Abstract
Cochlear implant systems that combine electric and acoustic stimulation in the same ear are now commercially available and the number of patients using these devices is steadily increasing. In particular, electric-acoustic stimulation is an option for patients with severe, high frequency sensorineural hearing impairment. There have been a range of approaches to combining electric stimulation and acoustic hearing in the same ear. To develop a better understanding of fitting practices for devices that combine electric and acoustic stimulation, we conducted a systematic review addressing three clinical questions: what is the range of acoustic hearing in the implanted ear that can be effectively preserved for an electric-acoustic fitting?; what benefits are provided by combining acoustic stimulation with electric stimulation?; and what clinical fitting practices have been developed for devices that combine electric and acoustic stimulation? A search of the literature was conducted and 27 articles that met the strict evaluation criteria adopted for the review were identified for detailed analysis. The range of auditory thresholds in the implanted ear that can be successfully used for an electric-acoustic application is quite broad. The effectiveness of combined electric and acoustic stimulation as compared with electric stimulation alone was consistently demonstrated, highlighting the potential value of preservation and utilization of low frequency hearing in the implanted ear. However, clinical procedures for best fitting of electric-acoustic devices were varied. This clearly identified a need for further investigation of fitting procedures aimed at maximizing outcomes for recipients of electric-acoustic devices.
Introduction
Continuing developments in implant electrode design and improved surgical techniques have resulted in increasing incidence of preservation of residual acoustic hearing in the implanted ear following cochlear implantation (Friedland & Runge-Samuelson, 2009; Talbot & Hartley, 2008). In turn, this has led to an expansion of CI candidature to include patients with significantly more residual hearing, and the development of cochlear implant technology for combined electric and acoustic stimulation (EAS) in the implanted ear. In the EAS or Hybrid application, a cochlear implant electrode array is implanted into the cochlea and electrical stimulation is used to convey high frequency information to the user. This is coupled with a hearing aid in the implanted ear, that is used to convey low frequency information to the user via acoustic stimulation.
To date, the focus of EAS research has been on the degree of postoperative hearing preservation outcomes (Arnoldner et al., 2010; Gantz & Turner, 2003; Gstoettner et al., 2006; James et al., 2006; Kiefer et al., 2005; Lenarz et al., 2006; Mukerjee et al., 2012; Skarzynski, Lorens, Piotrowska, & Skarzynski, 2010; von Ilberg et al., 1999; Woodson, Reiss, Turner, Gfeller, & Gantz, 2009) as well as the perceptual benefits of combined electric-acoustic applications for: speech perception (Büchner et al., 2009; Dorman & Gifford, Dorman, & Brown, 2010; Fraysse et al., 2006; Gantz, Turner, Gfeller, & Lowder, 2005; Helbig & Baumann, 2009; James et al., 2005; Lenarz et al., 2009; Lorens, Polak, Piotrowska, & Skarzynski, 2008; Simpson, McDermott, Dowell, Sucher, & Briggs, 2009; Skarzynski et al., 2012; Turner, Gantz, Karsten, Fowler, & Reiss, 2010); localization (Dunn, Perreau, Gantz, & Tyler, 2010); perception of music (Brockmeier et al., 2010; Dorman, Gifford, Spahr, & McKarns, 2008; Gfeller, Olszewski, Turner, Gantz, & Oleson, 2006; Gfeller et al., 2007; Gifford, Dorman, & Brown, 2010); and functional performance (Driver & Stark, 2010; Gstoettner et al., 2008; Gstoettner et al., 2011; Helbig et al., 2011). In general, results with EAS have been compared with either the cochlear implant used in isolation, or with the preoperative condition with hearing aids.
A variety of EAS fitting approaches have been reported in the literature. EAS outcome studies report utilising a range of practices for both electric and acoustic stimulation. As the number of recipients increases, research into how to best fit EAS devices to individuals with different degrees of postoperative residual hearing is crucial.
Previously, Talbot & Hartley (2008) have reported a descriptive review of the effectiveness of EAS intervention as compared with conventional cochlear implant use with respect to pitch perception. However, that review was not structured to address three critical issues:
what is the range of acoustic hearing in the implanted ear that can be effectively used for an electric-acoustic fitting?;
what benefits are provided by combining acoustic stimulation with electric stimulation?; and
what clinical fitting practices have been developed for devices that combine electric and acoustic stimulation?
The examination of these issues should lead to the development of guidelines and recommendations for fitting EAS devices.
