Implantation of an active transcutaneous bone conduction device via the middle fossa surgical approach: an institutional case series

Introduction

Bone conduction implants (BCIs) can be used to restore hearing in patients who cannot wear, or do not benefit from, conventional hearing aids due to chronic otitis media, congenital malformations, or previous middle ear surgeries (Snik et al., 2005). BCIs are indicated for conductive or mixed hearing loss, as well as single-sided deafness (SSD) (Magele et al., 2019). Consisting of a surgically implanted component and an external audio processor, BCIs can be classified as either percutaneous, where the skin overlying the implant is penetrated by an external abutment, or transcutaneous, where the skin remains intact. Transcutaneous devices can be further subdivided into passive or active types. Passive systems use an external processor to transmit sound vibrations through the skin to the implant, whereas active systems generate vibrations directly to the temporal bone via the implanted component (Gerdes et al., 2016).

The BONEBRIDGE bone conduction implant (BB-BCI; MED-EL GmbH, Innsbruck, Austria) is an active transcutaneous device which consists of an externally worn audio processor and an active implant. The implant transducer (bone conduction-floating mass transducer, BC-FMT) is surgically implanted into the temporal bone and a coil section is placed below the skin behind the ear which holds the audio processor in place with a magnet. Soundwaves collected by the audio processor are sent to the BC-FMT, which then directly vibrates the bone (Zernotti and Sarasty, 2015). In Canada, the BB-BCI is indicated for patients over the age of 5 years, with conductive or mixed hearing loss, and bone conduction (BC) thresholds better than or equal to 45 dB hearing level (HL) measured at 0.5 to 3.0 kHz, or for patients with SSD and air conduction (AC) thresholds of better than or equal to 20 dB HL measured at 0.5 to 3.0 kHz. A bilateral fitting can be considered for patients with a symmetric conductive or mixed hearing loss, with an interaural difference in BC thresholds less than 10 dB pure-tone average (PTA), or less than 15 dB at individual frequencies. The active, transcutaneous design of the BB-BCI results in excellent surgical and audiological outcomes for patients by reducing the skin complications associated with percutaneous systems, as well as the vibration attenuation through the skin associated with passive systems (Sprinzl and Wolf-Magele, 2016). However, the active nature of the implant makes it quite large and technically challenging to place surgically. Our group proposed a novel middle fossa surgical approach for implanting the BB-BCI in 2013 as an alternative to the transmastoid or retrosigmoid locations. This approach was shown to have advantages over the previous surgical approaches, including a smaller incision, minimal dissection to the bone, and improved wound healing (You et al., 2019).

In 2019, the second-generation BB-BCI device, the BCI 602, was developed (Plontke et al., 2020) to address some of the limitations of the previous generation. The smaller profile of the BCI 602 expanded the anatomical selection criteria of the original BCI 601 to include patients who previously could not be implanted due to unfavorable anatomy, for example patients with insufficiently thick skull bones, or in patients with previous radical cavity operation or mastoid obliteration (Król et al., 2020; Sprinzl et al., 2021). The BCI 602 features a thinner BC-FMT, which reduces the required drilling depth from 8.7 mm to 4.5 mm; a reduced overall implant length from 69 mm to 64.3 mm; a flexible section between the coil and transducer; and self-drilling screws for fixation to the bone, which obviates the need for pre-drilling. Figure 1 shows a comparison between the original BCI 601 and the updated BCI 602. This new model is associated with several advantages over the previous generation: reduced drilling time (and therefore overall surgical time), more flexible placement, simplified pre-operative planning, and reduced likelihood of dura exposure (Plontke et al., 2020; VIBRANT Technology Assessment Team MED-EL, 2021). The smaller dimensions of the BCI 602 and the added flexibility can also allow for implantation in younger children with smaller temporal bones (Brkic et al., 2023; Willenborg et al., 2023). Another benefit of the BCI 602 is the reduction of imaging artifacts on MRI with a middle fossa placement (Utrilla et al., 2021). Early experience with the BCI 602 has shown similar safety and audiological outcomes to the BCI 601 (Brooks et al., 2025; Cywka et al., 2021, 2022; Lein et al., 2024; Plontke et al., 2020; VIBRANT Technology Assessment Team MED-EL, 2021; Willenborg et al., 2022, 2023).

