should be standardized as much as possible. Clinicians can choose between closed-set tests (eg, forced choice of 1 answer from a list of 4) and open-set tests (auditory alone without context) of words and/or sentences. Closed-set tests and sentence tests typically produce substantially higher correct percentages than do open-set tests and tests of single words. This difference reflects the amount of contextual information available when word and sentence material are presented. Voice presentation can also affect speech-perception scores, with live presentations typically producing higher rates of correct responses than taped presentations.
Trends toward higher rates of open-set speech recognition with newer implant technology and longer implant experience have prompted calls for more stringent assessments of receptive capability. Increasing the difficulty of a speech-perception test has the effect of limiting the ceiling effect that results from testing with simple, everyday phrases. For the purpose of generating more meaningful comparative data, increasing the test difficulty tends to normalize distributions across populations, thereby enabling more powerful statistical analyses of differences between groups.
Tests of implant performance
The Minimum Speech Test Battery (MSTB) for adult cochlear implant users is a standardized set of comprehensive tests of preoperative and postoperative speech recognition. To minimize the effects of learning and memorization, the word and sentence tests have different lists for at least 6 testing trials. The average and range of performance of cochlear implant users are critical to defining audiologic performance boundaries for implant candidacy, monitor postimplantation results, and facilitate in comparisons across implant designs and coding strategies.
The major components of the MSTB are the Hearing in Noise Test (HINT) and the Consonant/Nucleus/Consonant (CNC) test. The HINT provides a measure of speech-reception thresholds for sentences in quiet and in noise. For high levels of recognition in quiet, the background noise is filtered to match the long-term average spectrum of the sentences. In the MSTB, the HINT sentence lists are presented at 70 dB in quiet and at a +10 dB signal-to-noise ratio (ie, noise at 60 dB). Smaller signal-to-noise ratios (eg, +5 dB or 0 dB) may also be used to avoid ceiling effects. Normal-hearing listeners can comprehend sentences effectively with signal-to-noise ratios down to −3 dB, whereas implant recipients typically show degraded speech recognition when signal-to-noise ratios are lowered beyond +10 dB.
The CNC test consists of monosyllabic words with equal phonemic distribution, with each list of words having approximately the same phonemic distribution as the English language. CNC lists enable performance testing that is likely to represent daily experience with speech stimuli. These tests measure the percentage of words correctly recognized. Revised CNC lists were developed to eliminate relatively uncommon words and proper nouns. More recent observations have stressed the importance of speech test materials that reduce contextual cues in the interest of assessing auditory performance (bottom-up processing) rather than cognitive (top-down processing) components of speech recognition.
Improved speech perception is the primary goal of cochlear implantation. Initial clinical series judged implant efficacy mostly on environmental sound perception and performance on closed-set tests, whereas greater emphasis is now placed on measures of open-set speech comprehension. Speech-perception results from early clinical trials have served to guide the evolution of cochlear implantation.
Clinical observations in patients with current processors indicate that for patients with implant experience beyond 6 months, the mean score on open-set word testing approximates 30% to 60%, with a range of 0% to 100%. Results achieved with the most recently developed speech-processing strategies reveal mean scores more than 75% on words-in-sentence testing, although once again with a wide range of 0% to 100%. Although patients perform substantially poorer on single-word testing, these mean scores continue to improve as the speech-processing strategy evolves. After implantation, speech recognition by telephone and music appreciation are often observed. Again, these benefits seem to be best achieved through more recently developed processing strategies.
The high prevalence of SNHL among the elderly has prompted evaluations of the benefit of cochlear implantation in this age group. For recipients of the cochlear implant after the age of 65 years, open-set speech-recognition scores are not as high as those reported in younger cohorts, potentially representing an effect of longer duration of deafness as opposed to age per se. Nonetheless, implant usage is high among elderly recipients, with nonuse observed in few patients.
Predictors of benefit in adults
The evaluation of the benefit of cochlear implantation in adults has largely focused on measuring gains in speech perception. Assessments of speech recognition in implanted adults offer the opportunity to develop models of benefit prediction. As investigators identify the salient predictive factors, patients’ choices regarding candidacy, device and processing strategy, and the degree of postoperative auditory rehabilitation necessary can be better informed. Various statistical methods have been used to assess speech comprehension using cochlear implants. Multivariate analysis, a statistical technique that determines the contribution of individual factors to variations in performance, is the most commonly used methodology. The following factors have been evaluated:
- •
Patient variables: age of onset, age of implantation, deafness duration, cause, preoperative hearing, survival and location of spiral ganglion cells, patency of the scala tympani, cognitive skills, personality, visual attention, motivation, engagement, communication mode, and auditory memory.
- •
Device variables: processor, implant, electrode geometry, electrode number, duration and pattern of implant use, and the strategy used by the speech processing unit.
Although the factors identified as most determinative have varied with different study populations, the most recent analyses showed that age at implantation carries minimal or even statistically insignificant predictive power on postimplantation outcomes. Rather, it was the duration of deafness and preoperative speech perception scores that had the highest predictive power on postimplantation outcomes across the adult population.
