Proefschrift_vd_Beek

led to deeper insertion. Although the intracochlear conductivity paths of the two groups did not show significant differences, a basal current drain was seen for the shallowly inserted non-positioner patients. It was concluded that a basally perimodiolar electrode design benefits speech perception. The combination of decreased distance from the modiolus, increased insertion depth, and the insulating properties of the electrode array have functional implications for the clinical outcomes associated with the perimodiolar electrode design. Further research is needed to elucidate these factors’ individual contributions to those outcomes. Chapter 4 focuses on the application of the cochlear implant’s ability to record the electrically evoked action potentials (eCAP) of the neurons in the cochlea to measure the effectiveness of the electrode-to- neural interface. The study investigated the spread of excitation (SOE) profiles using eCAP measures and analyzed the effects of various parameter settings. Measurements were performed intra-operatively in 31 users of the Advanced Bionics HiRes 90K cochlear implant. SOE was measured using the forward masking technique (selectivity) as well as with a “fixed stimulus, variable recording” (scanning) technique. SOE profiles were produced at three stimulus levels and at three sites along the array. Additionally, the effects of the position of the recording electrodes and artefact rejection methods were studied in five subjects. All data were analyzed using linear mixed models. The selectivity method produced narrower excitation profiles than the scanning method, showing asymmetry along the array with broader SOE apically. Moreover, the position of the recording electrode shifted the SOE curves towards the recording contact, enhancing asymmetry. Neither significant effects of the current level nor artefact rejection methods were observed, nor was any significant correlation with speech perception found. Chapter 5 reports a study that analyzed the predictability of fitting levels for individual cochlear implant recipients based on a review of cohort data. The data included the threshold levels (T-levels) and maximum comfort levels (M-levels) of 151 adult subjects who used a CII/HiRes 90K cochlear implant with a HiFocus 1/1 J electrode. The 10th, 25th, 50th, 75th and 90th percentiles of the T- and M-levels are reported. The subjects’ speech perception was measured using a HiRes speech coding strategy during routine clinical follow-up. The T-levels for most subjects were between 20 and 35% of their M-levels and were rarely (< 1/50) below 10% of the M-levels (which is the manufacturer’s default). Furthermore, both the T- and M-levels increased over the first year of follow-up. Interestingly, the levels expressed in linear charge units showed a clear increase in dynamic range (DR) over 1 year (29.8 CU; SD 73.0), whereas the DR expressed in decibels remained stable. The T-level and DR were the only fitting parameters for which a significant correlation with speech perception (r = 0.34, p < 0.01, and r = 0.33, p < 0.01, respectively) could be demonstrated. Additionally, the T- and M-level profiles expressed in decibels turned out to be independent of the subjects’ across-site mean levels, as demonstrated with mixed linear models. Based on the data set from 151 subjects, clinically applicable predictive models for the T- and M-levels of all separate electrode contacts were obtained. These closed-set formulae allow the close approximation of individual recipients’ T- and M-levels based on just one psychophysical measurement. Additionally, the analyzed fitting level data can serve as a reference for future patients.

150 | Summary