Objectives: Real-time electrocochleography (ECochG) has been used as a monitoring tool during cochlear implantation (CI), whereby, amplitude drops have been correlated with postoperative acoustic hearing results. However, no consensus has been reached as to how a single event of an amplitude drop should be characterized. The aim of this study was to identify ECochG events that predict loss of hearing 1 month after surgery. Design: Fifty-five patients were included in this prospective cohort study. Real-time ECochG measurements were performed during CI electrode insertion. Single ECochG events were characterized according to their amplitude loss and slope steepness. Results: Using receiver operating characteristic analyses, the most efficient cut-off criterion for a relative hearing loss of 25% was an amplitude loss of 61% at a fixed slope steepness of 0.2 µV/sec. Three-quarters of our population had at least one such event during implantation. Most events occurred shortly before full insertion. With increasing number of events, median residual hearing thresholds deteriorated for all frequencies. Larger amplitude drops trended toward worse hearing preservation. Signal recovery after an ECochG event could not be correlated to acoustic hearing outcomes. Conclusions: Our data suggest that amplitude drops exceeding 61% of the ongoing signal at a slope steepness of 0.2 µV/sec are correlated with worse acoustic hearing preservation. Clearly defined ECochG events have the potential to guide surgeons during CI in the future. This is essential if a fully automated data analysis is to be employed or benchmarking undertaken.
10 June 2021 in IEEE Transactions on Biomedical Engineering
Objective: The insertion of the electrode array is a critical step in cochlear implantation. Herein we comprehensively investigate the impact of the alignment angle and feed-forward speed on deep insertions in artificial scala tympani models with accurate macro-anatomy and controlled frictional properties. Methods: Motorized insertions (n=1033) were performed in six scala tympani models with varying speeds and alignment angles. We evaluated reaction forces and micrographs of the insertion process and developed a mathematical model to estimate the normal force distribution along the electrode arrays. Results: Insertions parallel to the cochlear base significantly reduce insertion energies and lead to smoother array movement. Non-constant insertion speeds allow to reduce insertion forces for a fixed total insertion time compared to a constant feed rate. Conclusion: In cochlear implantation, smoothness and peak forces can be reduced with alignment angles parallel to the scala tympani centerline and with non-constant feed-forward speed profiles. Significance: Our results may help to provide clinical guidelines and improve surgical tools for manual and automated cochlear implantation.
Sound intensity is a key feature of auditory signals. A profound understanding of cortical processing of this feature is therefore highly desirable. This study investigates whether cortical functional near-infrared spectroscopy (fNIRS) signals reflect sound intensity changes and where on the brain cortex maximal intensity-dependent activations are located. The fNIRS technique is particularly suitable for this kind of hearing study, as it runs silently. Twenty-three normal hearing subjects were included and actively participated in a counterbalanced block design task. Four intensity levels of a modulated noise stimulus with long-term spectrum and modulation characteristics similar to speech were applied, evenly spaced from 15 to 90 dB SPL. Signals from auditory processing cortical fields were derived from a montage of 16 optodes on each side of the head. Results showed that fNIRS responses originating from auditory processing areas are highly dependent on sound intensity level: higher stimulation levels led to higher concentration changes. Caudal and rostral channels showed different waveform morphologies, reflecting specific cortical signal processing of the stimulus. Channels overlying the supramarginal and caudal superior temporal gyrus evoked a phasic response, whereas channels over Broca’s area showed a broad tonic pattern. This data set can serve as a foundation for future auditory fNIRS research to develop the technique as a hearing assessment tool in the normal hearing and hearing-impaired populations.
Objectives: Real-time electrocochleography (rt-ECochG) is a method to detect intracochlear potential changes during cochlear implantation (CI). Steep amplitude drops of the cochlear microphonic (CM) signal (so called “ECochG events”) have been correlated with worse residual hearing outcomes. However, the sensitivity and specificity of monitoring CM amplitude on its own are too low to use it as a biomarker. The aim of this article was to establish if additional signal components would help to better predict postoperative hearing outcomes. Design and Setting: Single-center, prospective cohort study at a tertiary referral hospital. Participants and Interventions: Between 2017 and 2020, we included 73 adult patients receiving a lateral wall cochlear implant electrode. During electrode insertion, rt-ECochG measurements were performed. Main Outcomes: We calculated a multiple regression analysis for patients with one ECochG event. The dependant variable was the relative acoustic hearing result 4 weeks after surgery. Independent variables were CM latency, a ratio of the auditory nerve neurophonic to the CM (the ANN/CM index) as well as CM signal recovery. Results: The change of the ANN/CM index linearly correlated with acoustic hearing outcomes 4 weeks after surgery. Adding this factor led to a statistically significant increase in the variance accounted for by the regression model. Conclusions: When monitoring the implantation process with rt-ECochG, prediction of postoperative hearing thresholds is improved by addition of the ANN/CM index to a model that includes CM amplitude fluctuation.