Bimodal stimulation in children with inner ear malformation: One side cochlear implant and contralateral auditory brainstem implant

To determine audiological outcomes of children who use a cochlear implant (CI) in one ear and an auditory brainstem implant (ABI) in the contralateral ear.

reports showing the benefits of CI in IEM, CI became an accepted procedure in this patient population. 3 However, when the IEM are severe and occur with cochlear nerve (CN) or cochleovestibular nerve (CVN) deficiency or hypoplasia, the clinicians are faced with the dilemma of deciding between CI and an auditory brainstem implantation (ABI).
Paediatric ABI indications were clearly defined in two groups in the first ABI consensus meeting. The first group was definite congenital indications, including complete labyrinthine aplasia (Michel aplasia), cochlear aplasia, CN aplasia and cochlear aperture aplasia.
The second group was possible congenital indications, including hypoplastic cochlea with cochlear aperture hypoplasia, common cavity and incomplete partition type I cases with or without CN, the presence of a common CVN and the presence of hypoplastic CN. 4 Patients with hypoplastic CNs or thin, unbranched CVNs constitute the most controversial group in the CI and ABI decision. 5 Even in patients with confirmed CN hypoplasia or aplasia on magnetic resonance imaging (MRI), some auditory responses could be observed on certain frequencies during audiological evaluation in one or both ears, which is compatible with severe hearing loss. 5,6 Cochlear implant users with hypoplastic CN or CVN have only a limited benefit and lag behind the cochlear implant users with normal CN or CVN. 5,7 Another amplification option for these types of patients is the ABI, which improves environmental sound awareness, speech detection and language skills. 5,8 Thus, for cases with limited benefit from CI, the application of contralateral ABI can be considered.
The first report of CI with a contralateral ABI was presented by clinic. In the current paper, we report the audiological findings of 12 paediatric bimodal implant users with both cochlear implants and auditory brainstem implants.

| Ethical considerations
The study was authorised by the Hacettepe University Non-Interventional Clinical Research Ethics Board (GO 18/437).

| Participants
Twenty-five children with bilateral prelingual profound SNHL were implanted with a cochlear implant on one side and an ABI on the other side at the Hacettepe University Department of Otorhinolaryngology. Six subjects were implanted with a cochlear implant and an ABI simultaneously. One subject was implanted with an ABI initially and with a cochlear implant at a later date due to a reimbursement issue. The remaining 18 subjects were implanted with a cochlear implant initially and later with an ABI. Six out of the eighteen were excluded, one due to postoperative neurological problems and the remaining five due to their limited experience with ABIs. The remaining 12 children with severe IEM and bilateral profound hearing loss were included in this study. Simultaneous CI and ABI results have been submitted separately as another paper.
All subjects completed medical, audiological, and speech and language assessments, as well as temporal bone imaging preoperatively.
The preoperative audiological evaluation was composed of auditory brainstem responses (ABR) and behavioural testing with and without hearing aids. The hearing thresholds were assessed preoperatively using inserted earphones and hearing aids in free field with age-appropriate behavioural methods, such as visual reinforcement audiometry or play audiometry. The CI side was determined according to the results of behavioural responses and the status of the CN on magnetic resonance imaging (MRI). CI was planned in the ear with better sound detection during behavioural testing with inserted ear phones and with better CN as seen on MRI. Clinical features of the subjects are given in Table 1.
All 12 subjects (five male and seven female) were initially implanted with a cochlear implant and then subsequently implanted with an ABI between January 2013 and November 2016 due to their limited progress in auditory perception skills. The preoperative audiological test results were not reported in the current study, and only audiological performance with the CI and the ABI was retrospectively reviewed in this study. The data presented in this study were collected between January 2013 and July 2018.

| Intraoperative and postoperative audiological measurements
No major intraoperative complications were encountered during both CI and ABI surgeries. Initial activation of the cochlear implant was

