Factors Predictive of Binaural Hearing Restoration by Cochlear Implant in Single-Sided Deafness.

INTRODUCTION: Cochlear implants (CIs) can restore binaural hearing in cases of single-sided deafness (SSD). However, studies with a high level of evidence in support of this phenomenon are lacking. The aim of this study is to analyze the effectiveness of CIs using several spatialized speech-in-noise tests and to identify potential predictors of successful surgery. METHODS: Ten cases underwent standard CI surgery (MEDEL-Flex24). The speech-in-noise test was used in three different spatial configurations. The noise was presented from the front (N0), toward the CI (NCI), and toward the ear (Near), while the speech was always from the front (S0). For each test, the speech-to-noise ratio at 50% intelligibility (SNR50) was evaluated. Seven different effects were assessed (summation, head shadow [HS], speech released of masking [SRM], and squelch for the CI and for the ear). RESULTS: A significant summation effect of 1.5 dB was observed. Contralateral PTA was positively correlated with S0N0-B and S0NCI-B (CIon and unplugged ear). S0N0-B results were positively correlated with S0N0-CIoff (p < 0.0001) and with S0Near-CIoff results (p = 0.004). A significant positive correlation was found between delay post-activation and HS gain for the CI (p = 0.005). Finally, the HS was negatively correlated with the squelch effect for the ear. CONCLUSION: CI benefits patients with SSD in noise and can improve the threshold for detecting low-level noise. Contralateral PTA could predict good postoperative results. Simple tests performed preoperatively can predict the likelihood of surgical success in reversing SSD.

Comparison of Tonotopic and Default Frequency Fitting for Speech Understanding in Noise in New Cochlear Implantees: A Prospective, Randomized, Double-Blind, Cross-Over Study.

OBJECTIVES: While cochlear implants (CIs) have provided benefits for speech recognition in quiet for subjects with severe-to-profound hearing loss, speech recognition in noise remains challenging. A body of evidence suggests that reducing frequency-to-place mismatch may positively affect speech perception. Thus, a fitting method based on a tonotopic map may improve speech perception results in quiet and noise. The aim of our study was to assess the impact of a tonotopic map on speech perception in noise and quiet in new CI users. DESIGN: A prospective, randomized, double-blind, two-period cross-over study in 26 new CI users was performed over a 6-month period. New CI users older than 18 years with bilateral severe-to-profound sensorineural hearing loss or complete hearing loss for less than 5 years were selected in the University Hospital Centre of Rennes in France. An anatomical tonotopic map was created using postoperative flat-panel computed tomography and a reconstruction software based on the Greenwood function. Each participant was randomized to receive a conventional map followed by a tonotopic map or vice versa. Each setting was maintained for 6 weeks, at the end of which participants performed speech perception tasks. The primary outcome measure was speech recognition in noise. Participants were allocated to sequences by block randomization of size two with a ratio 1:1 (CONSORT Guidelines). Participants and those assessing the outcomes were blinded to the intervention. RESULTS: Thirteen participants were randomized to each sequence. Two of the 26 participants recruited (one in each sequence) had to be excluded due to the COVID-19 pandemic. Twenty-four participants were analyzed. Speech recognition in noise was significantly better with the tonotopic fitting at all signal-to-noise ratio (SNR) levels tested [SNR = +9 dB, p = 0.002, mean effect (ME) = 12.1%, 95% confidence interval (95% CI) = 4.9 to 19.2, standardized effect size (SES) = 0.71; SNR = +6 dB, p < 0.001, ME = 16.3%, 95% CI = 9.8 to 22.7, SES = 1.07; SNR = +3 dB, p < 0.001 ME = 13.8%, 95% CI = 6.9 to 20.6, SES = 0.84; SNR = 0 dB, p = 0.003, ME = 10.8%, 95% CI = 4.1 to 17.6, SES = 0.68]. Neither period nor interaction effects were observed for any signal level. Speech recognition in quiet ( p = 0.66) and tonal audiometry ( p = 0.203) did not significantly differ between the two settings. 92% of the participants kept the tonotopy-based map after the study period. No correlation was found between speech-in-noise perception and age, duration of hearing deprivation, angular insertion depth, or position or width of the frequency filters allocated to the electrodes. CONCLUSION: For new CI users, tonotopic fitting appears to be more efficient than the default frequency fitting because it allows for better speech recognition in noise without compromising understanding in quiet.

Frequency Fitting Optimization Using Evolutionary Algorithm in Cochlear Implant Users with Bimodal Binaural Hearing.