Method
Search Strategy
A systematic review of literature was conducted using methods that followed the guidelines provided by Cox (2005). The initial electronic databases searched included PubMed and MEDLINE. The following key words were entered into the search fields: “Cochlear implant OR cochlear implantation AND acoustic stimulation AND electric stimulation” Additional terms used in subsequent searches: “Cochlear Implantation OR Cochlear Implants AND Electric Stimulation AND Acoustic Stimulation AND Hearing Loss, Sensorineural OR Hearing Loss, High-Frequency. In addition, a textbook by Van de Heyning and Kleine Punte, 2010 was searched by hand to identify references that met the search criteria. Finally, a further electronic search of Web of Science (ISI), SCOPUS - V.4 (Elsevier), CINAHL PLUS (EBSCO), PsycINFO (CSA) databases and the reference lists of the relevant manuscripts were examined for articles that did not appear in prior searches.
Selection Procedure
The search strategy yielded 168 articles. Following a preliminary examination of all 168 article abstracts, 58 were selected for more comprehensive review. The full text articles were retrieved for the 58 articles and were analyzed with regard to the inclusion and exclusion criteria below.
A final 27 met the criteria and were identified for detailed analysis and included in this review.
Inclusion Criteria
Studies were selected for detailed analysis if they met the following inclusion criteria:
publication of the results appeared in a peer-reviewed article or textbook. Excluded repeat publications by the same author/research group using the same subject group and reporting no additional or new evidence.
randomized controlled trials (RCT), nonrandomized controlled trials, cohort studies, case-control studies;
studies that implanted “Hybrid or EAS” candidates with preoperative residual low frequency hearing (Pure tone thresholds ≤ 65 dB HL for 250 Hz and 500Hz) and severe to profound hearing loss in the high frequencies (Pure tone thresholds >70 dB HL at 1500 Hz);and
studies that report on preservation or performance data with a minimum of 6 months postoperative device experience only, to ensure the low frequency air-bone gap and cochlear implant MAP/program have stabilized (Helbig et al., 2011; Lenarz et al., 2009).
articles published between 2000 and 2012
Exclusion Criteria
Studies were excluded from detailed analysis on the following criteria:
expert opinion-based evidence and case studies;
studies that implanted only “traditional” or “conventional” cochlear implant candidates with severe to profound, sensorineural hearing loss;
studies reporting hearing preservation or performance data with less than 6 months postoperative device experience.
Rating of Quality Evidence
Two evidence-based practice (EBP) review rating schemes were used to evaluate the research evidence. The first scheme was the Valente et al. (2006) scale design which ranks studies according to the strength of research evidence. This method was used initially to filter the large number of studies based on: evaluation of the research design used (Level of evidence: 1-6); the quality and relevance of the data (Grade of recommendation: A-D); and the effectiveness (EV: real-world based) and efficacy (EF: laboratory based) of the study. Based on these, an overall strength of recommendation (Strength of evidence: I-III) was made. This process yielded similar strength of evidence for all the studies that met the search criteria. The second scheme was designed by MacDermid (2004). A number of research design principles (subject selection, methodology, endpoints, methods of analysis, results, and conclusions) were used to rate the source of evidence. A total of 24 items are assessed and a total score from (1-48) is calculated. A higher score on the MacDermid scale reflects a higher level of quality of evidence or high internal validity. Whilst a study by Olson and Shinn (2008) showed a strong correlation between the two rating scales used, MacDermid’s critical review worksheet “Evaluation of effectiveness Study design” worksheet and “Evaluation Guidelines for Rating the Quality of an Intervention Study” were used together to qualitatively score and rank the evidence from the final 27 studies in more detail. The evidence was also categorized according to whether assessments were carried out under laboratory/ ideal conditions (Efficacious, EF) or in the real world (Effectiveness, EV).
Results
A total of 27 articles were identified for detailed analysis and inclusion in this review. All articles were published between 2005 and 2012. The review of the studies yielded evidence of a moderate strength (B to D). No meta-analysis or truly randomized controlled trial studies were identified.
Discussion
Preservation of residual hearing for the purpose of combining electric and acoustic stimulation has been reported by several research groups using a variety of standard and research electrode arrays and atraumatic surgical techniques (Friedland & Runge-Samuelson, 2009; Talbot & Hartley, 2008). To determine the postoperative fitting range in the implanted ear that can be effectively aided, these published studies were classified into three sub-groups: firstly, Type I studies that implanted conventional, perimodiolar electrode arrays using the “Advanced Off Stylet” technique and a cochleostomy approach to achieve a full insertion (300°-430°) of array; secondly, Type II studies that implanted standard, full length or medium length electrode arrays with reduced insertion depth (360°), using either a cochleostomy or round window approach; and lastly, Type III studies that implanted shorter and thinner electrodes, such as the “Hybrid or EAS arrays,” that were specifically designed to increase the likelihood of preserving residual hearing, using either a cochleostomy or round window approach. A summary of the 14 studies that met the search inclusion criteria is shown in Table 1.
A Summary of 14 Articles That Provide Information on Postoperative Hearing Preservation Outcomes That Met the Systematic Review Inclusion Criteria.
