Gochlear implants from model to patients
Briaire, J.J.
Citation
Briaire, J. J. (2008, November 11). Gochlear implants from model to patients. Retrieved from https://hdl.handle.net/1887/13251
Version: Corrected Publisher’s Version
License: Licence agreement concerning inclusion of doctoral thesis in the Institutional Repository of the University of Leiden
Downloaded from: https://hdl.handle.net/1887/13251
Note: To cite this publication please use the final published version (if applicable).
Cochlear Implants: From Model to Patients
Cochlear implants (CI) are by now an accepted form of rehabilitation for pro- foundly deaf patients. CI users regain part of their hearing by direct electrical stimulation of the auditory nerve. With modern cochlear implants most users are able to achieve open-set speech understanding and are able to use the telephone. There are, however, still a lot of unanswered questions regard- ing the optimal design, stimulation paradigms, fitting methods and objective measurements. With the development of a realistic computer model of the implanted cochlea, as described in this thesis, these questions are analyzed from a fundamental perspective. This realistic model enables the analysis of clinical devices and gives insight in discrepancies between human and animal results. Insights gained from the model are used to improve clinical practice.
Based on the model outcomes presented the characteristics of an improved electrode design were defined, and finally tested in a temporal bone study.
Chapter 1 presented the basic principles of a cochlear implant and a histori- cal overview of the development of this device, from experimental devices to well accepted commercial products. At the end of this chapter the thesis was outlined.
Chapter 2 described the basic principle of modeling cochlear implants with a two step model. The first step is the modeling of the electrical conduc- tion through the cochlea, also known as the volume conduction problem. The second step is to model the behavior of the nerve fibers in response to the potential distribution calculated in the first step. This two step model was used throughout this thesis. The newly introduced spiral shaped cochlea model allows for the prediction of excitation thresholds and spatial selectivity in the implanted cochlea. The model outcomes describing differences between exci- tation areas for various electrode locations, were compared with electrophysi- ological experiments. It was concluded they showed good agreement.
In chapter 3 a detailed description was given of the volume conduction model with an explanation of how the spiral shaped cochlear meshes and implant designs used in this thesis were generated. The meshes have the flexibility to investigate clinically relevant issues. With this model of the guinea pig cochlea the insulating effects of the membranes surrounding the scala tympani have been investigated. The described mesh generating software has the flexibility to be used for the much more challenging creation of the human cochlea.
Chapter 4 described the potential distributions and current pathways in a spi- ral cochlear model as described in the previous chapters. The relatively well conducting scala tympani turned out to be the main one indeed, but the expo- nential decay (J ∼ e−z) of current was only a good description of the far-field 270
behavior. In the vicinity of the electrodes, i.e. near the fibers that are most easily excited, higher current densities were found, that were best described by a spherical spread of the current (J ∼ R12). We concluded that the current spread along the scala tympani and its dependence on the position of the cur- rent source in the cochlea is well described by the superposition of spherical and exponential decay.
In chapter 5 a comparison was made between the outcomes of a guinea pig computer model and a realistic model of the human cochlea, both implanted with a model of a HiFocus cochlear implant. Taking into account the large anatomical differences in size, location in the temporal bone and overall geom- etry of the cochlea and auditory nerve between both species. It turned out that a well-designed modiolus-hugging electrode yields reduced current thresholds and a high spatial selectivity without a reduction of the dynamic range. How- ever, in the second turn of the human cochlea the outcome was less favorable:
As in the guinea pig, cross-turn stimulation reduces the dynamic range in the perimodiolar position substantially. We concluded that the clinical success of cochlear implantation in man and the promising results with modiolus-hugging devices depend largely upon typically human cochlear anatomy.
The clinical evaluation of the HiFocus electrode array with electrode position- ing system combined with the CII implant electronics is described in chapter 6. The speech perception scores on CVC words without lip reading were monitored prospectively for the ten postlingually deafened patients. After one week all patients but one were able to use the telephone functionally. At the end of the study (follow-up 3 to 11 mo) the average CVC phoneme score was 84% (word score 66%). The Phoneme Recognition Threshold (50% of the performance in silence) was reached for Signal-to-Noise ratios between 0 and 5dB. NRI recordings were obtained with both the alternating polarity and the forward masking paradigm. The NRI system, although somewhat slow, gives good results and offers the ability to measure I/O curves and neural tuning.
The study presented in chapter 7 evaluated the effect of the duration of post- operative follow-up on the value of pre-operative predictors. The performance outcomes of a group of 91 subjects, implanted between 2000 and 2005 with a HiRes90K or CII implant with a HiFocus I electrode array, were evaluated with univariate linear regression analysis and with multiple regression anal- ysis based on the Iowa predictive model. The difference in learning curve between patients with a long and a short duration of deafness or with dif- ferences in implantation age introduces a dependence on the post-operative
Cochlear Implants: From Model to Patients
experience on the relative importance of predictive factors. Therefore, predic- tive models should be based on follow-up times of at least 2 years to allow poorer performers approach their ultimate performance level. When dealing with non-controlled retrospective studies, multiple regression analysis should be used to extract the influence of, for instance, electrode array design, or age at implantation, on the speech understanding scores; while reducing the effect of other parameters such as duration of deafness.
In chapter 8 the extension of a the human cochlea model, incorporating record- ing capabilities, is described. This model is used to investigate the individual single fibre action potential (SFAP) contribution to the eCAP at various stimu- lation levels. The single fiber contributions indicated that at high current levels the fibers located centrally in the excitation area, close to the stimulating elec- trode, yield an atypical response, without a clear negative peak in the SFAP.
The number of fibers with atypical responses also increased with stimulus level. Therefore, the overall eCAP amplitude decreases above a certain stim- ulus level. This phenomenon was recorded earlier, but was always regarded as a recording artifact.
Chapter 9 described an extensive model study on the consequences of the precise location of the electrode array in the scala tympani. The objective of this chapter was to find the optimal placement with respect to threshold, dy- namic range and spatial selectivity for both, degenerated and non-degenerated cochleae. The model predicted reduced threshold, increased dynamic range and higher spatial selectivity for the peri-modiolar position at the basal end of the cochlea, with minimal influence of neural degeneration. At the apical end of the array (1.5 cochlear turns), the dynamic range and the spatial selectivity were limited due to the occurrence of cross-turn stimulation, with the exception of the condition without neural degeneration and with the electrode array along the lateral wall of scala tympani. The benefits of eCAP recordings with respect to intra-operatived placement optimisation were investigated. The eCAP sim- ulations indicated that a largeP0peak occurs before theN1P1complex when the fibers are not degenerated. The absence of this peak might be used as an indicator for neural degeneration.
Based on the design criteria described in chapter 9 a design change of the current HiFocus electrode array has been proposed which should have a pe- rimodiolar position of the electrode array for the basal contacts, while a lat- eral position should be achieved for more apical ones. The initial temporal bone study with prototype electrodes meeting these criteria was described in chapter 10. The prototype electrodes were inserted in human temporal
272
bones and the position of the contacts was verified with high-resolution CT- scans prior to a careful dissection, documenting the insertion trauma. The new electrode was able to attain the desired position with minimal damage to the intracochlear structures.
In chaper 11 the capabilities of the current model are described as well as future steps needed for the creation of a patient specific model with direct clinical implications for the individual patient. Some of the ongoning develop- ments leading to a new generation of cochlear implants are highlighted. It was concluded that computational modeling can form a theoretical guideline along which new experiments and new developments for cochlear implants can be directed.
