Predictive value of automated cognivue cognitive assessment for cochlear implant outcomes – a preliminary study in a private otolaryngology practice

1) To complete a follow-up investigation of postoperative outcomes for adult cochlear implant (CI) recipients scoring ≥30% Consonant-Nucleus-Consonant (CNC) preoperatively, and 2) to describe the postoperative performance trajectory for this group of higher performing patients. Retrospective chart review. Tertiary referral center. One hundred four (105 ears) postlingually deafened adults who scored ≥30% CNC word recognition in the ear to be implanted preoperatively. One hundred four subjects underwent cochlear implantation. Pre- and postoperative CNC word scores and AzBio sentences in quiet and noise in the ear to be implanted as well as the bilateral-aided condition pre-CI and at 1, 3, 6, and 12 months post-CI. Statistically significant improvement was demonstrated for CNC and AzBio sentences in quiet and noise for the CI alone and bilateral listening conditions. Most improvement was demonstrated by 6-months postoperatively (p < 0.001) with the exception of AzBio sentences in noise demonstrating improvement within 3 months (p < 0.001). For patients with preop CNC scores up to 40% (n = 57), all recipients demonstrated either equivocal (n = 17) or statistically significant improvement (n = 40) for CNC word recognition in the CI-alone condition and none demonstrated a significant decrement in the bilateral condition. For patients with preop CNC scores >40% (n = 47, 48 ears), 89.3% (42 patients) demonstrated either equivocal (n = 24, 50%) or statistically significant improvement (n = 19, 39.6%) for CNC word recognition in the CI-only condition and none demonstrated a significant decrement in the bilateral condition. CI candidates with preoperative CNC word scores higher than conventional CI recipients derive statistically significant benefit from cochlear implantation for both the CI ear and best-aided condition. These data provide further support for the expansion of adult CI candidacy up to at least 40% CNC word recognition preoperatively with consideration given to further expansion possibly up to 60%.

(1) To determine the effect of hearing aid (HA) bandwidth on bimodal speech perception in a group of unilateral cochlear implant (CI) patients with diverse degrees and configurations of hearing loss in the nonimplanted ear, (2) to determine whether there are demographic and audiometric characteristics that would help to determine the appropriate HA bandwidth for a bimodal patient. Participants were 33 experienced bimodal device users with postlingual hearing loss. Twenty three of them had better speech perception with the CI than the HA (CI>HA group) and 10 had better speech perception with the HA than the CI (HA>CI group). Word recognition in sentences (AzBio sentences at +10 dB signal to noise ratio presented at 0° azimuth) and in isolation [CNC (consonant-nucleus-consonant) words] was measured in unimodal conditions [CI alone or HAWB, which indicates HA alone in the wideband (WB) condition] and in bimodal conditions (BMWB, BM2k, BM1k, and BM500) as the bandwidth of an actual HA was reduced from WB to 2 kHz, 1 kHz, and 500 Hz. Linear mixed-effect modeling was used to quantify the relationship between speech recognition and listening condition and to assess how audiometric or demographic covariates might influence this relationship in each group. For the CI>HA group, AzBio scores were significantly higher (on average) in all bimodal conditions than in the best unimodal condition (CI alone) and were highest at the BMWB condition. For CNC scores, on the other hand, there was no significant improvement over the CI-alone condition in any of the bimodal conditions. The opposite pattern was observed in the HA>CI group. CNC word scores were significantly higher in the BM2k and BMWB conditions than in the best unimodal condition (HAWB), but none of the bimodal conditions were significantly better than the best unimodal condition for AzBio sentences (and some of the restricted bandwidth conditions were actually worse). Demographic covariates did not interact significantly with bimodal outcomes, but some of the audiometric variables did. For CI>HA participants with a flatter audiometric configuration and better mid-frequency hearing, bimodal AzBio scores were significantly higher than the CI-alone score with the WB setting (BMWB) but not with other bandwidths. In contrast, CI>HA participants with more steeply sloping hearing loss and poorer mid-frequency thresholds (≥82.5 dB) had significantly higher bimodal AzBio scores in all bimodal conditions, and the BMWB did not differ significantly from the restricted bandwidth conditions. HA>CI participants with mild low-frequency hearing loss showed the highest levels of bimodal improvement over the best unimodal condition on CNC words. They were also less affected by HA bandwidth reduction compared with HA>CI participants with poorer low-frequency thresholds. The pattern of bimodal performance as a function of the HA bandwidth was found to be consistent with the degree and configuration of hearing loss for both patients with CI>HA performance and for those with HA>CI performance. Our results support fitting the HA for all bimodal patients with the widest bandwidth consistent with effective audibility.