Abbreviations: AOS= Advanced Off Stylet; CI: Cochlear implant; EAS=Electric-acoustic Stimulation; EC=Electrical complement; EF=Efficacious or laboratory based measures; ES=Electrical stimulation; EV= Effectiveness or real world measures; IE= Implanted ear; HA= Hearing aids; MD score= MacDermid score; NIE= No implanted ear; NH: Normal hearing; PDCI; Partial deafness cochlear implantation; PTA: pure tone eudiometry; PTT: pure tone thresholds; RW= round window; SN= Sensorineural hearing loss; SNR = signal to noise ratio; TL=Total length.
The preservation of residual hearing was reported to range from 70% to 100% of participants across the14 studies with minimum postoperative timeframes of 6 to 12 months. In these studies the term “Hearing preservation” encompassed both complete and partial preservation. As defined in the studies complete preservation referred to the maintenance of hearing thresholds within 10 dB HL of all preoperative thresholds, whereas partial preservation generally referred to a change in thresholds of greater than 10 dB to 15 dB HL of any preoperative threshold level. The degree of postoperative hearing loss and audiometric configuration varied both across individuals within studies, and across studies (see Table 1: Key Findings & Range).
No systematic reporting procedure exists to allow for direct comparisons of postoperative hearing results across various electrode arrays or surgical procedures aimed at improving hearing preservation. In order to compare changes in low frequency acoustic hearing across Type I, II & III studies, the preoperative and postoperative medians at 125 Hz, 250 Hz & 500 Hz were calculated using the raw data from each participant’s audiogram, for six studies out of 14 that provided individual threshold data across frequencies (Gstoettner et al., 2006, 2009; James et al., 2005; Podskarbi-Fayette, Pilka, & Skarzynski, 2010: Skarzynski et al., 2012; Simpson et al., 2009). Thresholds at ≥ 750 Hz were excluded because reporting of thresholds at the octave and intermediate frequencies was highly variable between studies and the preoperative thresholds were already in the severe to profound range. Figure 1 shows the median change in low frequency hearing at 125, 250, and 500Hz. Results from each study are represented by a data point ±SD for the various electrode array studies.
Median change in low frequency thresholds dB HL ± SD for various electrode array studies.
Figure 1 shows that the greatest shift in acoustic hearing thresholds occurred at 500 Hz across all studies. Median changes in postoperative hearing thresholds ranged from 5 dB to 25 dB HL at 125 Hz, 10 dB to 30 dB HL at 250 Hz, and 28 to 43 dB HL at 500 Hz. When examining threshold shifts it is important to also consider a shift in the context of the impact on individual hearing thresholds and whether they are still aidable. Threshold shifts may move postoperative hearing thresholds into a mild to moderate range that may still benefit from amplification or may also result in final thresholds that are in the severe to profound range that may not.
Table 2 summarizes the hearing preservation data for Type I, II & III studies based on six of the 14 studies that reported individual threshold data across frequencies.
Hearing Preservation Data for Type I, II & III Electrode Array Studies.
Note: 1,2,3Super script refers to the relevant reference in the Type row the calculation was based on.
Average changes to thresholds at and below 500Hz (3FA) ranged between approximately 24 dB to 37dB HL and median changes ranged from 18 dB HL to 34 dB HL. In general, better preservation outcomes were achieved in Type III studies that implanted electrode arrays specifically designed to preserve residual hearing for the purpose of electric and acoustic stimulation compared to Type I and II studies that implanted standard electrode arrays either partially or fully inserted. Table 2 shows a lower average threshold shift (3FA) postoperatively for Type III studies as compared to either Type I or Type II. In addition, a higher percentage of participants showed complete preservation (defined in this paper as <10dB HL at any 3 frequencies) or partial preservation (defined as ≥15dB HL at 1 frequency or ≥10dB HL at any 2 adjacent frequencies) of residual hearing for Type III studies as compared to either Type I or Type II . Finally, Type I and Type II studies reported total hearing loss percentages of 17% and 18 % respectively, whereas, Type III array studies report no cases of a participant losing all hearing as a result of surgery. There is a wide range of resulting auditory thresholds in the low frequencies following implantation (see Table 2; minimum and maximum pure tone thresholds). Low frequency thresholds in the implanted ear (125, 250, and 500 Hz) ranged from normal to profound. However, the majority of subjects showed thresholds in the severe to profound range for frequencies of 750 Hz and above. The data are too limited at this time to provide a definitive recommendation regarding the risk of this intervention with regard to the various electrode arrays.
It is important to note that the majority of studies that met the inclusion criteria had relatively small numbers of participants and reported results with minimum postoperative timeframes of 6 to 12 months. Whilst a few studies reported outcomes for longer timeframes, unfortunately, these results were combined with results from much shorter time periods. Examination of the study evidence revealed a high degree of variability in assessment time frames and reporting procedures, therefore, the evidence is insufficient to provide any recommendation regarding the longer term changes in postoperative hearing following this intervention.
