Tobias Rader

Speech Perception With Combined Electric-acoustic Stimulation and Bilateral Cochlear Implants in a Multisource Noise Field

Objective: The aim of the study was to measure and compare speech perception in users of electric-acoustic stimulation (EAS) supported by a hearing aid in the unimplanted ear and in bilateral cochlear implant (CI) users under different noise and sound field conditions. Gap listening was assessed by comparing performance in unmodulated and modulated Comité Consultatif International Téléphonique et Télégraphique (CCITT) noise conditions, and binaural interaction was investigated by comparing single source and multisource sound fields. Methods: Speech perception in noise was measured using a closed-set sentence test (Oldenburg Sentence Test, OLSA) in a multisource noise field (MSNF) consisting of a four-loudspeaker array with independent noise sources and a single source in frontal position (S0N0). Speech simulating noise (Fastl-noise), CCITT-noise (continuous), and OLSA-noise (pseudo continuous) served as noise sources with different temporal patterns. Speech tests were performed in two groups of subjects who were using either EAS (n = 12) or bilateral CIs (n = 10). All subjects in the EAS group were fitted with a high-power hearing aid in the opposite ear (bimodal EAS). The average group score on monosyllable in quiet was 68.8% (EAS) and 80.5% (bilateral CI). A group of 22 listeners with normal hearing served as controls to compare and evaluate potential gap listening effects in implanted patients. Results: Average speech reception thresholds in the EAS group were significantly lower than those for the bilateral CI group in all test conditions (CCITT 6.1 dB, p = 0.001; Fastl-noise 5.4 dB, p < 0.01; Oldenburg-(OL)-noise 1.6 dB, p < 0.05). Bilateral CI and EAS user groups showed a significant improvement of 4.3 dB (p = 0.004) and 5.4 dB (p = 0.002) between S0N0 and MSNF sound field conditions respectively, which signifies advantages caused by bilateral interaction in both groups. Performance in the control group showed a significant gap listening effect with a difference of 6.5 dB between modulated and unmodulated noise in S0N0, and a difference of 3.0 dB in MSNF. The ability to “glimpse” into short temporal masker gaps was absent in both groups of implanted subjects. Conclusions: Combined EAS in one ear supported by a hearing aid on the contralateral ear provided significantly improved speech perception compared with bilateral cochlear implantation. Although the scores for monosyllable words in quiet were higher in the bilateral CI group, the EAS group performed better in different noise and sound field conditions. Furthermore, the results indicated that binaural interaction between EAS in one ear and residual acoustic hearing in the opposite ear enhances speech perception in complex noise situations. Both bilateral CI and bimodal EAS users did not benefit from short temporal masker gaps, therefore the better performance of the EAS group in modulated noise conditions could be explained by the improved transmission of fundamental frequency cues in the lower-frequency region of acoustic hearing, which might foster the grouping of auditory objects.

Prevalence and Characteristics of Preoperative Balance Disorders in Cochlear Implant Candidates

Objectives: The aim of this study was to elucidate the frequency and characteristics of preoperative vertigo symptoms in patients who undergo cochlear implantation (CI), in order to differentiate them from CI-related symptoms. Methods: In a prospective observational study, 47 adult CI candidates were asked about vertigo problems on a questionnaire. A subdivision into 3 groups was done: Group A (probable otogenic vertigo), group B (possible otogenic vertigo), and group C (not otogenic vertigo). Horizontal semicircular canal function was measured. Patients with vertigo complaints were compared to patients without vertigo with regard to the presence of abnormal vestibular function findings. Results: Twenty-five patients (53%) reported preoperative vertigo problems. In 21 (84%), the patients history suggested a probable (group A) or possible (group B) otogenic origin. Patients with vertigo more often had abnormal findings on vestibular function testing than did patients without vertigo. This difference, however, was not statistically significant. Conclusions: A considerable number of CI candidates have preoperative vertigo symptoms. These cannot be explained by horizontal semicircular canal function alone. In order to understand why CI patients develop postoperative vertigo, analysis of prospective preoperative vestibular function test findings and vertigo symptoms is necessary.

Speech Perception With Combined Electric-Acoustic Stimulation: A Simulation and Model Comparison.