Cochlear implantation is highly effective at improving hearing outcomes, but results have been limited to groupwise analysis. That is, limited data are available for individual patients that report comparisons of preoperative aided speech recognition and postimplantation speech recognition. To assess changes in preoperative aided vs postoperative speech recognition scores for individual patients receiving cochlear implants when considering the measurement error for each speech recognition test. This cross-sectional study used a prospectively maintained database of patients who received cochlear implants between January 1, 2012, and December 31, 2017, at a tertiary, university-based referral center. Adults with bilateral sensorineural hearing loss undergoing cochlear implantation with 6- or 12-month postoperative measures using 1 or more speech recognition tests were studied. Cochlear implantation. Postoperative word recognition (consonant-nucleus-consonant word test), sentence recognition (AzBio sentences in quiet), and sentence recognition in noise (AzBio sentences in +10-dB signal-to-noise ratio) scores, and association of each speech recognition score change with aided preoperative score to each test's measurement error. Analysis of data from a total of 470 implants from 323 patients included 253 male (53.8%) patients; the mean (SD) age was 61.2 (18.3) years. Most patients had statistically significant improvement in all speech recognition tests postoperatively beyond measurement error, including 262 (84.8%) for word recognition, 226 (87.6%) for sentence recognition, and 33 (78.6%) for sentence recognition in noise. A small number of patients had equivalent preoperative and postoperative scores, including 45 (14.5%) for word recognition, 28 (10.9%) for sentence recognition, and 9 (21.4%) for sentence recognition in noise. Four patients (1.6%) had significantly poorer scores in sentence recognition after implantation. The associations between age at implantation and change in speech recognition scores were -0.12 (95% CI, -0.23 to -0.01) for word recognition, -0.22 (95% CI, -0.34 to -0.10) for sentence recognition, and -0.10 (95% CI, -0.39 to 0.21) for sentence recognition in noise. Patients with no significant improvement were similarly distributed between all preoperative aided speech scores for word recognition (range, 0%-58%) and sentence recognition (range, 0%-56%) testing. In this cross-sectional study, with respect to preoperative aided speech recognition, postoperative cochlear implant outcomes for individual patients were largely encouraging. However, improvements in scores for individual patients remained highly variable, which may not be adequately represented in groupwise analyses and reporting of mean scores. Presenting individual patient data from a large sample of individuals with cochlear implants provides a better understanding of individual differences in speech recognition outcomes and contributes to more complete interpretations of successful outcomes after cochlear implantation.

Assess relationships between patient, hearing, and cochlear implant (CI)-related factors and second-side CI speech recognition outcomes in adults who are bilaterally implanted. Retrospective review of a prospectively maintained CI database. Tertiary academic center. One hundred two adults receiving bilateral sequential or simultaneous CIs. Postimplantation consonant-nucleus-consonant (CNC) word and AzBio sentence scores at ≥12 months. Of patient, hearing and CI-specific, factors examined only postimplantation speech recognition scores of the first CI were independently associated with speech recognition performance of the second CI on multivariable regression analysis (CNC: ß = 0.471[0.298, 0.644]; AzBio: ß = 0.602[0.417, 0.769]). First-side postoperative CNC scores explained 24.3% of variation in second CI postoperative CNC scores, while change in first CI AzBio scores explained 40.3% of variation in second CI AzBio scores. Based on established 95% confidence intervals, 75.2% (CNC) and 65.9% (AzBio) of patients score equivalent or better with their second CI compared to first CI performance. Age at implantation, duration of hearing loss, receiving simultaneous versus sequential CIs, and preoperative residual hearing (measured by pure-tone average and aided speech recognition scores) were not associated with 12 month speech recognition scores at 12 months. The degree of improvement in speech recognition from first CI may predict speech recognition with a second CI. This provides preliminary evidence-based expectations for patients considering a second CI. Counseling should be guarded given the remaining unexplained variability in outcomes. Nonetheless, these data may assist decision making when considering a second CI versus continued use of a hearing aid for an unimplanted ear. III.