In summary, studies investigating preservation of residual acoustic hearing have shown that preservation of low frequency hearing was achieved in the majority of patients, particularly with the use of electric-acoustic arrays. However, there is on average, a reduction in the low frequency acoustic thresholds postoperatively. On an individual level, there is a wide range of resulting auditory thresholds and audiometric configurations, with the majority available for EAS application following implantation. Given this, there is a need for a range of amplification options to accommodate degrees of hearing loss from mild to profound in the implanted ear. The review also emphasizes the need for additional research studies utilizing longer timeframes with minimum of 2 to 5 years postoperatively and larger studies as postoperative hearing preservation is not always stable and changes in postoperative hearing can continue to occur. It is recommended that studies do not combine hearing preservation data from various postoperative timeframes or exclude any enrolled participants. Standard reporting of thresholds at octave and intermediate frequencies, median and mean postoperative threshold changes at all frequencies and preoperative and postoperative audiometric data for both ears for all enrolled participants would enable meta-analysis methods to be used to examine preservation results across studies, across surgical technique and electrode arrays.
A systematic and consistent reporting procedure for all hearing preservation results would also help clinicians in the preoperative counseling stage to provide a clearer understanding of risks of this intervention and assist patients considering this option in making a fully informed choice. An international consensus statement on recommended reporting practices for postoperative hearing preservation and performance data for EAS devices would be clinically useful and relevant.
Following review, 18 studies were selected for detailed analysis to address the second question concerning benefit of combining acoustic and electric stimulation. Findings are summarized in Table 3.
A Summary of 18 Articles That Provide Information on Postoperative Outcomes That Met the Systematic Review Inclusion Criteria.
Abbreviations: APHAB=Abbreviated Profile of Hearing Aid Benefit questionnaire; BTE= Behind the ear; CI: Cochlear implant; dB = decibel; EAS= Electric-acoustic Stimulation; EC=Electrical complement; EF=Efficacious or laboratory based measures; ES=Electrical stimulation; EV= EffectiVeness or real world measures; HA = hearing aids; IE= Implanted ear; ITE=In the ear; HA= Hearing aids; MD score= MacDermid score; NIE = nonimplanted ear; NH= Normal hearing; PDCI=Partial deafness cochlear implantation; PTA= Pure tone Audiometry; PTT=Pure tone thresholds; SN= Sensorineural hearing loss; SNR = signal to noise ratio.
Preoperative Versus Postoperative Speech Perception
To evaluate the speech perception benefits resulting from combining acoustic amplification with electric stimulation, preoperative speech perception scores obtained using hearing aids (HAs) were compared to postoperative scores in the implanted ear using electric stimulation alone (CI alone) and combining acoustic and electric stimulation (CI +HAi). Results consistently show that speech perception, on average, was significantly higher postoperatively with electric stimulation alone (CI) when compared to best aided preoperative condition, and for combined acoustic and electric stimulation in the implanted ear (CI+HAi) as compared to scores for electric stimulation alone (CI). These findings are shown in Figure 2 for the various electrode arrays studies. In all studies, speech was presented from a single frontal loudspeaker (0° azimuth) at a fixed distance from the participant. Word perception scores from listening in quiet are shown, with averaged results from each study calculated and represented by a data point.
Preoperative versus postoperative word scores in quiet (Mean ± SD) in implanted ear for various array studies.
Figure 2 shows speech performance in the preoperative best-aided condition (HAs) compared with postoperative monaural listening scores obtained for the implanted ear using electric stimulation alone (CI only) represented as the closed symbols. Any increase in measured performance is assumed to represent primarily benefit as a direct result from implantation. The graph shows that all studies, regardless of array used, reported significantly better mean word perception scores for postoperative electric stimulation alone (CI only) as compared to preoperatively (HAs) in quiet. The increase in word scores ranged from 30% to 39% across the various electrode studies. It could be argued, that testing this condition provides evidence to support this intervention for individuals with severe high frequency sensorineural hearing losses, even when there is complete hearing loss associated with the implantation.
Figure 2 also compares speech performance in the preoperative best aided condition (HAs) as compared with the unilateral listening condition in the implanted ear using combined acoustic and electric stimulation (CI+HAi). Data was taken from the same studies and represented as open symbols. Any increase in measured post implantation performance represents primarily benefit from use of preserved low frequency residual hearing through combining acoustic and electric stimulation following cochlear implantation. On average, word scores in quiet we significantly better postoperatively in the implanted ear combining acoustic and electric stimulation (CI+HAi) as compared to preoperatively (HAs). Postoperative EAS benefit ranged from 43.8% to 49% in word scores. The post implantation speech results show that there is benefit combining acoustic and electric stimulation.