Optimizing hearing in patients with a unilateral cochlear implant (CI) and contralateral acoustic hearing is a challenge. Evolutionary algorithms (EA) can explore a large set of potential solutions in a stochastic manner to approach the optimum of a minimization problem. The objective of this study was to develop and evaluate an EA-based protocol to modify the default frequency settings of a MAP (fMAP) of the CI in patients with bimodal hearing. METHODS: This monocentric prospective study included 27 adult CI users (with post-lingual deafness and contralateral functional hearing). A fitting program based on EA was developed to approach the best fMAP. Generated fMAPs were tested by speech recognition (word recognition score, WRS) in noise and free-field-like conditions. By combining these first fMAPs and adding some random changes, a total of 13 fMAPs over 3 generations were produced. Participants were evaluated before and 45 to 60 days after the fitting by WRS in noise and questionnaires on global sound quality and music perception in bimodal binaural conditions. RESULTS: WRS in noise improved with the EA-based fitting in comparison to the default fMAP (41.67 ± 9.70% versus 64.63 ± 16.34%, respectively, p = 0.0001, signed-rank test). The global sound quality and music perception were also improved, as judged by ratings on questionnaires and scales. Finally, most patients chose to keep the new fitting definitively. CONCLUSIONS: By modifying the default fMAPs, the EA improved the speech discrimination in noise and the sound quality in bimodal binaural conditions.

Correlation Between Cochlear Length, Insertion Angle, and Tonotopic Mismatch for MED-EL FLEX28 Electrode Arrays.

OBJECTIVE: To investigate the relationship between cochlear length, insertion angle, and tonotopic mismatch and to compare the tonotopic mismatches with respect to the spiral ganglion and the organ of Corti. STUDY DESIGN: Retrospective. SETTING: Tertiary referral center with cochlear implant program. PATIENTS: Analyses of patients' computed tomography images after cochlear implant surgery. INTERVENTION: Cochlear implantation with 28-mm-long straight lateral wall electrode arrays. MAIN OUTCOME MEASURE: Cochlear length, insertion angle, and insertion depth were assessed using the OTOPLAN software. Tonotopic mismatch for each electrode contact was estimated using the Greenwood (organ of Corti) and the Stakhovskaya (spiral ganglion) maps and compared. RESULTS: 106 cochleae were analyzed. 99% of the electrode arrays were located in the tympanic ramp. The insertion was complete in 96% of cases. The mean cochlear length was 34.5 mm and the mean insertion angle of the apical electrode was 545°. Cochlear length was negatively correlated with the insertion angle of the contacts E1 to E9 (all p < 0.004). The tonotopic mismatch was greater at the organ of Corti than at the spiral ganglion. It was also greater at the organ of Corti in larger cochleae (correlation with mismatch for E1 r = 0.421, p < 0.0001) and in the apical than in the middle and basal regions of the cochlea. CONCLUSION: Small cochlea size corresponded to higher insertion angle and reduction of tonotopic mismatch on a 28-mm-long straight lateral wall electrode array. Tonotopic mismatch could be minimized preoperatively by choosing electrode arrays according to the individual cochlear morphology and postoperatively by appropriate frequency fitting.

Evaluation of a New Bone Conduction Device for the Rehabilitation of Single-Sided Deafness: Effects on Speech Understanding in Noise.

INTRODUCTION: A new external, adhesive, no-pressure bone-conduction device provides rehabilitation for conductive hearing loss and single-sided deafness (SSD). The purpose of the study is to evaluate speech recognition performance with the bone-conduction contralateral routing of signal (aBC-CROS) and compare it to an air-conduction CROS (AC-CROS) used by subjects for at least 1 year. METHODS: Ten SSD patients underwent speech understanding in noise tests with their AC-CROS, the aBC-CROS, and unaided. The 1st test session took place the day the aBC-CROS was fitted, with the second session after 2 weeks of aBC-CROS use. Two configurations were used: speech presented on the deaf side and noise on the normal side and the reverse. RESULTS: The speech recognition threshold (SRT) improved with both devices when speech was presented to the deaf side. Nine patients showed significant improvement (p < 0.016) with the AC-CROS (mean: 2.8 dB) and the aBC-CROS (mean: 3.0 dB). Mean difference of improvement was significant between unaided and aBC-CROS (p = 0.001) or AC-CROS (p = 0.006). The SRT deteriorated by an average of 2.3 dB with the AC-CROS with noise presented on the deaf side, with significance found for six patients (p < 0.016). The aBC-CROS did not affect performance in this configuration (mean improvement: 0.3 dB) and only one patient had a significant SRT degradation (p < 0.016). Mean difference of improvement was significant between the AC-CROS and aBC-CROS (p = 0.021) or unaided (p = 0.05). DISCUSSION: The aBC-CROS is a good alternative to the existing CROS devices for SSD rehabilitation, as it offers the same benefit with none of the drawbacks when noise is on the patient's deaf side.

Speech polar plots for different directionality settings of SONNET cochlear implant processor.