Introduction: The use of cochlear implantation to rehabilitate moderate to profound sensorineural hearing loss has become more widespread; however, the adult utilization rate of cochlear implant candidates is still very less. The study aims to examine the percentage of adult patients in a heterogeneous group of cochlear implant recipients at a nascent cochlear implant program who demonstrate improvements in speech outcomes.Methods: Speech outcome scores were assessed preoperatively and postoperatively at three, six, and 12-month intervals using consonant-nucleus-consonant (CNC) words and AzBio sentences in quiet. Mean speech outcome scores at each time point and binomial distribution tables with 95% CI were used to assess individual improvement in speech understanding.Results: 45 patients underwent a total of 49 cochlear implantation surgeries. The mean age at surgery was 62 years. The mean preoperative CNC score in the ear to be implanted was 18%±18, while the mean postoperative CNC score at three, six, and 12 months was 35%±21, 44%±23, and 45%±25, respectively. The mean preoperative AzBio score in the ear to be implanted was 22%±26 while the mean postoperative AzBio score at three, six, and 12 months was 50%±29, 56%±27, and 63%±26, respectively. Of the implantations, 74% (32 of 43) and 69% (22 of 32) showed significant improvement at six months or one year using AzBio and CNC binomial distribution tables, respectively.Conclusions: Findings demonstrate significant improvements in speech perception following cochlear implantation for patients not benefiting from hearing aid aural rehabilitation. The study provides realistic expectations for new and emerging programs hoping to demonstrate cochlear implant utility for improving patients’ speech outcomes.

Determine whether asymmetric hearing loss (AHL) affects postoperative speech outcomes in cochlear implant (CI) patients. Retrospective cohort study. Tertiary care hospital. Adult English-speaking patients with unilateral CIs implanted between 2014 and 2018 were stratified into NonAHL and AHL groups based on preoperative AzBio scores in quiet from the nonimplanted ear (0-50% vs. 51-100%, respectively). CI surgery in the poorer performing ear. Postoperative consonant-nucleusconsonant (CNC) word and AzBio sentence test scores in quiet and/or noise at +5 dB signal-to-noise ratio (SNR). Of 512 patients, 33 non-AHL and 27 AHL patients were included. Average ages were 65.6 and 63.6 years, respectively. As expected, preoperative AzBio scores in quiet from the nonimplanted ear were higher in the AHL group (95% confidence interval [95%CI]: 66.4-76.4%) than the non-AHL group at baseline (95%CI: 12.3-23.6%). In both cohorts, AzBio scores in quiet from the implanted ear improved from baseline, with 24-month scores (95%CI: 73.8 - 84.9%) being higher than preoperative scores (95%CI: 13.2-23.1%). There were also significant differences in AzBio scores in quiet between cohorts overall (p = 0.0120) on mixed model analysis, with the AHL group performing ∼6.4% better than the non-AHL group; however, differences were not significant when scores were stratified by time. In addition, there were no significant differences in CNC in quiet and AzBio scores in noise at +5 dB SNR between cohorts (p = 0.1786 and p = 0.6215, respectively). After CI, patients with AHL can achieve scores on word and sentence tests at least comparable to traditional CI candidates, supporting the expansion of CI candidacy to include patients with AHL.

Objectives:Machine learning (ML) is an emerging discipline centered around complex pattern matching and large data-based prediction modeling and can improve precision medicine healthcare. Cochlear implants (CI) are highly effective, however, outcomes vary widely, and accurately predicting speech perception performance outcomes between patients remains a challenge. This study aims to evaluate the ability of ML to predict speech perception performance among CI recipients at 6-month post-implantation using only preoperative variables on one of the largest CI datasets to date, with an emphasis placed on identification of poor performers.Design:All patients enrolled in the national CI outcome tracking database, HERMES, and the institutional CI registry. Data were split 90/10 training/testing with hyperparameter tuning designed to optimize AUPRC performed during 10-fold cross-validation within 100 iterations. Multiple models were developed to predict final and delta (Δ) in consonant-nucleus-consonant (CNC) words and AzBio sentences at 6-month post-implantation. Two metrics, (1) final performance scores and (2) equally distributed 20th percentile performance ranking were used as primary outcomes. All models were compared with currently used “gold standard,” defined as linear or logistic regression models leveraging Lazard features (LF). Final metrics for comparison included mean absolute error (MAE), calibration curves, heat accuracy maps, area under the receiver operating curve (AUROC), and F1 score.Results:A total of 1877 patients were assessed through an ML pipeline. (1) XGBoost (XGB) predicted CNC with MAE of 17.4% (95% confidence interval [CI]: 17.34 to 17.53%) and AzBio with MAE of 20.39% (95% CI: 20.28 to 20.50%) and consistently outperformed linear regression with LF (CNC MAE 18.36% [95% CI: 18.25 to 18.47]; AzBio 21.62 [95% CI: 21.49 to 21.74]). Although statistically significant, the 1 to 2% boost of performance is clinically insignificant. (2) Predicting quintiles/20th percentile categories for CI performance, XGB outperformed logistic regression (Log-LF) across all metrics. XGB demonstrated superior calibration compared with Log-LF and provided a larger proportion of predicted probabilities predictions at the extremes (e.g., 0.1 or 0.9). XGB outperformed Log-LF predicting ≤40th percentile for CNC (AUROC: 0.708 versus 0.594; precision: 0.708 versus 0.596; F1 score: 0.708 versus 0.592) and AzBio (AUROC: 0.709 versus 0.572; precision: 0.710 versus 0.572; F1 score: 0.709 versus 0.572). This was consistent for ΔCNC and ΔAzBio. Last, accuracy heat maps demonstrated superior performance of XGB in stratifying sub-phenotypes/categories of CI performance compared with Log-LF.Conclusions:This study demonstrates how ML models can offer superior performance in CI speech perception outcomes prediction modeling compared with current gold standard (Lazard—linear or logistic regression). ML offers novel insights capable of capturing nonlinear complex relationships and can identify novel sub-phenotypes at the extremes of CI performance using preoperative clinical variables alone. This is the first study to our knowledge to offer any type of meaningful preoperative stratification for CI speech perception outcomes and may have significant implications that need to be carefully explored when it comes to patient counseling, auditory rehabilitation, and future CI clinical trials. While prospective validation is a necessary next step and performance is still limited based on current traditional CI variables, these results highlight the potential of artificial intelligence (AI) in CI care, the critical need to integrate novel variables that better account for CI performance, and the need for improved data collaboration and standardized registries moving forward.