Across all studies eligible for inclusion, results consistently show that speech perception, on average, was enhanced postoperatively with electric stimulation alone compared to preoperative condition. Greater postoperative advantage was evident when combining acoustic and electric stimulation in the implanted ear than for use of electric stimulation alone. It should also be noted that considerable individual variability was reported across these studies, with the majority of individuals showing benefit, whereas a small number showed no difference in preoperative versus postoperative performance. Age and duration of hearing loss accounted for significant variance and had a negative impact on speech performance (Gantz et al., 2009; Lenarz et al., 2009). A significant correlation was also found between preoperative articulation abilities and postoperative CI alone Speech Reception Threshold (SRT) sentence scores (Buchener et al., 2009).
Acoustic and Electric Stimulation Speech Perception Benefit in the Implanted Ear
A further question is the degree of benefit that can be gained from combining acoustic and electric stimulation when residual hearing is preserved following implantation? To investigate this, several studies have compared speech perception in the implanted ear using electric stimulation only (CI) with speech perception using combined acoustic and electric stimulation (CI+HAi) in quiet (Figure 3a) and in noise (Figure 3b). Speech perception scores for testing in quiet and noise are shown, with averaged results from each study calculated and represented by a data point.
EAS benefit for implanted ear in Quiet: Electric only stimulation (CI) versus Acoustic & Electric Stimulation (CI+HAi) word scores (Mean ± SD) for various electrode array studies.
EAS benefit for implanted ear in Noise: Electric only stimulation (CI) versus Acoustic & Electric Stimulation (CI+HAi) scores (Mean ± SD) for various electrode array studies.
As shown, speech perception scores for the implanted ear were consistently higher for the combined acoustic and electric stimulation as compared to electric stimulation alone, particularly for speech perception tests in noise. Figure 3a shows that in quiet, benefit ranged from 6% to 15% for word perception scores. Five out of the nine electrode array studies reported statistically significant benefit on average. Figure 3b shows that in noise (+10dB SNR) this benefit ranged from 4% to 29% for sentence perception scores. Five out of six studies reported statistically significant benefit, on average. The one study that did not reach statistical significance had reached ceiling affect.
Across all studies, speech perception scores were higher postoperatively in the implanted ear when using combining acoustic and electric stimulation (CI+HAi) as compared to electric stimulation alone (CI), on average, in both quiet and noise. The addition of acoustic stimulation potentially provides low frequency information to enhance the perception of F0 voicing cues (timing and presence), envelope periodicity cues, low-frequency segmental phonetic cues (nasal resonant, F1 onset and shape of transitions) for improved speech perception (Brown & Bacon, 2009; Kong & Carlyon, 2007; Qin & Oxenham, 2006; Zhang, Dorman, & Spahr, 2010).
Binaural Hearing Benefits
Preservation of low frequency hearing in the implanted ear also provides potential benefits through enabling provision of acoustic stimulation to both ears. Five out of the 18 inclusion studies examined the impact of utilizing both ears for binaural benefit in speech perception and localization. Three studies that evaluated word performance in quiet (Gantz et al., 2005; Lorens et al., 2008; Simpson et al., 2009) reported higher speech perception scores in the binaural condition (CI +HAi + HAc opposite ear) compared to the monaural condition (CI +HAi) on average, however, the studies did not reach statistical significance. Two studies evaluated sentence perception using an adaptive test with spatially separated target speech and noise sources (Dunn et al., 2010; Gifford et al., 2010). Binaural hearing did however provide a statistically significant benefit for sentence perception in spatially separated speech and noise sources. The binaural improvement for speech reception thresholds (SRTs) was about 3dB in both studies. Dunn and colleagues (2010) proposed that the potential advantage lies in the utilization of two similar acoustic signals with identical sound processing being presented to the central auditory system ear. One study to date that reported on horizontal localization found a significant difference in localization performance for the binaural condition compared to the monaural or bimodal condition (Dunn et al., 2010). The author concluded that the bilateral hearing aids providing similar processing from both ears improved the perception of audible low frequency input below 800 Hz, where primarily interaural time level differences (ITD) cues help to localize the sound source. The ability to localize the sounds was dependent on participants being able to perceive sounds in both ears with the acoustic amplification provided by the hearing aids, and independent of whether there was electrical stimulation or not.
Only a small number of studies have utilized test arrangements to evaluate binaural benefit derived through the provision of acoustic stimulation to both ears. The review highlights the need for the inclusion of more studies investigating and using outcomes measures that better quantify binaural advantages such as; spatially separated speech and noise, localization, and also functional benefit in everyday life or Effective (EV) evidence, rather than traditional outcome measures utilized in standard cochlear implantation.
EAS outcome studies report utilizing a range of fitting practices for both the acoustic and electric stimulation. A summary of the 16 studies selected to address the third clinical question are shown in Tables 4.