In 2020, we published an institutional review of the audiological and surgical outcomes in 37 patients implanted with the BCI 601 via our novel middle fossa approach from 2013 to 2018. Our findings revealed a functional gain of 40.3 dB ± 19.0 dB (mean ± standard deviation), favorable speech perception outcomes, and no complications or instances of revision surgery with a mean follow-up time of 32 months (range of 9–71 months) (Siegel et al., 2020). The purpose of the present study was to update our case series with an additional 35 patients implanted with either the BCI 601 or the new generation BCI 602 device via the middle fossa surgical approach at our center. A pooled analysis of all 72 patients is presented, with outcomes compared between device generations.

Materials and methods

A retrospective review was conducted at London Health Sciences Centre (LHSC) London, Ontario, Canada of patients who were implanted with the BCI 601 or 602 using the middle fossa surgical approach between April 2013 and March 2025. Data from the first 37 patients implanted between the dates of April 2013 and May 2018 have been previously published (Siegel et al., 2020).

The data for an initial cohort of patients was collected under Western University's Health Sciences Research Ethics Board (HSREB) approval #100309. The remaining patients' data were retrospectively collected for the purposes of a quality-assurance review of the implant program and were exempted from further review by the HSREB under Tri-council Policy Statement 2, article 2.5. All patients signed an informed consent at the time of their surgical consent, allowing their data to be used for quality assurance and/or research, as is standard at our research hospital.

Surgical procedure

For all patients, preoperative three-dimensional (3D) computed tomography (CT) planning was performed to model the ideal positioning of the implant and ensure safety. This modeling technique obviated the need for image guidance or intraoperative surgical templates. The optimal location was determined to be the thickest possible portion of the middle fossa bone, which typically occurs either just anterior to the transverse-sigmoid junction, or further posteriorly just superior to the transverse sinus. Using the 3D models, surface landmarks were made on the skin to mark the placement of the BC-FMT. The BC-FMT was then implanted into the temporal bone through a cylindrical craniotomy. A detailed, step-by-step description of the 3D modeling and surgical technique has been published, including an annotated surgical video (You et al., 2019).

Briefly, a 3-cm horizontal incision was made within the hairline above the ear, centered on the planned implant position determined using preoperative imaging and 3D Slicer. The temporalis muscle was divided horizontally, and a subperiosteal pocket was developed to accommodate the receiver coil, with positioning confirmed using the manufacturer's sizing template.

A cylindrical well slightly larger than the diameter of the BC-FMT was drilled in the squamous temporal bone to accommodate the implant, preserving sufficient surrounding bone for fixation with two anchoring screws. Bone removal was carried down to the dura, which was carefully freed from the edges of the craniotomy. Hemostasis and gentle dural tenting were achieved using Surgical, with bipolar cautery and bone wax as required. The craniotomy walls were fashioned perpendicular to the dura to ensure optimal implant seating.

Device positioning was dictated by temporal bone thickness and preplanned with 3D Slicer. Implants placed anterior to the transverse–sigmoid junction were typically oriented with vertically aligned screws and a 90° bend at the transition segment, whereas implants placed superior to the transverse sinus were oriented horizontally with a 45° anterior tilt to optimize coil positioning relative to the auricle. The device was secured using self-drilling screws, with lifts used as required.

The external sound-processor (Amadé or Samba) fitting was scheduled for 2 weeks following surgery to allow for local inflammation to resolve, as per the manufacturer's recommendations. In certain cases, patients were activated at 3 to 4 weeks post-operatively if they were unable to attend their scheduled appointment.

Sound processor fitting

All patients were fitted with either the Samba or Samba 2 audio processors from MED-EL GmbH. Initially, some patients were fitted with the Amadé processor, as it was the approved model at the time. However, these patients were subsequently upgraded to the Samba or Samba 2 processor based on the time the processors became available and the patients' willingness to upgrade their equipment.

Programming for the patients using the Amadé audio processors was completed using the Siemens Connexx fitting software and the MED-EL Symfit software patch. BC thresholds were entered as AC measures, or an in-situ vibrogram threshold was measured directly in the Connexx/Symfit software. The vibrogram measure was not available for use in the Symfit software for the first set of implanted patients, but became the preferred method as it most accurately represents hearing through the implant. Thresholds were used to calculate the fitting using desired sensation level (DSL) prescriptive targets (Scollie et al., 2005), with modifications made based on patient input. Programs were then created and downloaded into the audio processor, with the number and type varying by patient preference and experience with use of hearing technology. Typically, three programs were provided at initial activation, including a universal program, a speech-in-noise program, and a user-specific program.

