Bone conduction devices (BCDs) and middle ear implants (MEIs) are established treatments for persistent mixed hearing loss (MHL). While both can yield more and less comparable audiological outcomes within their indication ranges, differences may arise in cases of asymmetrical MHL. In such cases, better bone conduction thresholds in the contralateral ear may lead to cross-hearing with BCDs. MEIs, offering side-specific stimulation, are expected to avoid this issue. This retrospective study aims to clinically assess spatial hearing and sound localization in patients with talk-over risk.
Patients with BCDs or MEIs and at least a 10 dB difference in bone conduction thresholds at two or more frequencies on the implanted side were included. Spatial hearing was evaluated in S₀N₀, SᵢN𝚌, and S₀Nᵢ conditions, both aided and unaided. Localization ability was also assessed.
Preliminary data from the BCD group (N=10) show improved signal-to-noise ratios of 2 dB (S₀N₀) and 6 dB (SᵢN𝚌). However, no improvement—and slight deterioration—was observed in the S₀Nᵢ (squelch) condition. No benefit in localization was found with BCD use. Data from the MEI group are still being collected and will be presented.
Initial findings suggest limitations in spatial hearing and localization with BCDs. The comparison will become more insightful once MEI data are available.

Background:
The objective of this study was to ascertain a range of sensorineural hearing thresholds for which the application of two different power Bone Conduction Devices (BCDs) would result in sufficient speech intelligibility. This fitting range is based on the maximum stable gain (MSG, feedback limit). Furthermore, the available aided dynamic range (DR) at this MSG is evaluated.
Methods:
A feedback test was performed in 42 individual patients with a new BCD on their percutaneous implant, using the proprietary BCD-fitting software. Subsequently, the BCD gain was temporarily set at MSG, and the gain and output were measured on a skull-simulator for a range of speech levels. The fitting range was derived using the speech intelligibility index and the DR from the output measurements.
Results:
Based on an average MSG of 11.6 ± 4.1 dB for device 1 and 15.9 ± 5.8 dB for device 2, the fitting range of a power-BCD is 35–40 dB HL for good perception of soft speech, and up to 45–50 dB HL for moderate-to-sufficient understanding of normal-level speech in at least half of the users with this level of hearing loss. The DR was on average 17,4 dB for device 1 and 15.6 dB for device 2. The device output was not constrained by the maximum force output for conventional speech levels, except at low frequencies.
Conclusions:
With the gain set at the maximum stable gain, power BCDs enable sufficient speech understanding for mild-to-moderate sensorineural hearing loss cases, and the aided dynamic range is adequate for perceiving loudness differences.

Background:
Objective verification of bone conduction devices (BCDs) remains underdeveloped compared to air conduction hearing aids. Aided thresholds—commonly used—are vulnerable to tester variability, listener state, and device setting artifacts. In contrast, output verification using a surface microphone can offer a direct, standardized method for assessing aided audibility. However, clinical use of the Real-Head to Dial Difference (RHDD), analogous to RECD, requires normative data, especially across diverse BCD coupling methods where device behavior and tissue impedance differ.

Methods:
This prospective study aims to establish average RHDD values across three coupling types: percutaneous, softband, and active transcutaneous. We are enrolling up to 60 participants: 20 percutaneous, 20 softband, and as many active transcutaneous users as available. For each, we measure skull-simulator output at dial settings (using Verifit2) and on-skin SPL via the Audioscan SurfMic in the in-situ position. RHDD is calculated as the difference between these measures across key frequencies (0.5–4 kHz). Age, head circumference, and device model are recorded to explore potential predictors of RHDD variability.

Results:
Preliminary data from 16 participants show coupling-type-dependent differences in RHDD, with wider variance observed in softband configurations. Early trends suggest anatomical and device-specific contributions to RHDD deviations.

Conclusion:
Establishing normative RHDD values will enable clinicians to interpret surface microphone data confidently, improving BCD verification beyond the limitations of aided thresholds. This work lays the groundwork for integrating RHDD-based methods into routine clinical practice, aligning with the goals of DSL-BCD and modern prescriptive verification.