Speech recognition with a cochlear implant (CI) tends to be better for younger adults than older adults. However, older adults may take longer to reach asymptotic performance than younger adults. The present study aimed to characterize speech recognition as a function of age at implantation and listening experience for adult CI users. Retrospective review. A retrospective review identified 352 adult CI recipients (387 ears) with at least 5 years of device listening experience. Speech recognition, as measured with consonant-nucleus-consonant (CNC) words in quiet and AzBio sentences in a 10-talker noise masker (10 dB signal-to-noise ratio), was reviewed at 1, 5, and 10 years postactivation. Speech recognition was better in younger listeners, and performance was stable or continued to improve through 10 years of CI listening experience. There was no indication of differences in acclimatization as a function of age at implantation. For the better performing CI recipients, an effect of age at implantation was more apparent for sentence recognition in noise than for word recognition in quiet. Adult CI recipients across the age range examined here experience speech recognition benefit with a CI. However, older adults perform more poorly than young adults for speech recognition in quiet and noise, with similar age effects through 5 to 10 years of listening experience. 3 Laryngoscope, 131:2106-2111, 2021.

Combined Acoustic and Electric Stimulation

Many cochlear implant centers screen patients for cognitive impairment as part of the evaluation process, but the utility of these scores in predicting cochlear implant outcomes is unknown. To determine whether there is an association between cognitive impairment screening scores and cochlear implant outcomes. Retrospective case series of adult cochlear implant recipients who underwent preoperative cognitive impairment screening with the Montreal Cognitive Assessment (MoCA) from 2018 to 2020 with 1-year follow-up at a single tertiary cochlear implant center. Data analysis was performed on data from January 2018 through December 2021. Cochlear implantation. Preoperative MoCA scores and mean (SD) improvement (aided preoperative to 12-month postoperative) in Consonant-Nucleus-Consonant phonemes (CNCp) and words (CNCw), AzBio sentences in quiet (AzBio Quiet), and Cochlear Implant Quality of Life-35 (CIQOL-35) Profile domain and global scores. A total of 52 patients were included, 27 (52%) of whom were male and 46 (88%) were White; mean (SD) age at implantation was 68.2 (13.3) years. Twenty-three (44%) had MoCA scores suggesting mild and 1 (2%) had scores suggesting moderate cognitive impairment. None had been previously diagnosed with cognitive impairment. There were small to medium effects of the association between 12-month postoperative improvement in speech recognition measures and screening positive or not for cognitive impairment (CNCw mean [SD]: 48.4 [21.9] vs 38.5 [26.6] [d = -0.43 (95% CI, -1.02 to 0.16)]; AzBio Quiet mean [SD]: 47.5 [34.3] vs 44.7 [33.1] [d = -0.08 (95% CI, -0.64 to 0.47)]). Similarly, small to large effects of the associations between 12-month postoperative change in CIQOL-35 scores and screening positive or not for cognitive impairment were found (global: d = 0.32 [95% CI, -0.59 to 1.23]; communication: d = 0.62 [95% CI, -0.31 to 1.54]; emotional: d = 0.26 [95% CI, -0.66 to 1.16]; entertainment: d = -0.005 [95% CI, -0.91 to 0.9]; environmental: d = -0.92 [95% CI, -1.86 to 0.46]; listening effort: d = -0.79 [95% CI, -1.65 to 0.22]; social: d = -0.51 [95% CI, -1.43 to 0.42]). In this case series, screening scores were not associated with the degree of improvement of speech recognition or patient-reported outcome measures after cochlear implantation. Given the prevalence of screening positive for cognitive impairment before cochlear implantation, preoperative screening can be useful for early identification of potential cognitive decline. These findings support that screening scores may have a limited role in preoperative counseling of outcomes and should not be used to limit candidacy.