Programming for patients using the Samba 2 audio processor was completed in the Symfit 8.0 software. AC and BC thresholds obtained at the pre-assessment appointment were entered prior to programming, after which vibrogram thresholds were measured directly in the software.

Audiometry

Complete audiological testing was performed pre- and post-operatively by experienced implant audiologists. Pre-operative ear-specific AC and BC thresholds were measured in a sound treated room using ER-3A insert earphones coupled to foam tips and mastoid placement of a bone oscillator to present pure tones from 250 Hz to 8,000 Hz. A standard Hughson-Westlake threshold search was used for audiometric measures. Post-operatively, sound field thresholds were measured with the BB-BCI in place using a narrow band noise stimulus and the previously described audiometric techniques. Aided audiometry was completed at activation and at regular post-operative follow-ups. Plugging or masking of the non-implanted ear was completed as required using standard audiometric masking protocols to minimize the influence of the non-implanted ear on aided measures. According to standard protocol, the contralateral ear was not masked if hearing thresholds of the non-test ear were poorer than the sound field presentation level. Implants were set to DSL v.5 adult bone conduction device (BCD) prescriptive targets with consideration to patients' preferred listening levels. All patients underwent testing with noise reduction disabled. For their selected program, patients used a mix of omnidirectional and directional microphone settings, although a front-facing speaker was employed.

Speech perception

Speech perception was measured using the consonant-nucleus-consonant (CNC) word test and AzBio sentences test pre-operatively, at activation, and at each regular follow-up visit until the patient reported they were adjusted and comfortable with the sound, or at the 6-month visit, whichever came first. Open-set, recorded CNC word list and AzBio sentences materials were presented in quiet at 60 dB sound pressure level (SPL) in the sound field.

Outcome measures

Intraoperative complications, post-operative adverse events, and instances of implant failure or explanation were documented. Pure-tone averages (PTAs) were calculated as the average of the thresholds at frequencies 0.5, 1.0, 2.0, and 3.0 kHz for pre-operative AC and BC (unaided condition) and the post-operative aided conditions (with BB-BCI in use). The air-bone gap (ABG) was calculated as the difference between AC thresholds and BC thresholds of the implanted ear. Functional gain (FG) was calculated as the difference between pre-operative AC thresholds and the most recent post-operative aided thresholds. Effective gain (EG) was calculated for the patients with mixed hearing loss as the difference between pre-operative BC thresholds and the most recent post-operative aided thresholds. A two-sample t-test assuming unequal variances was performed to compare the mean FG between implant types (BCI 601 vs. BCI 602).

Results

Seventy-two patients were included in the review (44 implanted with the BCI 601 and 28 implanted with the BCI 602). There were 30 males and 42 females, with an average age at implantation of 43.9 years. All patients either had a documented diagnosis of conductive or mixed hearing loss (48 and 24 patients, respectively). Hearing loss etiology was chronic otitis media and/or mastoid cavities or congenital cholesteatoma, which precluded consistent use of conventional hearing aids. Medically, all patients were deemed by the implant team to be in good general health and without cognitive impairments precluding implantation. Thirty-nine patients were implanted on the right side and 33 on the left. All surgeries were performed by one of three experienced neurotologists at LHSC (SKA, MDB, and LSP). Aided audiometric data were available at a mean follow-up time from surgery of 5.2 months (range 3-13.3 months). An overview of patient demographics is given in Table 1.

Table 1. Demographics of all patients (n = 72) implanted with either the BONEBRIDGE BCI 601 or BCI 602.

^*Pure tone average (PTA) measured as the mean of thresholds at 0.5, 1.0, 2.0, and 3.0 kHz.

AC, air conduction; BC, bone conduction; ABG, air-bone gap; MHL, mixed hearing loss; CHL, conductive hearing loss.

Surgical outcomes

Patient charts were reviewed over the 12 year period (all patients: mean 69 months, range 6-149 months; patients implanted with the BCI 601: mean 103 months, range 46-149 months; patients implanted with the BCI 602: mean 16 months, range 6-39 months). Patients were discharged same day from the hospital and staples were removed 1 week post-operatively. Incisions were located within the hairline with good cosmesis. The implants were activated at 2 to 4 weeks post-operatively and good subjective user satisfaction was reported from all patients.