Introduction
Accurately predicting the output of bone conduction hearing aids (BCHAs) is critical for optimizing device performance and ensuring patient satisfaction. A surface microphone mounted on the forehead offers a non-invasive approach to estimate BCHA output; however, its accuracy across different devices and frequency ranges remains uncertain. This study aimed to evaluate the predictive accuracy of the surface microphone for transcutaneous and adhesive BCHAs by comparing its output with objective experimental measurements.

Methods
Three BCHAs (one ADHEAR and two BONEBRIDGE devices) were installed on five fresh-frozen human cadaver heads. A stepped sine stimulus ranging from 0.1 to 10 kHz was applied under all conditions. During stimulation, three-dimensional cochlear promontory motion and intracochlear pressures in the scala vestibuli and scala tympani (accessed via mastoidectomy) were recorded. These objective parameters were compared with sound pressure measurements obtained from a forehead-mounted surface microphone attached using a soft band.

Results
Strong linear relationships were observed between forehead surface sound pressure, three-dimensional promontory motion, and differential intracochlear pressure across all three BCDs, particularly in the frequency range of 700 Hz to 1.5 kHz. At frequencies above 2 kHz, the sensitivity of the surface microphone declined for all devices, approaching the noise floor beyond 3 kHz and limiting its predictive accuracy at higher frequencies.

Conclusions
The forehead-mounted surface microphone reliably predicts BCHA output at lower frequencies but demonstrates reduced sensitivity above 2 kHz. While the sensor shows promise for non-invasive verification, refinements or complementary methods may be necessary to achieve accurate output predictions across the full clinically relevant frequency spectrum.

Background:
Clinicians often compare audiometric bone-conduction thresholds to manufacturer fitting guidelines to guide implant and device selection. However, clinical skull-simulator measurements suggest that achieving sufficient aided audibility for moderate-to-severe mixed hearing loss can be difficult, creating uncertainty in BCD and implant selection. This study aimed to develop a web-based tool to predict the fitting characteristics of percutaneous bone-conduction devices (BCDs) using pre-surgical audiograms.
Methods:
A quasi-experimental, prospective study with a repeated-measures design was conducted with 96 adults using BCDs on skin-penetrating abutments. Audiometric and in-situ bone-conduction thresholds, skull-simulator output, and QuickSIN scores were collected. Regression analyses examined the relationship between audiometric thresholds at 0.5, 1, 2, and 4 kHz (predictors) and in-situ thresholds, skull-simulator output, and QuickSIN scores (outcomes).
Results:
Regression models significantly predicted in-situ thresholds and skull-simulator outputs. A statistically significant relationship was also found between audiometric thresholds and QuickSIN scores. These models were used to create a web-based tool that estimates the expected skull-simulator output for percutaneous BCDs based on audiometric BC thresholds.
Conclusion:
Audiometric bone-conduction thresholds can predict the aided frequency response of percutaneous BCDs by estimating in-situ thresholds and skull-simulator output. This predictive tool may support clinical decision-making and could be extended to active transcutaneous systems in the future. A beta version of the tool is under development.

Introduction: The BONEBRIDGE (BB, MED-EL, Austria) is an active transcutaneous bone conduction implant for treatment of conductive and mixed hearing loss. For the initial fitting, a fitting formula based on patient’s audiogram or Vibrogram is used. Currently, the DSL-V5 for children and adult are used as standard procedures. This study aimed to assess the benefit and feasibility of a new fitting formula, DSL-BC, and compare it to the standard approach and users own fine fitting.
Methods: Twenty-two participants (children, n=2; adults, n=20) were included in this clinical study, which evaluated both audiological and subjective outcomes in an acute test setting. Three fitting strategies—standard first fit, DSL-BC, and the patient’s own fitting—were tested in a randomized order. Audiological assessments included sound field (SF) thresholds, OLSA in quiet at 45dB and 65dB SPL, and OLSA in noise at 65 dB SPL (S₀/N₀).
Results: Preliminary results showed average PTA4 in SF of 30 dB HL for DSL-BC, 40 dB HL for the standard fit, and 35 dB HL for the patient’s own fitting. OLSA in quiet at 65 dB SPL yielded similar results across all conditions. However, at 45 dB, DSL-BC showed a significant advantage compared to other conditions. In noise, DSL-BC improved speech understanding by approximately 2 dB SNR compared to the standard fit.
Conclusion: This feasibility study demonstrates that the DSL-BC fitting formula offers clear audiological benefits and may serve as a reliable first-fit strategy for BONEBRIDGE users.