Review the relationship of tonotopic mismatch with the speech recognition of cochlear implant (CI) users with unilateral hearing loss (UHL; also known as single-sided deafness). Twenty-seven adults (≥18 yr of age) with late-onset UHL. Cochlear implantation. Speech recognition was assessed at 6 months post-activation with consonant-nucleus-consonant (CNC) words in the CI alone condition (contralateral ear masked). In the combined condition (CI plus the normal-hearing ear), masked speech recognition was assessed using AzBio sentences in a 10-talker masker (0-dB signal-to-noise ratio) in three target-to-masker configurations. Tonotopic mismatch was calculated as the semitone deviation between the center filter frequency and the cochlear place frequency of the most apical electrode contact. There was a significant, negative association between tonotopic mismatch and CNC scores ( r27 = -0.43, p = 0.013) and masked speech recognition when the target was from the front and the masker was presented toward the normal-hearing ear ( r27 = -0.36, p = 0.033). The speech recognition of adult CI users with UHL in the CI alone and bilateral listening conditions may be significantly influenced by tonotopic mismatches. These findings support the need for prospective investigation of methods to reduce or eliminate tonotopic mismatches (e.g., implantation of electrode arrays that approximate cochlear place and/or individualized mapping of filter frequencies) for CI users with UHL.

Predictors of second-side cochlear implant performance have not been well studied. We sought to assess whether speech recognition scores from first-side cochlear implant (CI1) could predict second-side cochlear implant (CI2) scores in sequential bilaterally implanted adults. Retrospective review using a prospectively collected database. Academic tertiary care hospital. Fifty-seven adults with postimplantation speech recognition testing performed at least 12 months after CI2. Sequential bilateral CI. CI2 performance at ≥12 months as measured using consonant-nucleus-consonant (CNC) words and AzBio sentences in quiet and +10 dB signal-to-noise ratio (S/N). CI1 performance scores at ≥12 months were independently associated with CI2 performance scores at ≥12 months for CNC words (β = 0.371 [0.136-0.606], p = 0.003), AzBio sentences in quiet (β = 0.614 [0.429-0.80], p < 0.0001), and AzBio +10 dB S/N (β = 0.712 [0.459-0.964], p < 0.0001). CI1 scores on AzBio in quiet at 0 to 6 months were also independently associated with CI2 AzBio in quiet scores at ≥12 months (β = 0.389 [0.004-0.774], p = 0.048). Hearing loss etiology and duration, age at implantation, interval between CI1 and CI2, duration of hearing aid use, and preimplantation speech recognition testing scores were not consistently associated with CI2 scores at ≥12 months. CI1 performance is an independent predictor of second-side performance as measured ≥12 months postimplantation. This may be a clinically useful metric when considering adult sequential bilateral implantation.

While single-sided deafness cochlear implants (SSD-CIs) have now received regulatory approval in the United States, candidate-ear candidacy criteria (no better than 5% word-recognition score) are stricter than for traditional CI candidates (50 to 60% speech recognition, best-aided condition). SSD implantation in our center began before regulatory approval, using a criterion derived from traditional candidacy: 50% consonant-nucleus-consonant (CNC) word-identification score in the candidate ear. A retrospective analysis investigated whether SSD patients exceeding the 5% CNC criterion nevertheless benefitted from a CI as assessed by spatial-hearing tests (speech understanding in noise [SIN] and localization) and by a patient-reported outcome measure quality-of-life instrument validated for patients with CIs. A retrospective chart review assessed the clinical experience of a single CI center. Subjects consisted of 27 adult CI recipients with SSD (N = 21) or asymmetric hearing loss (AHL; N = 6) implanted since September 2019 with at least 3 months of postoperative follow-up. Patients with revision surgery or simultaneous labyrinthectomy and CI surgery were excluded from the sample. Subjects were divided into 2 groups based on preoperative CNC scores measured under best-aided conditions with a behind-the-ear hearing aid in the sound field at 0.9 m from a front loudspeaker, and the better ear masked using an insert earphone with 45 dB HL speech-weighted noise. The "MEETS" group had preoperative CNC word scores <5%; the "EXCEEDS" group had scores >5%. The clinical protocol also included intelligibility tests using AzBio sentences in the same test conditions as CNC; binaural spatial testing (broadband-noise sound localization, and matrix-sentence speech-reception thresholds in spatially separated noise) using a custom-built 7-speaker array; and the CI Quality of Life (CIQOL) instrument. To evaluate CI benefit, preoperative unaided performance was compared with postoperative binaural (acoustic ear + CI ear) performance at a clinic visit closest to 6 months postsurgery. Of 27 SSD-CI recipients, 11 subjects exceeded the 5% preoperative CNC candidacy criterion. Both the MEETS and EXCEEDS groups improved significantly on all 5 primary study outcome measures (CI-alone CNC and AzBio, binaural SIN and sound localization, and CIQOL). The only statistically significant differences observed between the MEETS and EXCEEDS groups were that preoperative CNC and AzBio scores were significantly higher for the EXCEEDS group, as expected given that the groups were defined based on preoperative speech-perception scores in quiet. There were no statistically significant differences between the MEETS and EXCEEDS groups in postoperative scores in any test or in the magnitude of the improvement from preoperative to postoperative assessment. SSD- and AHL-CI recipients exceeding the 5% CNC preoperative candidacy criterion significantly improved in CI-alone speech perception, spatial hearing, and subjectively reported CIQOL outcomes and the observed benefits were indistinguishable from SSD- and AHL-CI recipients who met the 5% criterion. A less-restrictive SSD-CI and AHL-CI candidacy criterion should be considered, and larger-scale clinical trials to evaluate CI efficacy using a less-stringent candidate-ear criterion are warranted.