There were no major intra- or post-operative complications found for any of the patients. There were no instances of device failure or explantation. One patient (6A, BCI 601) noted tenderness at the incision site in the night post-operative, and another (11A, BCI 602) reported feelings of numbness and tingling at the scalp region superior to the implant location at 1 month post-operative. At 4 years post-implantation, one patient in the BCI 601 group (8A) reported increasing pain at the magnet site and the screw site. Physical examination was unremarkable except for tenderness to touch under the scar line, and there was no abnormality seen or felt at the magnet site. The patient was referred to Neurosurgery for follow-up with chronic pain.

Audiometric outcomes

Auditory thresholds improved for all patients following BB-BCI implantation. Audiometric outcomes across the cohort, as well as for subgroups defined by device generation and hearing loss type, are summarized in Table 2. The two-sample t-test showed no significant difference in FG between groups (BCI 601: 38.5 ± 15.0 dB; BCI 602: 39.1 ± 15.2 dB; p = 0.85). Closure of the air-bone gap was seen in both device generation groups, with overclosure achieved at 2.0 kHz in the BCI 602 group (Figure 2) and at 1.0, 2.0, 3.0, and 4.0 kHz in patients with MHL (Figure 3). The mean EG PTA (measured at 0.5, 1.0, 2.0, and 3.0 kHz) for patients with mixed hearing loss was −0.16 ± 11.6 dB (n = 24).

Table 2. Audiometric outcomes by implant generation (BONEBRIDGE BCI 601 or BCI 602) and hearing loss type (mixed or conductive).

All values are in dB.

^*Pure tone average (PTA) measured as the mean of thresholds at 0.5, 1.0, 2.0, and 3.0 kHz.

^§ABG calculated as AC-BC at 0.5, 1.0, 2.0, and 3.0 kHz. AC, air conduction; BC, bone conduction; MHL, mixed hearing loss; CHL, conductive hearing loss.

Speech perception outcomes

Paired speech perception results were available for a subset of patients, as summarized in Table 3. Results are shown pre-operatively, at device activation, and post-operatively at an average of 6.5 months. Aided improvements in speech perception can be seen for both tests, across all patients tested, with a mean increase of 16.8% and 50.3% for AZBio Sentences (n = 28) and CNC Words (n = 12), respectively, from pre-operative to post-operative time points.

Table 3. Speech perception outcomes for a subset of patients with paired assessments.

Included patients had measurements at T1 (pre-operative visit or activation) and T2 (most recent post-operative visit), with a mean follow-up interval of 6.5 months between time points.

Discussion

We previously conducted an institutional review of 37 patients implanted with the BB-BCI 601 via the middle fossa approach, which demonstrated a FG of 40.3 dB ± 19.0 dB (mean ± SD) 3 months postoperative, improved speech perception outcomes, and no complications or instances of revision surgery over a mean follow-up time of 32 months (range of 9–71 months) (Siegel et al., 2020). Since the introduction of the second-generation BB-BCI 602 in 2019, we have continued to implant patients with the middle fossa approach. The present report aimed to evaluate the safety and efficacy of the BCI 602 in an updated cohort of patients and to compare the outcomes with our previous findings. Our review included an additional 35 patients (7 implanted with the BCI 601 and 28 implanted with the BCI 602), bringing the total to 72 patients. Findings again showed no cases of explantation or revision surgery over a 12-year period (mean follow-up time of 16 months, ranging from 6–39 months, for patients implanted with the BCI 602). The mean FG across the cohort was 38.7 ± 15.0 dB at an average follow-up of 5 months post-implantation. Moreover, there was no significant difference in FG between the two implant types (p = 0.85). These findings further support the excellent safety and performance of both the BB-BCI 601 and 602, consistent with previously published reports.