Objective: The aim of this study is to simulate speech perception with combined electric-acoustic stimulation (EAS), verify the advantage of combined stimulation in normal-hearing (NH) subjects, and then compare it with cochlear implant (CI) and EAS user results from the authors’ previous study. Furthermore, an automatic speech recognition (ASR) system was built to examine the impact of low-frequency information and is proposed as an applied model to study different hypotheses of the combined-stimulation advantage. Signal-detection-theory (SDT) models were applied to assess predictions of subject performance without the need to assume any synergistic effects. Design: Speech perception was tested using a closed-set matrix test (Oldenburg sentence test), and its speech material was processed to simulate CI and EAS hearing. A total of 43 NH subjects and a customized ASR system were tested. CI hearing was simulated by an aurally adequate signal spectrum analysis and representation, the part-tone-time-pattern, which was vocoded at 12 center frequencies according to the MED-EL DUET speech processor. Residual acoustic hearing was simulated by low-pass (LP)-filtered speech with cutoff frequencies 200 and 500 Hz for NH subjects and in the range from 100 to 500 Hz for the ASR system. Speech reception thresholds were determined in amplitude-modulated noise and in pseudocontinuous noise. Previously proposed SDT models were lastly applied to predict NH subject performance with EAS simulations. Results: NH subjects tested with EAS simulations demonstrated the combined-stimulation advantage. Increasing the LP cutoff frequency from 200 to 500 Hz significantly improved speech reception thresholds in both noise conditions. In continuous noise, CI and EAS users showed generally better performance than NH subjects tested with simulations. In modulated noise, performance was comparable except for the EAS at cutoff frequency 500 Hz where NH subject performance was superior. The ASR system showed similar behavior to NH subjects despite a positive signal-to-noise ratio shift for both noise conditions, while demonstrating the synergistic effect for cutoff frequencies ≥300 Hz. One SDT model largely predicted the combined-stimulation results in continuous noise, while falling short of predicting performance observed in modulated noise. Conclusions: The presented simulation was able to demonstrate the combined-stimulation advantage for NH subjects as observed in EAS users. Only NH subjects tested with EAS simulations were able to take advantage of the gap listening effect, while CI and EAS user performance was consistently degraded in modulated noise compared with performance in continuous noise. The application of ASR systems seems feasible to assess the impact of different signal processing strategies on speech perception with CI and EAS simulations. In continuous noise, SDT models were largely able to predict the performance gain without assuming any synergistic effects, but model amendments are required to explain the gap listening effect in modulated noise.

Long-term Hearing Preservation Outcomes After Cochlear Implantation for Electric-Acoustic Stimulation.

Objective: This study reviewed outcomes of hearing preservation (HP) surgery in a cochlear implant patient population, with clinical follow-up results up to 11 years after implantation. Study Design: Retrospective case review. Setting: Tertiary referral university hospital. Patients: Ninety six patients (103 ears) with partial deafness who underwent HP surgery at the University Hospital Frankfurt since 1999 were included. Electrode carriers were Cochlear Slim Straight, MED-EL Standard, Medium, Flex20, and Flex24. Intervention: Cochlear implantation using the HP surgery technique with either the cochleostomy or round window approach. Main Outcome Measures: Pure-tone averages for low frequencies (125 Hz, 250 Hz, 500 Hz, PTAlow) and speech perception scores of the Freiburg monosyllable and number tests in quiet. PTAlow shifts were used to evaluate HP as complete for ⩽10 dB, partial between 10 and 30 dB, and minimal for ≥30 dB. Time intervals were: preoperative, postoperative, after 12 months, and long-term (>24 months, mean 51.4 months, range 2–11 years). Impacts of electrode design and surgical approach were analyzed. Results: Postoperatively (n = 103), HP was complete in 32 (31.1%), partial in 49 (47.6%), minimal in 14 (13.6%), and loss of hearing occurred in 8 cases (7.8%). After 12 months (n = 81), HP was complete in 22 (27.2%), partial in 33 (40.7%), minimal in 11 (13.6%), and loss of hearing occurred in 7 additional cases. For long-term outcomes (n = 62) HP was complete in 7 (11.3%), partial in 24 (38.7%), minimal in 9 (14.5%), and loss of hearing occurred in 7 additional cases (total 22/103, 21.4%). Cases with residual hearing who could utilize acoustic amplification (i.e., PTAlow < 80 dB HL) were 82/95 (85.3%) postoperatively, 58/66 (87.9%) after 12 months, and 38/40 (95.0%) for long-term outcomes. Conclusions: Long-term HP is feasible in a subset of patients. Patients with sufficient long-term residual hearing had the prerequisite to benefit from additional acoustic stimulation. No correlation of total hearing loss with etiology, electrode design, or surgical approach was evident. Apart from individual effects of structural damage or inflammation, genetic factors are suggested to influence HP. Cases with total hearing loss still demonstrated successful speech perception in long-term monosyllable recognition scores.

Hearing preservation and improved speech perception with a flexible 28-mm electrode.