A Summary of 16 EAS Articles That Provide Information on Fitting Practices for Devices That Combine Electric and Acoustic Stimulation.
Abbreviations: ACE= Advanced combination encoder; BTE= Behind the ear; CI: Cochlear implant; dB = decibel; CIS= Continuous interleaved sampling DSL = Desired Sensation Level; DSLi/o = Desired Sensation Level Input/Output; EF=Efficacious or laboratory based measures; EV= EffectiVeness or real world measures; HA = hearing aids; IE= Implanted ear; ITE=In the ear; HA= Hearing aids; MD= MacDermid score;MP= Monopolar stimulation; NAL RP = National Acoustic Laboratories-Revised Profound; NAL-NL1 = National acoustic laboratories-Non-linear 1; NIE = nonimplanted ear; NH= Normal hearing; PDCI=Partial deafness cochlear implantation; PTA= Pure tone Audiometry; PTT=Pure tone thresholds; SN= Sensorineural hearing loss; SNR = signal to noise ratio; PPS=Pulses per second.
Low Frequency Acoustic Amplification for Combined Acoustic and Electric Stimulation
The provision of acoustic amplification to the residual low frequency hearing preserved in the implanted ear in EAS studies has been either via an independently operating, commercially available digital in-the-ear hearing aid worn together with a behind-the-ear cochlear implant sound processor, or more recently, via EAS sound processors that deliver both acoustic and electrical stimulation through one integrated device. A wide range of conventional hearing aid prescriptive rules have been used in the studies to determine the amplification applied to the low frequency hearing in the implanted ear. Procedures for linear and nonlinear amplification such as: Half gain rule (Kiefer et al., 2005; Helbig et al., 2011), NAL-RP (Gantz & Turner, 2009; Dunn et al., 2010), DSL [I/O] (James et al., 2005; Fraysse et al., 2006) and NAL-NL1 (Karsten et al., 2012, Simpson et al., 2009) have been applied. The majority of studies apply a variation to these standard hearing aid fitting rationales for the high frequency region of the audiogram. In EAS device fittings the high frequency information is provided via electrical stimulation via the cochlear implant part, therefore, acoustic amplification is only provided up to the point where there is useful residual hearing or cut off where hearing is not measurable. This frequency up to which acoustic amplification is provided ranged from the frequencies where the hearing threshold exceeded 60 dB HL (Helbig et al., 2009; Nopp & Polak, 2010); exceeded 80 dB HL (Fraysse et al., 2006; Gstoettner et al., 2008; James et al., 2005; Lenarz et al., 2009) or 85 dB HL (Gantz et al., 2009; Simpson et al., 2009) on the individual’s audiogram. Practical considerations such as prevention of feedback and off- frequency listening, providing sufficient loudness, improving mould or loudness comfort and prolonging battery life, have also been reported as the reason for applying variation to the standard fitting rationale.
Only two studies reported on the broadband amplification applied to the contralateral ear for subjects in their study (Dunn et al., 2010; Gantz et al., 2009). Hearing aid verification using real ear probe measures to obtain objective measures of hearing aid fitting were reported in only two studies (Dunn et al., 2010; Vermeire, Anderson, Flynn, & Van de Heyning, 2008). Four studies report performing loudness balancing tasks between the electric and acoustic signal in the implanted ear. Loudness rating scales were used and adjustments to gain of hearing aid were made in response to individual feedback to live voice recorded speech or noise signals at different input levels (Karsten et al., 2012; Kiefer et al., 2005; Simpson et al., 2009). One study reported using a loudness scaling test performed at 500 Hz with the CI component deactivated to determine the setting of the AGC threshold (Helbig et al., 2011). The majority of the studies examined did not provide any information on whether a loudness balancing procedure between the electric and acoustic signal was used.
To date, two studies have evaluated the influence of different acoustic fitting parameters in the implanted ear on EAS outcomes (Vermeire et al., 2008; Polak, Lorens, Helbig, McDonald, & Vermeire, 2010). A small study by Vermeire and colleagues (2008) investigated the effect of two parameters; acoustic amplification (gain) and frequency range of amplification on speech performance, and subjective measure of benefit in 4 adults with a MED-EL M- electrode array and ITE hearing aid in the same ear. Results showed that additional low frequency gain (+6dB SPL in the low frequency range) and providing a wider frequency range for amplification (amplification up to thresholds of 120dB HL on the audiogram) provided the highest mean speech scores. Unfortunately, small participant numbers and limited listening experience of 30 min with each condition limit the strength of evidence and therefore the ability to support definitive conclusions. A second study by Polak et al. (2010) evaluated compression, compression threshold, and low frequency slope in the implanted ear (nonimplanted ear plugged) in 24 EAS participants. The authors reported that whilst all the parameters tested had an effect on speech perception in quiet and noise in the implanted ear, the parameters that had the most impact on overall benefit was the compression ratio and the frequency where electrical stimulation started. No statistical analysis of the results was reported to evaluate significance. Limited listening experience with each condition (2 hr or 1 day) also limits the strength of evidence from this study. Further investigation is required. A summary of hearing aid parameters used in EAS fittings is shown in Table 3.