Cywka et al. (2021) reported significant improvements in hearing and speech understanding, as well as subjective benefit in a cohort of 22 children ages 8 years and older implanted with the BCI 602, with a mean FG of 28.8 dB (±10.5 dB) at 6 months follow-up. The authors then confirmed stable audiological improvements and favorable surgical outcomes in a group of 42 adults, reporting a FG of 27 dB (±10.7 dB) at 12 months follow-up (Cywka et al., 2022). A reduction in total surgical time was reported for the BCI 602, with surgery times ranging from 19–66 minutes (mean 28.3 min ± 9.4) (n = 42) vs. 28–57 min (mean 35.6 min ± 8.1) (n = 10) for the BCI 601. Šikolová et al. (2022) prospectively compared BCI generations in 12 pediatric patients, with reported mean FGs of 23.2 dB and 33.8 dB in the BCI 601 and 602 groups, respectively. In a retrospective study of 20 adult patients implanted with the BCI 602, Brkic et al. (2023) observed good audiological outcomes with an average FG of 27.9 dB after 3 months follow-up, no intra- or postoperative complications, and demonstrated that the use of optional lifts did not adversely affect audiological outcomes or surgical time. Willenborg et al. (2023) reported on the use of the BCI 602 in 14 children, with three under the age of five, showing safe and successful implantation in all patients and improvements in hearing performance. In a multicenter study, Sprinzl et al. (2023) reported significant audiological benefit, preserved residual hearing, high rates of compliance, and low complication rates in 33 adult and pediatric patients implanted with the BCI 602, with a FG of 25.5 dB across all patients. A systematic review and meta-analysis of patients implanted with the BCI 602 by Lein et al. (2024) summarized findings across 116 adult and pediatric patients, demonstrating clinically meaningful improvements in hearing thresholds, speech perception in noise, and word recognition scores, with low post-operative complication rates (3.1% in adults, and 3.8% in children). In six studies with a total of 112 patients, all patients showed significant improvement in sound field thresholds (mean difference of −27.05 dB) postoperatively and a significant pooled improvement of word recognition scores in the aided condition (mean difference of 68.89%, n = 96). Of the studies that reported duration of surgery with the BCI 602, the time ranged from 19–90 min, with a total pooled mean duration of surgery of 41.9 min (±11.2). Most recently, Loader et al. (2025) published a retrospective review of 50 patients implanted with either the BCI 601 (n = 24) or 602 (n = 26) over a period of 8 years, showing an average sound field threshold increase of 24.3 dB, and significantly better word recognition scores at 80 dB SPL in the BCI 602 group (p = 0.0076). The authors reported two explantations in the BCI 601 cohort and one surgical wound revision in the BCI 602 cohort, without implant loss.

The mean FG reported here (38.7 ± 15.0 dB) is greater than what has been reported in the majority of the other studies of the BB-BCI. A possible reason for this finding may be the middle fossa surgical placement utilized at our center. BCDs have been shown to produce greater stimulation of the cochlea and improved hearing thresholds when positioned closer to the cochlea (Sprinzl and Wolf-Magele, 2016). This finding has been widely studied experimentally in cadavers (Eeg-Olofsson et al., 2008; Ghoncheh et al., 2022; Stenfelt and Goode, 2005). Schurzig et al. (2025) were the first to clinically demonstrate this effect in a retrospective cohort of 28 patients implanted with the BB-BCI. Postoperative CT data were used to correlate the position of the BC-FMT with maximum output using patients' audiometric data. A significant correlation was found between the distance of the BC-FMT center to the cochlea, with maximum output levels decreasing as distance increased, with a decrease of approximately 1 dB/mm. This study also showed that an increase in longitudinal distance between the BC-FMT and the external auditory canal significantly decreased maximum output levels up to 2 kHz, thereby supporting an inferior rather than superior placement of the implant relative to the external auditory canal, however further research is needed to substantiate this finding. Notably, our group previously conducted an experimental cadaveric study on the effects of BB-BCI placement on sound transmission, finding no significant difference related to implant location when the BC-FMT was placed in either the transmastoid, retrosigmoid, or middle fossa locations (Rohani et al., 2020). While the surgical and post-operative healing benefits of the middle fossa approach have been well documented, further research is needed to confirm if there is a true audiological benefit associated with this surgical approach over the other surgical placements.