It is difficult for professionals to feel confident when referring patients for cochlear implant (CI) candidacy evaluations, as eligibility is largely determined by speech recognition measures that are typically unavailable before formal assessment. The Spectral-Temporally Modulated Ripple Test (SMRT) is an automated psychoacoustic measure that could be administered in audiology clinics before CI referral and may aid in identifying likely CI candidates. This study evaluated the relationship between aided SMRT performance and aided speech recognition in adult CI candidates. Retrospective analysis of prospectively collected data. Tertiary academic medical center. Twenty-five adult, native English-speaking CI candidates. Diagnostic and research testing before cochlear implantation. Aided SMRT scores obtained in a research setting using participants' own device(s) (everyday listening configuration) were correlated (Spearman) with bilateral and unilateral aided speech recognition scores collected during the clinical CI evaluation, including Consonant-Nucleus-Consonant (CNC) word recognition and AzBio sentence recognition in quiet and noise. Significant correlations were observed between laboratory SMRT scores and all clinical unilateral (ear to be implanted) best-aided speech recognition measures (CNC: n=25, ρ=0.54, P=0.006; AzBio quiet: n=25, ρ=0.59, P=0.002; AzBio noise: n=11, ρ=0.72, P=0.012). Significant correlations were also observed between laboratory SMRT and clinical bilateral best-aided speech recognition in quiet (CNC: n=25, ρ=0.50, P=0.01; AzBio: n=25, ρ=0.53, P=0.007), but not with bilateral AzBio speech recognition in noise (n=18, ρ=0.28, P=0.26). SMRT performance in everyday listening configuration correlates with unilateral best-aided speech recognition in quiet and noise in the ear to be implanted in adult CI candidates. These findings set the stage for future work to investigate if SMRT, alone or in combination with other metrics, can be used to predict CI candidacy versus noncandidacy.