The resulting models of postimplantation outcomes follow a similar mathematical structure with a patient’s postoperative word score starting at a constant value k , which is either increased by the addition of a term dependent on the pre–cochlear-implant sentence score or decreased by the subtraction of a term dependent on the duration of deafness.
where CI is cochlear implant, Dur Yrs df is duration of deafness in years from onset and % words pre-CI is consonant-nucleus-consonant (CNC) monosyllabic word score before implantation.
In addition to the previously mentioned factors, the choice of which ear to implant has been a frequently discussed issue. Several studies, particularly the Iowa model, have emphasized the utility of implanting the better-hearing ear. At Johns Hopkins Hospital, the authors have advocated the implantation of the poorer-hearing ear. Although greater data are needed, the authors’ studies thus far reveal no significant difference in implant performance based on whether the better- or worse-hearing ear is implanted. Fig. 1 shows a regression plot of the predicted postoperative word scores for each patient as modeled by the Johns Hopkins (implant poorer ear) and Iowa formulas (better ear). There are virtually identical scores predicted from each patient’s duration of deafness and preoperative sentence recognition scores. These data suggest that results obtained through cochlear implantation of the poorer-hearing ear are statistically equivalent to results obtained through implantation of the better-hearing ear. The similarity of results obtained through both methods suggests that implantation may have a beneficial effect on central auditory pathway development regardless of the choice of ear to be implanted, a finding which was later confirmed by Francis and colleagues (2005).
Another variable that influences speech perception is technological sophistication of the implanted device. Improvements in speech perception have been associated with generational improvements in signal processing strategies, speech processors, and electrode arrays, but may reflect clinical trends in patient selection as well as technological advances.
Implant performance in children
The era of pediatric cochlear implantation began with House-3M single-channel implants (a collaboration between House Ear Institute and Minnesota Mining and Manufacturing Company) in 1980. Investigational trials with multiple-channel cochlear implants began with adolescents (aged 10 through 17 years) in 1985 and with children (aged 2 through 9 years) in 1986. Implantation of infants and toddlers younger than 2 years of age began in 1995. Although clinical experience with cochlear implantation is considerably shorter in children than in adults, a large body of evidence is now available (reviewed by ).
Auditory performance assessments for children
Auditory performance in children is assessed with a battery of audiological tests that can address the wide range of perceptual skills exhibited by children with severe to profound sensorineural hearing loss. Although substantial auditory gains are apparent in implanted children, the range of quantifiable improvement varies widely between children and depends heavily on the duration of use of the device as well as preoperative variables. For this reason, testing should survey a range of levels of speech recognition, including simple awareness of sound, pattern perception (discrimination of time and stress differences of utterance), closed-set (multiple choice) speech recognition, and open-set (auditory only) recognition.
Methodological variables must be considered when attempting to objectively rate the effect of cochlear implants on the development of speech perception in children who are deaf. There are obvious difficulties inherent in objectively rating communication competence in very young children; older children may exhibit advantages by virtue of greater familiarity with a test’s context, independent of perceptual skills. Objective assessment also mandates a structured setting. Given its unfamiliarity, children may not be in an optimal frame of mind in cooperating with testing. Investigators must also account for discontinuity in the age-appropriate measures necessary for longitudinal assessment. Kirk and colleagues (1997) have examined the methodological challenges and developmental considerations inherent in pediatric implant assessment and categorized variables relating to
- •
The child’s age and level of language and cognitive development (internal variables)
- •
The child’s ability and willingness to respond as influenced by reinforcement and required memory task (external variables)
- •
The procedure of voice presentation, the test administered, and the available options from which to choose a response (methodological variables)
Tests of speech perceptions typically used for childhood assessment have been described in detail and typically consist of closed-set tests that assess word identification among a limited set of options with auditory cues only, open-set tests (scored by percentage of individual words correctly repeated), and structured interviews of parents using criteria-based surveys to assess the response to sound in everyday situations and behaviors related to spoken communication.
Speech comprehension results in children with cochlear implants
Early assessments of pediatric hearing outcomes were performed by House and colleagues (1983) and showed substantial improvement in auditory thresholds and closed-set speech recognition, albeit with limited open-set speech recognition using early technology (House-3M single-channel implant). In 1994, Miyamoto and colleagues provided systematic, well-controlled assessments of childhood cohorts and consistently demonstrated performance advantages of multichannel over single-channel implants. Other early studies by Fryauf-Bertschy and colleagues (1992), Waltzman and colleagues (1994), Miyamoto and colleagues (1993), and Gantz and colleagues (1994) observed that implanted children gain substantial speech-perception capabilities for 5 years after implantation. Furthermore, the fact that many of the implanted children tracked in these studies were congenitally or prelingually deaf indicates that implantation can provide auditory access during critical developmental stages to form the early correlates of spoken language. More recent auditory outcomes publications reflect advances in implant technology and information processing, yielding ever-improving means in speech recognition results.