Regarding safety profiles, a systematic review and meta-analysis of the BCI 601 found a minor complication rate of 7.7% over a cumulative period of reported mean follow-up of 12.7 years (mean: 11.7 months ± 4.5), and a major complication rate of 1.7% (Magele et al., 2019). A more recent systematic review found a minor complication rate of 6.1% and a major complication rate of 5.7% for the BB-BCI (Souza et al., 2022). In their systematic review of the BCI 602, Lein et al. found that 97% of implantations were not associated with postoperative complications and reported an overall minor complication rate of 4% and a rate of < 1% of surgical intervention (Lein et al., 2024). The present review identified no major complications, device explantations, or revision surgeries for either device generation. However, because the length of follow-up was longer for patients implanted with the BCI 601 (mean 103 months, range 46-149 months) than for those implanted with the BCI 602 (mean 16 months, range 6-39 months), our findings support the short- to mid-term safety profile of the 602 but are insufficient to determine its long-term safety.

The BONEBRIDGE system was the first active transcutaneous bone conduction device available on the market. In 2019, the Cochlear^TM Osia^® active transcutaneous bone conduction device (Cochlear Bone-Anchored Solutions AB, Mölnlycke, Sweden) was introduced in the United States, and the most recent system was approved by the FDA in 2023. The Osia OSI300 implant is placed subcutaneously under the skin behind the ear and uses a piezoelectric transducer anchored to the temporal bone with an osseointegrated screw (BI300 implant). Similar to the BB-BCI, the transducer generates mechanical force in response to an external sound processor which is magnetically attached to the implant receiver coil. The Osia is indicated for patients 5 years of age and older with conductive or mixed hearing loss and BC PTAs of better than or equal to 55 dB, or in patients with SSD and AC PTAs of better than or equal to 20 dB. Bilateral fitting is achievable in patients with symmetrical hearing loss, where the difference between bone conduction thresholds is less than 10 dB PTA or less than 15 dB at individual frequencies.

Studies of the Osia system have demonstrated good audiological benefit (especially in higher frequencies), safety, and user satisfaction in adults and children (Briggs et al., 2022; Cowan et al., 2024; Lau et al., 2020; Mylanus et al., 2020; Ordoñez-Ordoñez et al., 2025; You et al., 2022). In a meta-analysis of 14 studies (n = 314), Key et al. (2024) found a pooled FG for all types of hearing loss of 35 dB against unaided hearing, with a pooled FG from subgroup analysis for conductive or mixed hearing loss of 37.7 dB (n = 97), and low complication and explantation rates. In another systematic review, Jukic et al. (2024) found a weighted mean PTA gain of 35.4 dB and a mean high-frequency gain (6 kHz and above) of 16.1 dB. Kim et al. (2023) conducted a comparative study of the audiological benefits in 17 patients implanted with the Osia compared to a composite group of patients implanted with either the Baha Attract or the BB-BCI. Significant improvements in FG were reported in patients with SSD at 4 kHz for the Osia group (37.5 ± 8.9 dB; n = 8) compared to the composite group (26.9 dB ± 13.8; n = 24; p = 0.05), and in EG in patients with CHL/MHL at 2 kHz for the Osia group (11.1 ± 14.9 dB; n = 9) compared to the composite group (−2.7 ± 12.6 dB; n = 26). It is important to note, however, that the analyses that showed statistical significance combined the BAHA Attract and BB-BCI into one group, even though the BAHA is a passive device and the BB-BCI is an active device, which may have skewed the results (Baumgartner et al., 2023). In a retrospective cohort study of 14 patients, audiometric and quality-of-life outcomes were compared between the BB-BCI and the Osia, showing statistically significant audiometric improvements in mid and high frequencies with the Osia, but outperformance in speech understanding at 50% with the BB-BCI (Lorente-Piera et al., 2024). Crowder et al. (2021) examined medical device reports submitted to the Food and Drug Administration (FDA) to identify adverse events with the BB-BCI and Osia and found that while device malfunctions were more frequently associated with the BB-BCI and patient injuries more commonly associated with the Osia, the overall complication rates for both remain low. A recent meta-analysis comparing adverse events between the BB-BCI and Osia found significantly lower risks of minor events with the BB-BCI (on average 9.1 events per 100 patients per year) compared to 25.8 events per 100 patients per year with the Osia (Caversaccio et al., 2025). Similarly for major events, there were average lower risks (3.5 events per 100 patients per year) with the BB-BCI compared to the Osia (5.6 events per 100 patients per year), but this difference was not statistically significant. No significant differences were found in the risk for explantation, revision surgery, re-implantation, or permanent non-use. The authors proposed that the observed differences may be due to the size and shape variations between the device transducers, or the minimal skin displacement associated with the BB-BCI.