Objectives: The newest CI processor from MED-EL company, the SONNET, has two new directional microphone settings. Besides the Omnidirectional microphone mode, it has the possibility to switch to Natural or Adaptive directionality. Both new modes favour perception of sound coming from a front-facing direction compared to sounds from sources at alternate azimuths. Natural directionality mimics the pinna effect of the normal external ear. Design: We undertook to verify the effect of these options in vivo by means of clinical audiological tests. Speech reception thresholds were successively measured for a variety of speech presentation azimuths while keeping the noise azimuths constant. Complete 'Speech Reception Threshold (SRT)-Polar-Plots' were obtained from these data for the Omnidirectional and Natural directionality modes of the SONNET. In addition, one 'SRT-point' was also measured in the 'Adaptive' mode for speech coming from 45° azimuth. Study sample: A group of 13 adult CI recipients participated. Only one of these subjects had previous experience with the SONNET processor. Results: Complete 'SRT-Polar-Plots' could be measured in Natural and Omnidirectional modes in CI recipients within an acceptable timeframe. The pinna-following directionality for Natural mode could be confirmed. Median SRT in noise for speech coming from the 45° azimuth speaker was -5.6 dB SNR for Omnidirectional, -9.1 dB SNR for Natural and -12.8 dB SNR for Adaptive microphone. Natural and Adaptive significantly improved performance compared to Omnidirectional mode at this optimal azimuth of 45° with a median improvement in SRT of 3.5 and 7.2 dB respectively. Conclusions: A novel audiological method, 'SRT-Polar-Plot', was developed and described. Significant directionality benefits for Natural and Adaptive mode were confirmed in vivo using this technique.

Using Electrically-evoked Compound Action Potentials to Estimate Perceptive Levels in Experienced Adult Cochlear Implant Users.

HYPOTHESIS: The cochlear implant (CI) fitting level prediction accuracy of electrically-evoked compound action potential (ECAP) should be enhanced by the addition of demographic data in models. INTRODUCTION: No accurate automated fitting of CI based on ECAP has yet been proposed. METHODS: We recorded ECAP in 45 adults who had been using MED-EL CIs for more than 11 months and collected the most comfortable loudness level (MCL) used for CI fitting (prog-MCL), perception thresholds (meas-THR), and MCL (meas-MCL) measured with the stimulation used for ECAP recording. Linear mixed models taking into account cochlear site factors were computed to explain prog-MCL, meas-MCL, and meas-THR. RESULTS: Cochlear region and ECAP threshold were predictors of the three levels. In addition, significant predictors were the ECAP amplitude for the prog-MCL and the duration of deafness for the prog-MCL and the meas-THR. Estimations were more accurate for the meas-THR, then the meas-MCL, and finally the prog-MCL. CONCLUSION: These results show that 1) ECAP thresholds are more closely related to perception threshold than to comfort level, 2) predictions are more accurate when the inter-subject and cochlear regions variations are considered, and 3) differences between the stimulations used for ECAP recording and for CI fitting make it difficult to accurately predict the prog-MCL from the ECAP recording. Predicted prog-MCL could be used as bases for fitting but should be used with care to avoid any uncomfortable or painful stimulation.

Combined electric and acoustic hearing performance with Zebra® speech processor: speech reception, place, and temporal coding evaluation.

OBJECTIVE: To assess the auditory performance of Digisonic(®) cochlear implant users with electric stimulation (ES) and electro-acoustic stimulation (EAS) with special attention to the processing of low-frequency temporal fine structure. METHOD: Six patients implanted with a Digisonic(®) SP implant and showing low-frequency residual hearing were fitted with the Zebra(®) speech processor providing both electric and acoustic stimulation. Assessment consisted of monosyllabic speech identification tests in quiet and in noise at different presentation levels, and a pitch discrimination task using harmonic and disharmonic intonating complex sounds ( Vaerenberg et al., 2011 ). These tests investigate place and time coding through pitch discrimination. All tasks were performed with ES only and with EAS. RESULTS: Speech results in noise showed significant improvement with EAS when compared to ES. Whereas EAS did not yield better results in the harmonic intonation test, the improvements in the disharmonic intonation test were remarkable, suggesting better coding of pitch cues requiring phase locking. DISCUSSION: These results suggest that patients with residual hearing in the low-frequency range still have good phase-locking capacities, allowing them to process fine temporal information. ES relies mainly on place coding but provides poor low-frequency temporal coding, whereas EAS also provides temporal coding in the low-frequency range. Patients with residual phase-locking capacities can make use of these cues.

Speech performance and sound localization abilities in Neurelec Digisonic® SP binaural cochlear implant users.