Keypoints
• CN hypoplasia presents a challenge in the decision-making process concerning the choice of CI, ABI, or CI and ABI together.
• Depending on the audiological and radiological results, bilateral stimulation should be advised for children with IEM and CN hypoplasia.
• In cases of hardly visible cochlear nerve, we have observed that children implanted with CI and contralateral ABI showed better performance when compared to either device alone.
• Depending on our clinical experience contralateral ABI should be done within 12-18 months after CI. performed 2-4 weeks after CI surgery in all subjects. The programming parameters were selected as default at initial fitting, and the minimum duration levels and pulse width values were increased during the follow-up visits due to hypoplastic CN. After CI, all subjects were followed by the same audiologists. Due to the limited auditory and speech progress with the cochlear implant, ABI was performed on the contralateral ear in all subjects. The limited progress with cochlear implants was evaluated by the experienced auditory implant team during the follow-up visits through auditory performance and improvement in auditory perception skills. The electrical ABR was used intraoperatively during the ABI surgery to evaluate the placement of the ABI electrode. The initial activation of the ABI was executed four weeks after surgery with monitoring of the vital functions.
Electrical impedances were measured in every programming session.
Programming parameters of the ABI side and auditory/non-auditory sensations were recorded during each session. The follow-up programming visits were planned every two to three months in the first two years and every three months after the second year.

| Evaluation of the auditory perception skills
Auditory perception tests used during the follow-up period contained the following test battery: Meaningful Auditory Integration

Scale (MAIS), pattern perception test, word recognition test, Speech
Intelligence Rating (SIR) and Category of Auditory Performance (CAP) scale. The MAIS is a parent-reported questionnaire that assesses the listening skills in children with hearing loss. Each item was rated together with parents and scored from 0 to 4 (0 = never, 1 = rarely, 2 = occasionally, 3 = frequently and 4 = always), with a total score of 0 to 40. 11 The CAP is a rating scale from 0 (no awareness of sound) to 7 (the use of telephone with a familiar speaker) in order to evaluate the hearing outcomes. 12 The SIR measures the speech intelligibility and shows the overall progress in speech over time with a rating of 1 (unintelligible speech with a manual primary mode of communication) to 5 (understandable speech to all listeners). 13 The pattern perception and word recognition tests were applied from the Children's Auditory Perception Skills Test in Turkish (CIAT) test battery. 8 The pattern perception and word recognition tests were applied verbally in three different conditions: ABI only, CI only and bimodal condition. All auditory perception tests were administered by the same experienced audiologist with a live voice during the follow-up visits.

| Demographics
Mean age of the subjects was 84.00 ± 33.94 months (range 60-108 months). Age at CI surgery was 25 ± 10.98 months (range 14-49 months), and age at ABI surgery was 41.50 ± 16.14 months (range 21-80 months). Mean duration between CI and implantation of the ABI was 16 Two of the subjects were reimplanted due to facial nerve stimulation (FNS) (Subject #6) or device failure (Subject #3) in the CI side.
One subject (Subject #8) was reimplanted with another implant from a different company at the request of the family due to insufficient performance with the previous implant in the ABI side.
Except for one subject, all subjects used both CI and ABI sound processors regularly. Subject #6 refused to use the CI processor after ABI surgery, complaining of not hearing with CI as well as ABI.

| Intraoperative testing and audiological performance
Electrical ABR was performed intraoperatively during ABI surgery in all subjects. The eABR results and programming parameters of the subjects are given in Table 2. Programming parameters were selected as default for strategy, rate and pulse width/duration at the initial fitting session of the ABI. For Cochlear™ ABI systems, the ini- Hearing thresholds with CI only and ABI only are given in Table 3.
Thresholds with CI only were not assessed for Subject #6 due to the subject not using the CI processor and were reported as "not applicable" (NA) in Table 3. Subject #11 was diagnosed with Down syndrome, and since it was not possible to evaluate thresholds with CI and all thresholds with ABI due to her additional disability, this is reported as NA in Table 3.