Discontinuation of Low Frequency Acoustic Amplification Postoperatively
A number of studies report discontinuation or rejection of amplification in the implanted ear by some individuals. Typically, they can be divided into three groups. The first group consisted of individuals with a severe to profound hearing loss in the implanted ear (typically >80 dB HL across the frequency range) (Gantz et al., 2009; Gstoettner et al., 2008; Helbig et al., 2011; Simpson et al., 2009; Skarzynski et al., 2010) and without sufficient useful hearing for combined electric acoustic stimulation. It is not clear whether this is due to limited benefit or the “perceived” limited benefit by the recipient or as a result of practical issues concerning physical or loudness discomfort around fitting such losses, the inconvenience of wearing an additional component or a combination of factors. In addition Helbig et al. (2011) reported that “amplification was often reported to be insufficient, especially in the frequencies below 500Hz, which made fitting to provide sufficient gain very difficult.” A study by Helbig and Baumann (2009) examined the relationship between the postoperative pure tone audiometric thresholds in the implanted ear and wearer acceptance of a combined DUET EAS sound processor in 15 individuals. The study found that the 11 participants continued to use combined electric and acoustic stimulation, had residual hearing that was <75dB HL in the 500Hz frequency region or below. The remaining four participants who rejected acoustic amplification, had a hearing loss of greater than 55dB HL at 125 Hz, 70 dB HL at 250 Hz and 98 dB HL at 500Hz. The authors deduced that acceptance of acoustic amplification was dependent on postoperative hearing thresholds.
The second group that did not use amplification in the implanted ear consists of individual with significant amounts of residual hearing between 125-500Hz and PTA within the normal to mild range (Gantz et al., 2009; Lenarz et al., 2009; Skarzynski, Lorens, Piotrowska, & Anderson, 2007; Skarzynski & Lorens, 2010). This group did not require amplification as they could combine their natural acoustic hearing with electric stimulation. Skarzynski et al. (2010) proposed a classification system where electrical stimulation alone should be used in the implanted ear if low frequency hearing (125 Hz - 500 Hz) is ≤ 30 dB HL or ≥ 80 dB HL, and acoustic amplification considered in the implanted ear for all others.
The final group consists of individuals that report management and wearing of a supplementary acoustic instrument was cumbersome, inconvenient or contraindicated due to issues of the external ear canal such as acute external otitis media and was the reason to reject the use of the hearing aid (Gstoettner et al., 2008; Helbig, Baumann, Helbig, von Malsen-Waldkirch, & Gstoettner, 2008; Helbig et al., 2011) despite some having sufficient residual hearing to utilize and potentially benefit from ipsilateral electric and acoustic stimulation. Whilst the introduction of less cumbersome EAS sound processors which deliver both acoustic and electrical stimulation in one integrated device has helped reduce some of these wearing issues, there are still a number of participants that have discontinued amplification (Helbig et al., 2008; Helbig & Baumann, 2009). Further systematic investigation of fitting procedures aimed at maximizing outcomes for recipients of electric-acoustic devices could potentially facilitate the continued use of acoustic amplification following implantation, reduce individual variability with regard EAS benefit, increase recipient outcomes and wearer satisfaction.
Cochlear Implant Electric Fitting for Combined Acoustic and Electric Stimulation
To date, the electric parameters selected for electric stimulation in EAS applications are generally default settings determined by the cochlear implant system used. The default parameters utilized in studies include sound coding strategies, stimulation rates, number of channels. A summary of electric parameters used in EAS fittings is shown in Table 3.
One electric fitting parameter that has received a considerable amount of attention is the electric-to-acoustic frequency boundary allocation, also known as the crossover frequency. The crossover frequency refers to the way in which frequencies in the input signal are selected and mapped for acoustic stimulation and for electric stimulation in the implanted ear (the frequency at which stimulation changes from acoustic to electrical). There is potential to provide overlapping inputs for low frequencies with the CI sound processor via electrical stimulation (200 Hz - 1000Hz) and the residual low frequency acoustic hearing (0 Hz -1000 Hz). Therefore, the same acoustic frequency can be represented in different cochlea locations via two very different modalities within the implanted ear. A number of studies attempted to quantify how much overlapping, if any, of the frequency range would provide the best outcomes for speech perception. The majority of published evidence that suggests that limiting or completely restricting the range of incoming stimulus frequencies that could be represented both acoustically and electrically provided potential speech perception benefits and was subjectively preferred by the majority of listeners. There was also a trend for better speech perception scores in noise when a restricted electrical bandwidth was used in the CI program. (Fraysse et al., 2006; Helbig et al., 2011; James et al., 2005; Karsten et al., 2012; Nopp & Polak, 2010: Simpson et al., 2009; Vermeire et al., 2008). The minority of listeners reported in the studies that preferred full electrical frequency ranges and overlapping acoustic and electric inputs, typically had more severe to profound losses or fluctuating losses compared to the steeply sloping losses (Fraysse,et al., 2006; Gantz et al., 2009; Vermeire et al., 2008; Nopp & Polak, 2010). While listeners with normal to mild hearing thresholds in the low frequency region reported “echo” and “distorted” sound quality with full unrestricted electric or overlapping acoustic and electric programs in a few studies (Helbig & Baumann, 2009; Karsten et al., 2012) .