There are several key differences between the Osia and BB-BCI systems. The BB-BCI requires the drilling of a bony well to house the BC-FMT, whereas the Osia necessitates only bone polishing rather than removal. The smaller size of the BCI 602 does however address this limitation. The Osia system relies on screws which must be osseointegrated, whereas the BB-BCI system does not. If osseointegration fails, revision surgery or explantation may be required (Crowder et al., 2021). Lastly, the BONEBRDIGE has a wider frequency range of 250-8,000 Hz vs. 400-7,000 Hz for the Osia (Ellsperman et al., 2021).

The Sentio System (Oticon Medical AB, Askim, Sweden) is the most recently approved active transcutaneous bone conduction system on the market, receiving both FDA approval and the CE mark in 2024, and Health Canada approval in 2025. The Sentio consists of the Sentio Ti implant, which is placed under the skin, and the Sentio 1 Mini sound processor, which is held over the implant by a magnet. The Sentio uses electromagnetic stimulation like the BB-BCI. The Sentio is the smallest available transcutaneous implant, allowing for flexibility in surgical placement. In early clinical reports, the Sentio has shown good performance and safety, with a FG at 3 months of 29.3 dB (± 9.4 dB) and no serious device-related events at 6 months (n = 51) (Hol et al., 2025), however long-term data is needed.

Limitations

The present report has several limitations. Inherent to its retrospective design, our review encountered missing and incomplete data. Speech perception was not routinely measured in all patients; therefore, the results we were able to present were sparse. Furthermore, our analyses were based on pooled datasets that combined different testing intervals and protocols, which may have introduced variability. Direct comparisons between our current and previous report may be skewed as the updated patient series included pediatric cases, whereas the initial cohort comprised only adult patients. Another limitation is the primary outcome measure used to reflect efficacy in our study was FG, however evidence suggests that this measure does not accurately reflect patients' subjective hearing quality in everyday listening situations or impacts on their quality of life (Auinger et al., 2022). Future prospective studies aiming to evaluate the BB-BCI would benefit from the inclusion of subjective questionnaires, such as the Health Utility Index (HUI) (Furlong et al., 2001) or the Speech, Spatial and Qualities Questionnaire (SSQ) (Gatehouse and Noble, 2004). Moreover, since our cohort included patients with mixed and conductive hearing losses, FG may fail to reflect the true effectiveness of the device as it is heavily influenced by the pre-existing air-bone gap (Consensus Statement on Bone Conduction Devices and Active Middle Ear Implants in Conductive and Mixed Hearing Loss, 2022). To address this limitation, we stratified our results by hearing loss type and reported the aided thresholds. Effective gain (EG) was additionally calculated for the subgroup of patients with mixed hearing loss, showing a close approximation of aided thresholds to preoperative bone conduction thresholds. Additionally, audiological outcomes at our center are measured using DSL prescriptive targets rather than the maximum available gain in the device, which may further limit comparison with other studies and the interpretability of FG. FG is also subject to variability by sound transmission through the sound field speakers. Objective measures have been proposed in the literature, such as the use of a skin microphone system (Persson et al., 2025). For these reasons, outcomes reported using FG should be interpreted with caution, and future studies should prioritize more accurate and clinically meaningful measures as well as patient-reported outcome measures.

Lastly, although an independent-samples t-test indicated no significant difference between groups, the analysis was underpowered to detect small effects given the small sample size in each group. As such, this result should be interpreted with caution, and future comparative studies with larger samples are needed to assess the effect of BB-BCI device generation on hearing outcomes.

Conclusion

The middle fossa approach for BONEBRIDGE implantation demonstrates favorable audiological benefit and an excellent safety profile. This study reports an institutional experience implanting both the BCI 601 and BCI 602 models, spanning 12 years. No statistically significant difference was observed between device generations, supporting the ongoing use of this technique for appropriately selected patients. Speech perception improved postoperatively, and no intraoperative complications or revision surgeries were reported. A mean functional gain of 38.7 ± 15.0 dB was reported across all patients, consistent with previously published literature of the BCI 601, 602, and other active bone conduction implants.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: William E. Hodgetts, University of Alberta, Canada

Reviewed by: Mohammad Ghoncheh, Hannover Medical School, Germany

Byanka Cagnacci Buzo, Santa Casa of Sao Paulo, Brazil