| Non-auditory sensations in ABI side
The mean number of active electrodes was found 67.62 ± 17.49% (between 33.33% and 93.33%). The three causes of electrode deactivation were defined as non-auditory sensations, inadequate auditory stimulation and impedance/voltage problems. Non-auditory sensations were not observed in 1 (Subject #8) out of 12 children.
The observed non-auditory sensations were FNS (58.3%), balance problems (8.3%), gag reflex (8.3%), and shoulder (25%) or neck pain (8.3%) in the remaining subjects. In 2 out of 12 children, more than 1 non-auditory sensation was observed, and the electrode was deactivated. after ABI (Figure 1). There was a statistical improvement in CAP scores with bimodal stimulation (P = .002). The speech intelligibility of all subjects advanced rapidly with the use of both CI and ABI (P = .001).

| Auditory perception skills before and after ABI
Pattern perception and word recognition scores were signifi-  Table 4, pattern perception scores increased in the bimodal condition, even when the scores were significantly lower in both CI only (P = .017) and ABI only (P = .018) conditions.
Word recognition scores were also significantly higher in the bimodal condition when compared to CI only (P = .012) and ABI only (P = .008) conditions. Despite pattern perception scores not being significantly different (P = .680), the word recognition scores were significantly higher (P = .027) in the CI only condition than in the ABI only condition.

| Factors influencing auditory performance with ABI
It was found that hearing performance with ABI was negatively cor-

F I G U R E 1 Category of Auditory Performance and Speech Intelligibility
Rating scores before and after ABI  Another remedy for these patients may be the bilateral application of CI. In fact, it was reported that bilateral CI provides more benefit for users with CN hypoplasia compared to unilateral CI in CVN deficiency. 18 Considering the limited benefit of CI alone for these patients, ABI could provide better outcomes. An option in this case may be the removal of the CI electrode and ABI application to malformations. 19 Improvement in language skills after the application of ABI for CI users with inner ear malformations may be related to adequate stimulation of areas in higher auditory pathways, which were not optimally stimulated during CI usage. This improvement could be increased in time with regular use and result in more favourable outcomes for users. It was also mentioned in the second consensus meeting that hearing thresholds with ABI were between 30 and 60 dB in most patients with ABI. 19 In our case series, the hearing thresholds with ABI were found to be similar to the ABI consensus paper. In our paper presenting the long-term results of the ABI, it was found that better thresholds were associated with better CAP scores. 20 Similar to the children with unilateral ABI, better hearing performance with ABI resulted in higher bimodal pattern perception scores for children with CI and contralateral ABI.
Another positive prognostic factor affecting the auditory performance with ABI was defined as chronological age and age at ABI surgery in the present study. Age at ABI activation was also found to be one of the most important prognostic factors in the same paper. 19 In the study of Peng et al, 9  The CI and contralateral ABI can be used together safely and synergistically. After ABI surgery, a relatively rare side effect was observed for Subject #2. Three days after the ABI surgery, when the patient activated his CI device for the first time since the ABI surgery, he had an excessive FNS on the CI side. This was remedied by decreasing the C-levels globally on the CI side. Later, after initial activation of the ABI, the same FNS was observed both with individual electrode stimulation and with live voice through CI. In order to remove this FNS on the CI side, the pulse width of the stimulation was increased from 50 to 88, and C-levels were decreased according to his responses. A possible reason for this problem may be the increased impedance values on the CI electrode after ABI surgery.

| Study limitations
Even though the present study is the first and therefore the largest study of children with IEM using CI and contralateral ABI, the study sample is still too small to conclusively show the effect of the bimodal stimulation with CI and ABI. The present study is also limited by the heterogeneity of the study population and reported test results. Only the auditory performance data from before ABI and the last follow-up visit were included in this study. In future studies, long-term results in large cohorts can be presented during the follow-up visits.

CO N FLI C T O F I NTE R E S T
All authors declare that they have no conflict of interest.