Objective The present study aimed to determine the extent of hearing preservation retrospectively after atraumatic cochlear implant (CI) surgery using a specialized surgical technique and specially designed flexible electrode to minimize cochlear trauma. Study Design Retrospective study. Setting Academic tertiary care center. Patients A consecutive series of 34 patients who had some preoperative residual hearing were included in this study. Intervention Patients underwent CI surgery with a flexible 28-mm electrode using a round window insertion technique. Main Outcome Measures All patients had at least 6 months of postoperative follow-up including audiometric testing and speech perception determined using the Freiburg monosyllable word test and the Oldenburger Sentence Test in noise. Audiometric testing served as a proxy for the evaluation of cochlear trauma and hearing preservation. Results Hearing was preserved to within 20 dB of preoperative low-frequency pure-tone audiometry (PTA) in 40.7% of patients. Hearing was preserved to within 20 dB of preoperative high-frequency PTA in 35.7% of patients. Overall, a deterioration in hearing thresholds was observed between preoperative assessment and first fitting. Speech perception improved significantly over time after surgery. Conclusion Using appropriate surgical techniques, and electrodes specially designed to minimize cochlear trauma, hearing preservation can be achieved.

Cochlear Reimplantation after Surgery for Electric-Acoustic Stimulation

Background/Aims: Combined electric-acoustic stimulation (EAS) of the auditory system usually results in better speech understanding than electric stimulation only, assuming low-frequency hearing can be preserved. Treatment options and outcomes for 4 subjects experiencing compromised hearing after EAS surgery are discussed. Methods: Each subject underwent revision surgery to replace the original 21-mm electrode arrays used in EAS surgery with cochlear implants (CIs) with 31-mm arrays. Our aims were: (1) to investigate whether deeper insertion is possible, and (2) to evaluate the influence of electrode insertion angle by comparing speech perception scores before and after revision surgery. Results: Deeper insertion was feasible in all subjects. Speech understanding scores after reimplantation were comparable to those seen after the first intervention. Conclusion: A 360-degree insertion under EAS conditions provides sufficient speech understanding, even in cases of additional hearing loss. Reimplantation with a longer electrode array is feasible in former EAS patients.

Towards a Unified Testing Framework for Single-Sided Deafness Studies: A Consensus Paper

Background: While hearing aids for a contralateral routing of signals (CROS-HA) and bone conduction devices have been the traditional treatment for single-sided deafness (SSD) and asymmetric hearing loss (AHL), in recent years, cochlear implants (CIs) have increasingly become a viable treatment choice, particularly in countries where regulatory approval and reimbursement schemes are in place. Part of the reason for this shift is that the CI is the only device capable of restoring bilateral input to the auditory system and hence of possibly reinstating binaural hearing. Although several studies have independently shown that the CI is a safe and effective treatment for SSD and AHL, clinical outcome measures in those studies and across CI centers vary greatly. Only with a consistent use of defined and agreed-upon outcome measures across centers can high-level evidence be generated to assess the safety and efficacy of CIs and alternative treatments in recipients with SSD and AHL. Methods: This paper presents a comparative study design and minimum outcome measures for the assessment of current treatment options in patients with SSD/AHL. The protocol was developed, discussed, and eventually agreed upon by expert panels that convened at the 2015 APSCI conference in Beijing, China, and at the CI 2016 conference in Toronto, Canada. Results: A longitudinal study design comparing CROS-HA, BCD, and CI treatments is proposed. The recommended outcome measures include (1) speech in noise testing, using the same set of 3 spatial configurations to compare binaural benefits such as summation, squelch, and head shadow across devices; (2) localization testing, using stimuli that rove in both level and spectral content; (3) questionnaires to collect quality of life measures and the frequency of device use; and (4) questionnaires for assessing the impact of tinnitus before and after treatment, if applicable. Conclusion: A protocol for the assessment of treatment options and outcomes in recipients with SSD and AHL is presented. The proposed set of minimum outcome measures aims at harmonizing assessment methods across centers and thus at generating a growing body of high-level evidence for those treatment options.

Management of Cochlear Implant Electrode Migration.