THE ARGUMENT FOR FITTING BIMODALLY If you see a child tomorrow with a hearing loss in both ears, will you recommend one hearing aid or two? The obvious answer is two. You would have a hard time finding a dispensing professional today who does not agree that the benefits of bilateral hearing aid fitting make it the standard of care for those with binaural hearing loss. While the benefits of binaural hearing and the advantage of bilateral fitting are beyond the scope of this article (e.g., see Litovsky et al.,1 Kochkin2), these facts are undisputed in hearing healthcare circles. The industry's confidence in bilateral hearing aids is supported by current trends in fitting. In 1980 only 27% of hearing aid fittings were bilateral.3 Today, it is an amazing 86% for those with binaural hearing loss.4 So, what is bimodal fitting and why should dispensing professionals care? Bimodal fitting means different stimuli are presented to each ear. For the purposes of this paper, it means a cochlear implant in one ear and a hearing aid in the other. But, you may ask, don't cochlear implant audiologists take care of that? The answer is no, at least not usually. Personal experience (first author), communication with cochlear implant audiologists, and the literature5 suggest that most hearing aids in bimodal devices are fitted outside the cochlear implant center. Thus, if you have a patient who receives a cochlear implant in one ear, you will most likely be the one responsible for the continuing care of the hearing aid in the contralateral ear. It is in the best interests of both your patient and you to know how to optimize the hearing aid fitting for the best bimodal performance. If you fit hearing aids on children, the question is not if you will be responsible for managing a child with bimodal devices, but rather when. The number of unilateral cochlear implant recipients who continue to use contralateral hearing aids is clearly increasing (Figure 1). The conventional wisdom that cochlear implants and hearing aids should not be used simultaneously is archaic,6,7as we will show in this paper.Figure 1: Percentage of unilateral cochlear implant users choosing to wear a hearing aid in the contralateral ear. Sources: Tyler et al.,8 Cowan and Chin-Lenn9.BIMODAL DEVICE USE IN CI WEARERS Significant advances over the years in cochlear implant technology, speech-coding strategies, and surgical techniques have resulted in substantial improvements in the auditory-only speech-understanding abilities of cochlear implant recipients.10 As a result, the candidacy criteria approved for cochlear implantation in the United States has progressively expanded. When Cochlear Corporation, Ltd., introduced the original Nucleus® cochlear implant in 1985, the only candidates approved by the Food and Drug Administration were adults with profound bilateral sensorineural hearing loss of post-linguistic origin who had 0% open-set speech recognition using hearing aids. Now, under the FDA criteria approved in 2005, candidates can be adults or children aged 12 months and older, and can have either pre- or post-lingual onset of hearing loss. Although mid- and high-frequency hearing must still be profound (hearing thresholds >90 dB HL), low-frequency hearing loss can be moderate for adults (hearing thresholds >40 dB HL) and severe for children over age 2 (hearing thresholds >70 dB HL). Further, best-aided pre-operative speech-recognition criteria have been raised from 0% to <60%. Figure 2 shows the current criteria for each age group.Figure 2: Current FDA-approved audiometric and speech-recognition criteria for cochlear implantation with the Nucleus device, by age group. (For children, the open-set word-recognition test recommended is the Lexical Neighborhood Test [LNT] or Multisyllabic Lexical Neighborhood Test [MLNT], which are available from www.auditec.com.)For persons with bilaterally profound sensorineural deafness (the purple-shaded area in Figure 2), cochlear implants are clearly the intervention of choice because many obtain little or no benefit from hearing aids. However, for children aged 2 years and up and for adults, there is a range of low-frequency thresholds (the green and yellow areas, respectively) that fall within the approved audiometric range for cochlear implants. Hearing aids often fail to provide adequate performance for these patients,11but a unilateral cochlear implant alone does not provide all the known benefits that arise from listening with two ears rather than one. Binaural benefits from perception of interaural differences in time and intensity are well known to improve speech-recognition performance, particularly in background noise, due to a combination of head shadow, binaural redundancy, and binaural squelch effects (e.g., see Byrne, 198112 for a review). Further, bilateral inputs provide the potential for good localization ability. Finally, a strong argument can be made for bilateral stimulation, especially in children, in light of the impact of auditory deprivation on perception. When a hearing-impaired ear remains unaided, speech-recognition ability in that ear significantly deteriorates over time,13,14 and there appears to be a limited window of opportunity for auditory system stimulation if the patient is to achieve maximal binaural functioning.15 Bilateral implantation is not for everyone. For example, there might be significant usable hearing in one ear. There may be insurance reimbursement or financial barriers. Parents may worry about surgery or preserving one ear for possible future technology or treatments. These concerns may or may not be well-founded. Insurance reimbursement is not the obstacle it once was. Cochlear brand implants are designed to be “backward compatible” so future advances can be applied to implants done today. Cotanche reported that treatment, e.g., hair cell regeneration, may be 20 years or more away.16 However, unilateral versus bilateral implantation in children is ultimately the parents' choice and their wishes must be respected. The less expensive, non-invasive fitting of a hearing aid on the ear contralateral to a cochlear implant allows preservation of hearing in that ear and may provide the benefits of binaural stimulation. SUMMARY OF THE LITERATURE The bimodal fitting approach was first reported in the literature in the early 1990s (e.g., Shallop et al., 199217). Concerns were initially expressed that patients might be unable to combine the two very different sound sources for central processing. Fortunately, this has not proven to be the case. In fact, some researchers have argued that bimodal stimulation may provide “complementary” cues for processing of signals that may be advantageous to