In this prospective study the outcome of the Digisonic® SP Binaural cochlear implant (CI), a device enabling electric stimulation of both cochleae by a single receiver, was evaluated in 14 postlingually deafened adults after 12 months of use. Speech perception was tested using French disyllabic words in quiet and in speech-shaped noise at +10 dB signal-to-noise ratio. Horizontal sound localization in quiet was tested using pink noise coming from 5 loudspeakers, from -90 to +90° along the azimuth. Speech scores in quiet were 76% (±19.5 SD) in the bilateral condition, 62% (±24 SD) for the better ear alone and 43.5% (±27 SD) for the poorer ear alone. Speech scores in noise were 60% (±27.5 SD), 46% (±28 SD) and 28% (±25 SD), respectively, in the same conditions. Statistical analysis showed a significant advantage of the bilateral use in quiet and in noise (p < 0.05 compared to the better ear). Significant spatial perception benefits such as summation effect (p < 0.05), head shadow effect (p < 0.0001) and squelch effect (p < 0.0005) were noted. Sound localization accuracy improved significantly when using the device in the bilateral condition with an average root mean square of 35°. Compared with published outcomes of usual bilateral cochlear implantation, this device could be a valuable alternative to two CIs. Prospective controlled trials, comparing the Digisonic SP Binaural CI with a standard bilateral cochlear implantation are mandatory to evaluate their respective advantages and cost-effectiveness.

Speech perception performance for 100 post-lingually deaf adults fitted with Neurelec cochlear implants: Comparison between Digisonic® Convex and Digisonic® SP devices after a 1-year follow-up.

CONCLUSION: Patients implanted with the Digisonic® SP device showed better identification scores than those implanted with the Convex device, with skills continuing to improve over a longer time period. Technological improvements were beneficial in terms of speech perception in quiet. OBJECTIVE: To compare speech perception skills for post-lingually deaf patients implanted with a previous Neurelec device, the Digisonic® Convex, with those implanted with a more recent one, the Digisonic® SP, which provides more electrodes and a faster stimulation rate. METHODS: This was a retrospective study of 100 implanted patients, 45 with the Digisonic® Convex implant and 55 with the Digisonic® SP. Speech perception (dissyllabic words and sentences, in open set) was evaluated until 1 year after implantation. RESULTS: Patients fitted with the Digisonic® SP implant showed significantly better scores after 3, 6, and 12 months (mean scores: 53%, 62%, and 68% for words; 58%, 69%, and 75% for sentences) than those fitted with the Convex implant (34%, 42%, and 43% for words; 38%, 59%, and 51% for sentences). The improvement in speech perception after implantation for SP patients continued throughout the 12 months for words and 6 months for sentences, versus 6 months for words and 3 months for sentences for Convex patients.

Intelligibility of interrupted and interleaved speech for normal-hearing listeners and cochlear implantees.

Speech intelligibility is degraded in the presence of a competing talker for cochlear implantees, presumably because of impaired tracking and integration of speech segments glimpsed in the masker valleys. This hypothesis was tested by assessing the intelligibility of periodically-interrupted bisyllables produced by a male and female talker, for normal-hearing listeners and implantees. A 4-Hz square-wave modulator with random phase was used to interrupt bisyllables from each talker. Stimuli were either presented alone (Experiment I) or interleaved (Experiment II: the two talkers were alternated). In Experiment I, the mean identification score for each voice was 88% for normal-hearing listeners and 35% for implantees. In Experiment II, the mean score corresponding to correct identification of both voices was 50% for normal-hearing listeners and 5% for implantees. Implantees identified at least one bisyllable among the two well above chance level but showed difficulties assigning it to the correct talker. This suggests that implantees can make use of partial information, but cannot track and integrate the non-adjacent components of interleaved speech as well as normal-hearing listeners. Additional results obtained with normal-hearing listeners tested with tone-vocoded syllables suggest that impaired tracking/integration for implantees stems from limited reception of spectral and temporal fine structure cues.

Effects of spectral smearing and temporal fine structure degradation on speech masking release.

This study assessed the effects of spectral smearing and temporal fine structure (TFS) degradation on masking release (MR) (the improvement in speech identification in amplitude-modulated compared to steady noise observed for normal-hearing listeners). Syllables and noise stimuli were processed using either a spectral-smearing algorithm or a tone-excited vocoder. The two processing schemes simulated broadening of the auditory filters by factors of 2 and 4. Simulations of the early stages of auditory processing showed that the two schemes produced comparable excitation patterns; however, fundamental frequency (F0) information conveyed by TFS was degraded more severely by the vocoder than by the spectral-smearing algorithm. Both schemes reduced MR but, for each amount of spectral smearing, the vocoder produced a greater reduction in MR than the spectral-smearing algorithm, consistent with the effects of each scheme on F0 representation. Moreover, the effects of spectral smearing on MR produced by the two schemes were different for manner and voicing. Finally, MR data for listeners with moderate hearing loss were well matched by MR data obtained for normal-hearing listeners with vocoded stimuli, suggesting that impaired frequency selectivity alone may not be sufficient to account for the reduced MR observed for hearing-impaired listeners.