The general approach for fitting these devices has been based on the principle that electric stimulation by the CI processor should begin at the point where acoustic hearing is no longer beneficial. The method used to determine this input frequency range however, has varied widely in studies reviewed. The majority of studies report crossover frequency selection being based on the individual’s unaided audiogram and where the dead regions of the cochlea might limit benefit. In these studies the selection of the crossover frequency based on the audiogram can be grouped into two: the frequency at which the individual’s thresholds are ≥65 dB HL on the audiogram (Helbig et al., 2011; Nopp & Polak, 2010) or ≥ 80 dB HL on the audiogram (Fraysse et al., 2006; Gantz et al., 2009; Gstoettner et al., 2008; James et al., 2005; Lenarz et al., 2009; Simpson et al., 2009). In addition, Greenwood’s frequency-to-place map calculations (Kiefer et al., 2005) and individually optimized frequency derived from psychophysical measurements (Baumann, Rader, Helbig, & Bahmer, 2011; Simpson et al., 2009) methods have also been reported. The findings do not provide a clear understanding of where to set the restricted electric bandwidth range based on an individual’s hearing. All studies have used sentence or monosyllabic word recognition in quiet and noise as outcome measurements. Although several studies include comments on participant’s subjective preference, only two study (Karsten et al., 2012; Vermeire et al., 2008) used a qualitative rating scales such as the visual analogue scale (VAS) to rate “ease of listening” effort during testing or in everyday life. The question of whether systematic adjustment of electric and acoustic output levels and parameters improved music perception, localization or functional performance in real life usage outcome, remains unclear.
Pediatric Fitting for Combined Electric and Acoustic Stimulation
Finally, two studies by Skarzynski and colleagues (2007), and Skarzynski & Lorens (2010) describe fitting procedures for EAS devices in the pediatric population. Skarzynski et al. (2007) described using a standard fitting procedure developed for adults when fitting children. No special consideration had to be made and there were no specific issues relating to these fittings. The author stated that “children readily accepted the combination of electric and acoustic information.”
Whilst there is some evidence that suggests a systematic fine-tuning of the electric and acoustic stimulation may be beneficial, further research is needed to investigate fitting procedures to maximize outcomes for adults and children with different degrees of residual hearing who use EAS implant systems.
Conclusion
The preservation of low frequency residual hearing following cochlear implantation provides the capacity for combining electric and acoustic stimulation in the implanted ear to benefit the user. This emphasizes the importance of continued development of electrode design and surgical techniques aimed at preserving residual low frequency hearing in the implanted ear. This knowledge also highlights the need for systematic reporting procedure for all postoperative hearing preservation studies and underscores the importance of appropriate preoperative counseling to ensure realistic expectations of potential benefits versus and risks.
The EAS research evidence to date has also shown that benefit from combining electric and acoustic stimulation in both the monaural and binaural condition can be consistently demonstrated over electric alone condition. This potential EAS advantage highlights the value of the endeavors to preserve low frequency hearing in the implanted ear and to continue amplification where appropriate. Consideration for the inclusion of additional outcome measures that help effectively evaluate the benefit of the additional acoustic low frequency hearing in both ears for speech perception, localization and functional performance everyday life is crucial.
A wide variety of acoustic and electric fitting parameters and methods have been used and the individual variations applied to amplification of the high frequencies. Further research is also needed to investigate fitting procedures to maximize outcomes for recipients with different degrees of hearing loss who use devices that combine electric and acoustic stimulation. Ensuring that the electric and acoustic stimulation devices are fitted optimally through empirically validated procedures and incorporating current knowledge of potential binaural benefit into a comprehensive individual rehabilitation program will facilitate their continued use following implantation and maximize outcomes.
Footnotes
Acknowledgements
The authors are grateful for the financial support of the Commonwealth of Australia through the HEARing Cooperative Research Centre, an Australian Government initiative.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the authorship and publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research and authorship of this article: This work is supported by the Commonwealth of Australia through the HEARing Cooperative Research Centre.