speech perception.18 Specifically, the lower frequencies provided by the hearing aid can provide information about the fundamental frequencies of a talker's voice and vowel information, while the mid- and high-frequency information from the cochlear implant can provide information needed on manner and place of articulation of consonants. It has also been suggested that localization ability, sound quality, and music perception may be enhanced by bimodal devices compared with bilateral cochlear implants.19,20 Studies have reported significant speech-recognition improvements for bimodal listening compared to either the patients' pre-operative bilateral hearing aid use or their post-operative use of the hearing aid or cochlear implant alone. This has been shown in adults17,21–23 and in children.24–26 For example, in a study by Luntz et al.,26 12 subjects (3 post-lingually impaired adults and 9 pre-lingually impaired adults and children aged 7 and older) were tested on sentences in noise after 7 to 12 months of using bimodal devices. Both speech (at 55 dB HL) and noise (at 45 dB HL) were presented from a frontal loudspeaker. Average speech-recognition scores were only 12.9% for the hearing aid alone and 60.7% for the cochlear implant alone, but bimodal listening produced an average score of 75.6% correct. Localization abilities have been shown to improve with bimodal devices relative to use of either device alone for some, although not all, adult2728 and pediatric1,24 patients. Many users of bimodal devices have also reported higher levels of satisfaction and perceived benefit than with hearing aids worn pre-implantation, although cosmetic and handling concerns of using the two devices have sometimes been expressed,29 emphasizing the need for sufficient counseling and training. It is also important to consider that children may need more time to learn to use bimodal cues.25 There is debate over the relative effectiveness of bilateral cochlear implants versus bimodal devices. Overall, however, the published literature on bimodal devices has been quite positive (e.g. see Ching et al. for a review18). A judicious approach would be to fit a hearing aid contralaterally to the implant on patients who show sufficient benefit from the hearing aid and are able to use the binaural cues provided. FACTORS IN FITTING THE HEARING AID Certain aspects of the fitting need to be considered and possibly modified for optimal use of bimodal devices. Dispensing professionals who follow proven, evidence-based protocols for hearing aid fitting, however, will require minimal adaptation of their normal procedure. The American Academy of Audiology has published a Pediatric Amplification Protocol and all professionals dispensing hearing aids to children should be familiar with it.30 Optimization of the hearing aid in bimodal fittings essentially requires three steps. First, the cochlear implant map must be stable. You will need to communicate with the cochlear implant audiologist to know when this has been accomplished. Second, a frequency response should be selected for the hearing aid that will provide the best speech intelligibility. This is established by starting with a hearing aid that has been fitted and verified using a prescriptive formula. While the first author has had success using NAL-NL1,31 and Ching recommended it as an optimal starting point,32 those who are proficient with DSL[i/o]33 or another validated prescriptive approach should not be discouraged from using it as the starting point. From the initial prescription, two alternate frequency responses should be programmed into the hearing aid and adjusted for equal loudness. This is easy in multiple-memory digital hearing aids. As the limits of the hearing aid permit, program one should be the selected prescriptive formula frequency response. Program two should have 6-dB per octave less amplification in the low frequencies (-6 dB at 1000 Hz, -12 dB at 500 Hz, and -18 dB at 250 Hz). Program three should have 6-dB per octave more amplification in the low frequencies (+6 dB at 1000 Hz, +12 dB at 500 Hz, and +18 dB at 250 Hz). Once the programs are established, the child should listen to connected discourse while the audiologist switches between programs to determine which one provides the clearest speech. This can be done by playing a recorded story or watching a child-friendly video. The cochlear implant should be turned off during this frequency response selection process. Ching reported that this procedure is appropriate for children as young as 6 years.32 For younger children, the professional may choose to default to the prescriptive response. Finally, the third step in the fitting protocol is to match overall loudness between the hearing aid and cochlear implant. Both the implant and the aid are turned on and the child is asked to report if the hearing aid is louder or softer than the cochlear implant. The aid is then adjusted accordingly. This can also be done while the child listens to a recorded story or watches a video. A chart like that in Figure 3 can be helpful for this task. Some children might experience loudness discomfort from amplification. If so, Ullauri et al. suggest starting with a lower volume setting on the hearing aid and raising it over time as acclimatization occurs until the level of balanced loudness is achieved.34Figure 3: Loudness balancing scale. Source: Cochlear in-house material.A flow chart for fitting the hearing aid in bimodal devices is shown in Figure 4. This recommended protocol has been validated in children and found to provide good binaural benefits.24 For the reader wishing more in-depth training, a tutorial is available at www.cochlearcollege.com. Ching et al. have also published excellent articles on fitting and adjusting the hearing aid for children wearing bimodal devices.2,35Figure 4: Optimizing the hearing aid in bimodal fitting. Source: Cochlear in-house material.CONCLUSIONS The use of bimodal devices is the recommended treatment option for children who meet cochlear implant candidacy but who either have some usable hearing in one ear or for other reasons get only one implant. Bimodal devices can be a successful alternative to bilateral hearing aids or to one cochlear implant alone. It is important to remember these three vital rules: (1) Work with the implant center to make sure the implant map is stable. (2) Fit the hearing aid frequency response for maximal speech intelligibility. (3) Balance the loudness with the cochlear implant and hearing aid. Bimodal fitting can provide optimal use of the different, but potentially complementary, bilateral cues provided by the acoustic amplifier and the electric stimulation from the implant.