Objective: The present study reviewed a cochlear implant (CI) patient population after surgery, which received a free-fitting electrode carrier designed for hearing preservation. The aim was to determine the rate of electrode migration of the CI electrodes and present clinical and surgical implications. Study Design: Retrospective patient review. Setting: Tertiary referral university hospital. Patients: Two hundred seventy-eight patients implanted uni- or bilaterally with lateral wall electrodes designed for hearing preservation (358 implants). The control group was 323 patients implanted uni- or bilaterally with preformed perimodiolar electrodes (468 implants). Interventions: Determination of CI electrode migration was conducted according to a clinical test protocol. Revision surgery was offered in confirmed patients of electrode migration. A bone groove was considered to improve the fixation of the electrode. Main Outcome Measures: Audiological testing including speech audiometry, subjective sound quality rating, and bilateral pitch comparison in bilateral patients, as well as radiological examinations, were conducted. Results: Electrode migration was observed solely in patients implanted with lateral wall electrodes; 10 of 358 patients with free-fitting electrodes (2.8%) had electrode migration, which was successfully confirmed by the proposed clinical test protocol. Nine of the 10 confirmed patients underwent reinsertion surgery. Mean perception score decreased from 75.0% to 62.1% after electrode migration and recovered completely after reinsertion surgery. A flowchart to detect electrode migration was designed for clinical practice. Conclusion: Although electrode migration is a rare complication in CI surgery, long-term follow-up diagnostics should include a test protocol to detect electrode shifts of lateral wall electrode arrays. A reinsertion surgery should be conducted in confirmed patients to recover speech perception.

Impact of a moving noise masker on speech perception in cochlear implant users

Objectives Previous studies investigating speech perception in noise have typically been conducted with static masker positions. The aim of this study was to investigate the effect of spatial separation of source and masker (spatial release from masking, SRM) in a moving masker setup and to evaluate the impact of adaptive beamforming in comparison with fixed directional microphones in cochlear implant (CI) users. Design Speech reception thresholds (SRT) were measured in S0N0 and in a moving masker setup (S0Nmove) in 12 normal hearing participants and 14 CI users (7 subjects bilateral, 7 bimodal with a hearing aid in the contralateral ear). Speech processor settings were a moderately directional microphone, a fixed beamformer, or an adaptive beamformer. The moving noise source was generated by means of wave field synthesis and was smoothly moved in a shape of a half-circle from one ear to the contralateral ear. Noise was presented in either of two conditions: continuous or modulated. Results SRTs in the S0Nmove setup were significantly improved compared to the S0N0 setup for both the normal hearing control group and the bilateral group in continuous noise, and for the control group in modulated noise. There was no effect of subject group. A significant effect of directional sensitivity was found in the S0Nmove setup. In the bilateral group, the adaptive beamformer achieved lower SRTs than the fixed beamformer setting. Adaptive beamforming improved SRT in both CI user groups substantially by about 3 dB (bimodal group) and 8 dB (bilateral group) depending on masker type. Conclusions CI users showed SRM that was comparable to normal hearing subjects. In listening situations of everyday life with spatial separation of source and masker, directional microphones significantly improved speech perception with individual improvements of up to 15 dB SNR. Users of bilateral speech processors with both directional microphones obtained the highest benefit.

Effectiveness of Directional Microphones in Bilateral/Bimodal Cochlear Implant Users—Impact of Spatial and Temporal Noise Characteristics

OBJECTIVE To measure speech reception thresholds (SRTs) in co-located (S0N0) and diffuse noise conditions (multi-source noise field, MSNF) and to assess the impact of beamforming algorithms in MSNF in cochlear implant (CI) users. STUDY DESIGN Non-randomized, open, prospective study. SETTING Tertiary referral cochlear implantation center. PATIENTS Participants included 14 CI users (7 bimodal, 7 bilateral) and 14 normal hearing young adults. INTERVENTIONS Cochlear implantation. MAIN OUTCOME MEASURES SRTs were assessed by means of a German matrix sentence test in either continuous or modulated noise. Loudspeakers were configured in two different conditions: S0N0 and MSNF (speech source in front, four speakers distributed at ±28.6 and ±151.4 degrees). In MSNF, the CI speech processor microphone was set in different directional sensitivity settings: standard (sub-cardioid), fixed (super-cardioid), and adaptive. RESULTS In continuous noise, SRTs of both CI groups were comparable. In modulated noise, bimodal CI users showed lower SRTs than bilateral CI group, but significant benefit from glimpsing was only demonstrated in normal hearing participants. All subject groups showed significant spatial release from masking (i.e., SRT improvement in MSNF compared with S0N0 condition) in continuous noise. A tendency of improved SRT (1 dB bimodal, 2 dB bilateral) with fixed and adaptive directional sensitivity was found which could not be statistically confirmed due to large between-subject variations. CONCLUSIONS The absence of the glimpsing effect in CI users was reaffirmed in the present study. Although very effective in single noise source conditions, the beneficial impact of beamforming algorithms in multiple noise source conditions is poor.