Intraoperative EABR Testing Predicts Strength of Cochlear Implant Stimulation Optimized After Long-Term Use in Pediatric Malformation Ears.

OBJECTIVE: This study focused on the intensities of cochlear implant (CI) stimulation in pediatric CI users with inner ear malformation or cochlear nerve deficiency (CND). In this population, CI programming is difficult because a large intensity of CI stimulation is required to achieve sufficient hearing, but the excess CI stimuli often induce facial nerve stimulation. We aimed to assess whether the results of intraoperative electrically evoked auditory brainstem responses (EABRs) testing predict maximum current levels of CI stimuli (cC levels) optimized by a behavioral-based method after long-term CI use. STUDY DESIGN: A retrospective case review. SETTING: A tertiary referral CI center. PATIENTS: A total of 116 ears with malformations (malformation group) and 63 control ears (control group) from patients younger than 18 years who received CI. The malformation group comprised 23 ears with a common cavity (CC), 26 with incomplete partition type 1 (IP-1), 26 with incomplete partition type 2 (IP-2), and 41 with CND. INTERVENTIONS: Diagnostic. MAIN OUTCOME MEASURES: Correlation between intraoperative EABR results and cC levels determined by the behavioral-based CI programming after long-term CI use. RESULTS: The CC, IP-1, and CND ears required significantly larger cC levels than the IP-2 ears and control groups. However, the cC levels increased to reach the plateau 1 year after surgery in all groups. Among the malformation group, 79 ears underwent intraoperative EABR testing. Greater than 80% of the CC, IP-1, and IP-2 ears and 54.8% of the CND ears exhibited evoked wave V (eV) and were included in the eV-positive category. Myogenic responses but no eV were observed in 18.2, 15.0, and 35.5% of the CC, IP-1, and CND ears, defined as the myogenic category. No eV or myogenic response was elicited in 9.7% of the CND ears. We focused on minimum current levels that elicited eV (eV levels) in the eV-positive category and maximum current levels that did not elicit any myogenic responses (myogenic levels) in the myogenic category. A significant relationship was detected between the eV levels and the cC levels. When analyzed in each malformation type, the eV levels significantly correlate with the cC levels in the CC and CND ears but not in the IP-1 and IP-2 ears, probably because of slight variation within the IP-1 group and the small number of the IP-2 group. The myogenic category did not show a significant relationship between the myogenic levels and cC levels, but the cC levels were similar to or smaller than the myogenic levels in most ears. CONCLUSIONS: This study confirmed that intraoperative EABR testing helps predict the optimal cC levels in malformation ears. EABR-based CI programming immediately after cochlear implantation, followed by behavioral-based CI programming, may allow us to achieve early postoperative optimization of CI maps even in young children with severe malformations.

Audio-visual speech perception in prelingually deafened Japanese children following sequential bilateral cochlear implantation.

OBJECTIVES: An effect of audio-visual (AV) integration is observed when the auditory and visual stimuli are incongruent (the McGurk effect). In general, AV integration is helpful especially in subjects wearing hearing aids or cochlear implants (CIs). However, the influence of AV integration on spoken word recognition in individuals with bilateral CIs (Bi-CIs) has not been fully investigated so far. In this study, we investigated AV integration in children with Bi-CIs. METHODS: The study sample included thirty one prelingually deafened children who underwent sequential bilateral cochlear implantation. We assessed their responses to congruent and incongruent AV stimuli with three CI-listening modes: only the 1st CI, only the 2nd CI, and Bi-CIs. The responses were assessed in the whole group as well as in two sub-groups: a proficient group (syllable intelligibility ≥80% with the 1st CI) and a non-proficient group (syllable intelligibility < 80% with the 1st CI). RESULTS: We found evidence of the McGurk effect in each of the three CI-listening modes. AV integration responses were observed in a subset of incongruent AV stimuli, and the patterns observed with the 1st CI and with Bi-CIs were similar. In the proficient group, the responses with the 2nd CI were not significantly different from those with the 1st CI whereas in the non-proficient group the responses with the 2nd CI were driven by visual stimuli more than those with the 1st CI. CONCLUSION: Our results suggested that prelingually deafened Japanese children who underwent sequential bilateral cochlear implantation exhibit AV integration abilities, both in monaural listening as well as in binaural listening. We also observed a higher influence of visual stimuli on speech perception with the 2nd CI in the non-proficient group, suggesting that Bi-CIs listeners with poorer speech recognition rely on visual information more compared to the proficient subjects to compensate for poorer auditory input. Nevertheless, poorer quality auditory input with the 2nd CI did not interfere with AV integration with binaural listening (with Bi-CIs). Overall, the findings of this study might be used to inform future research to identify the best strategies for speech training using AV integration effectively in prelingually deafened children.

SLC26A4 p.Thr410Met homozygous mutation in a patient with a cystic cochlea and an enlarged vestibular aqueduct showing characteristic features of incomplete partition type I and II.

Mutations of SLC26A4 are associated with incomplete partition type II (IP-II) and isolated enlargement of the vestibular aqueduct (EVA). We experienced a congenitally deaf 6-year-old boy with a rare p.Thr410Met homozygous mutation in SLC26A4 who underwent bilateral cochlear implantation. He had bilateral inner ear malformation, in which the dilated vestibule and EVA were identical to those in IP-II, but the cochlea lacking a bony modiolus resembled that in incomplete partition type I. These results suggest that homozygous mutations in SLC26A4 are always associated with EVA, while the severity of cochlear malformation may vary depending on the type of SLC26A4 mutation.