Over the past 15 years, a wealth of reports has documented further gains in speech recognition in young children who are deaf using multichannel cochlear implants. Miyamoto and colleagues (1993) noted that in 29 children with 1 to 4 years of experience with a cochlear implant, roughly half achieved open-set speech recognition. Subsequent assessments using greater duration of postimplantation follow-up suggest that this percentage has increased through the rest of the 1990s and 2000s, with open-set speech-recognition scores averaging as high as 80%.
Variability in speech-perception performance across patients is widely recognized. Factors implicated in speech recognition variability include
- •
Amount of residual hearing
- •
Age of implantation
- •
Mode of communication
- •
Family support
- •
Length of deafness
Miyamoto and colleagues (1994) found that the duration of deafness, communication mode, age at onset of deafness, and the processor used accounted for roughly 35% of the variance in closed-set testing, with the length of implant use accounting for the largest percentage of variance in measures of speech perception. O’Donoghue and colleagues (2000) found that age at implantation and mode of communication had a significant effect on speech-perception development in young children after implantation.
Zwolan and colleagues (2004) and Manrique and colleagues (2004) reported improved speech perception in children implanted at younger than 2 years of age compared with children implanted at an older age. Multicenter data reported by Osberger (2002) indicate that implantation performance of children implanted at younger than 2 years of age is significantly better than that of children implanted between 2 and 3 years of age. However, Osberger also identified an important confounding variable that exists in the children who receive a cochlear implant at a younger age: they are more likely to use an oral mode of communication. This finding, by itself, may be a predictor of higher implant performance, which is an observation borne out in early studies of a national childhood cohort assembled by Geers and colleagues (2000).
Osberger (2002) also found that children with more residual hearing were undergoing implantation relative to earlier cohorts. Gantz and colleagues (2000) compiled data from across centers that indicate children with some degree of preoperative open-set speech recognition obtain substantially higher levels of speech comprehension. Taken together, these studies suggest the strongest potential for benefit exists with implantation at a young age, when intervention is provided early and, in the case of a progressive loss, before auditory input is lost completely.
Cheng and colleagues (1999) performed a meta-analysis of relevant literature on speech recognition in children with cochlear implants. Of 1916 reports on cochlear implants published since 1966, 44 provided sufficient patient data to compare speech recognition results between published (n = 1904 children) and unpublished (n = 261) trials. Meta-analysis was complicated by the diversity of tests required to address the full spectrum of speech reception in implanted children. An expanded format of the Speech Perception Categories was designed to integrate results across studies. The main conclusions of this meta-analysis were that earlier implantation is consistently associated with a greater trajectory of gain in speech-recognition performance with an absence of a plateau in speech-recognition benefits over time. More than 75% of the children with cochlear implants reported in peer-reviewed publications have achieved substantial open-set speech recognition after 3 years of implant use.
In an effort to provide the first reference for evaluating postimplant speech recognition in children with cochlear implants, Wang and colleagues (2008) mapped the speech-recognition trajectory of implanted children from baseline up to the 24-month post–cochlear-implant evaluation ( Fig. 2 ). The growth in speech-recognition development over the first 24 months after the implantation was spread widely among children with cochlear implants. A substantial number of children implanted at younger ages demonstrated growth patterns very similar in range or well into the trajectories of the normal-hearing children. A few children implanted at older ages showed slower trajectories of development after implantation. In contrast, the trajectories of speech-recognition development among normal-hearing children showed much less variability, forming a much tighter band of normal development.
Language development in children
The above-mentioned studies have helped to characterize gains in speech recognition. However, the primary goal of implantation in children is to facilitate comprehension and expression through the use of spoken language.
By improving auditory access, cochlear implants augment sound and phrase structure. Although difficult to characterize, benefits in receptive language skills and language production after implantation are the crucial measure by which effectiveness of implants in young children should be assessed. One approach is to compare language performance on standardized tests.
The Reynell Developmental Language Scale evaluates both receptive and expressive skills independently. These scales have been normalized by performance levels of hearing children over an age range of 1 to 8 years and have been used extensively in populations of children who are deaf. Children who are deaf without cochlear implants achieved language competence at half the rate of their normal-hearing peers, whereas implanted patients exhibited language-learning rates that matched, on average, those of their normal-hearing peers (Niparko and colleagues, 2010). In a study of 188 children deafened before 3 years of age assessed language development following cochlear implantation. The average age of implantation in this cohort was approximately 27 months. They found that cochlear implantation is consistently associated with a significant improvement in comprehension and expression of spoken language over the first 3 years of implant use. The development of spoken language was positively associated with younger age at implantation and greater residual hearing before implantation. The rate of improvement in performance on spoken-language measures was less steep in children undergoing cochlear implantation at later ages, with clinical gaps that persist with longitudinal follow-up ( Fig. 3 ). The implication of this study is that cochlear implantation not only improves spoken-language expression and comprehension of children who are severely to profoundly deaf but does so early at a significantly increased rate in infants and toddlers.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
