Proefschrift_vd_Beek

by anodal and cathodal stimuli (Klop et al, 2004). This effect of averaging latencies has a noticeable effect on individual eCAP amplitudes and waveforms, but evidently has limited effect when relative amplitudes are compared in SOE measurements. This limited effect on SOE measures may make scanning data using different artefact rejecting methods comparable in future research. However, for a final answer with respect to the comparability of the artefact rejections methods further research is indicated. The large range in age and duration of deafness suggests that at least some difference in surviving nerve fibers would be present among the patients, which, in turn, could affect the SOE results. As these parameters turned out not to correlate significantly with our data (age, p = 0.8; duration of deafness, p = 0.5), it is concluded that they do not play a major role in explaining differences in SOE found in this study. CONCLUSIONS The main conclusions from our analysis of 31 subjects are that broader excitation profiles are measured using the scanning method compared to the forward masking technique. Secondly, along the electrode array an asymmetric SOE is seen, with wider spread apically. Thirdly, at high stimulation levels no clear effect of level on SOE was observed. Additionally, in this study no correlation between the width of SOE curves and speech understanding could be seen. With respect to the recording parameters, even in a subgroup of five patients it was possible to demonstrate that the choice for position of the recording contact influences the measurement of the SOE, shifting the curves towards the recording contact and enhancing measured asymmetry. Furthermore, a relative robustness of measurements was indicated by the facts that (1) for the selectivity measures no significant effect of the choice of fixed or varying probe was observed, and (2) comparable data could be obtained with different artefact rejection routines for scanning. However, the latter conclusions were based on the analysis of subgroups of just five patients each. To make a definitive statement on these two issues, a further study in larger patients groups is warranted. ACKNOWLEDGEMENTS The authors like to thank Ron Wolterbeek for statistical advice, Paul Boyd for language editing, Peter-Paul Boermans for ongoing support, and Advanced Bionics Corp., Valencia, California, USA and the Heinsius Houbolt Fund, The Netherlands for financial support. The data were presented under the title ‘Effects of parameter manipulations on spread of excitation measured with electricallyevoked compound action potentials’ at Objective Measures in Auditory Implants: 6th International Symposium, September 23, 2010, St Louis, Missouri, USA.

86 | Chapter 4