Background:
Selecting between bone-conduction implants (BCI) and active middle-ear implants (aMEI) for conductive or mixed hearing loss can be challenging due to overlapping candidacy. Over the past decade, our interdisciplinary team developed a decision tree to guide treatment choice based on audiological, anatomical, surgical, device availability, and funding factors.

Objective:
To review our decision tree and compare hearing and quality-of-life outcomes for BCI versus aMEI using an intention-to-treat analysis.

Methods:
We conducted a retrospective review of adults with conductive or mixed hearing loss at our tertiary implant centre. Candidates were categorized as BCI-only, aMEI-only, or both-eligible based on the decision tree. We compared unaided air- and bone-conduction thresholds, aided free-field thresholds (PTA4), effective gain, and 12-month APHAB scores.

Results:
Among 89 patients, 54 received BCI and 35 aMEI. By candidacy: 30 were BCI-only (100% underwent BCI), 37 were aMEI-only (95% received aMEI), and 22 were both-eligible (100% chose BCI). Preoperative audiograms showed similar AC/BC thresholds in BCI-only and both-eligible groups; the aMEI-only group had significantly poorer high-frequency AC thresholds and lower BC thresholds. Aided free-field PTA4 and frequency-specific thresholds were equivalent across groups (p > 0.05). The aMEI group achieved slightly greater effective gain—attributable to higher MPO and closer transducer placement—but APHAB global scores at 12 months did not differ between groups.

Conclusion:
Our decision tree produces comparable audiological and subjective outcomes for BCI and aMEI. Device selection should follow hearing-loss characteristics, expected MPO needs, and surgical risk. When both options are feasible, BCI is preferred to avoid middle-ear access, minimising operative morbidity.

Background:
Some adults with middle-ear pathology, anatomical malformation, or maximal conductive hearing loss cannot benefit from conventional hearing aids. Bone-conduction implants (BCI) and active middle-ear implants (aMEI) offer alternative solutions, but optimal device fitting remains challenging.

Objective:
To determine whether cortical auditory evoked potentials (CAEPs) can guide and improve BCI and aMEI fitting, thereby enhancing aided hearing thresholds and speech-in-noise performance.

Methods:
Fourteen adult implant users (BCI or aMEI) underwent CAEP recordings at Cz and Fz in response to three speech tokens (/U/, /A/, /ʃ/), delivered in the sound field. Each recording was analyzed using Hotelling’s T² statistical test, supplemented by visual waveform inspection. If no CAEP response was detected with the participant’s clinical device settings, device parameters were adjusted iteratively until a response appeared. Hearing outcomes were quantified as pure-tone average aided thresholds (PTA4) and adaptive speech-in-noise SNR scores, measured before and after optimisation.

Results:
Of 14 participants, three already exhibited CAEP responses with clinical settings, two could not be optimised despite parameter adjustments, and seven achieved clear CAEPs post-optimisation. Both univariate and multivariate analyses revealed significant improvements for the mid- and high-frequency tokens (/A/ and /ʃ/) (p < 0.05), while the low-frequency /U/ token showed no significant change. Optimisation yielded a mean PTA4 improvement of 11.3 dB (SD 7.9 dB) and a mean speech-in-noise gain of 5.1 dB SNR (SD 4.2 dB).
Conclusion:
CAEP-guided fitting successfully enhanced mid- and high-frequency audibility and speech-in-noise performance in adult BCI and aMEI users.