gms | German Medical Science

Introduction: Electrically evoked auditory brainstem response (eABR) is an objective measurement on cochlear implant (CI) users, where trains of single-pulses are used as stimuli. It is known that thresholds (THR) depend not only on the stimulation amplitude but also on stimulation rate and the duration of the stimuli. When stimulating with single pulses, eABR THRs show strong correlation with behavioural THRs. However, for high-rate stimulation, such as used for clinical purposes, the correlation is weaker, because temporal effects are missing when using low-rate stimulation. In this study, we employed multi-pulse trains as stimuli to partially compensate the effects of fast temporal processes induced by high-rate stimulation.

Material and Methods: Six CI users (10 ears, 2 male) with Med-El implants participated in this study. Multi-pulse trains of 1-, 2-, 4-, 8-, and 16-pulses with a repetition rate of 37 stimulation blocks-per-second were delivered to a medial electrode. Individual pulses in each block were 45 µs/phase symmetric biphasic pulses with 2.1 µs inter-phase gap. First, subjects indicated the psychophysical THR and MCL in each condition. Clinical THRs and MCLs were also measure in response to stimuli of rate of 1,000 pulses-per-second (pps). eABR amplitude growth functions (AGF) were then measured for different number of pulses, starting from a level of 95% of the DR down to a point where eABR waves amplitudes were just above the noise floor. Amplitudes of waves III and V were used for analysis. Processing of raw data included epoching, artefact rejection, curve fitting, filtering and averaging. An exponential curve was fitted to the AGF for all multi-pulse conditions. eABR threshold (eABR THR) was defined where the extrapolated fitted curve was zero.

Results: The morphology of eABR did not change in multi-pulse conditions. Moreover, the AGF in multi-pulse conditions looked similar as in single-pulse (conventional) eABR. Over all subjects, the median correlation between estimated eABR THR and psychophysical THR was 0.79 and 0.78 for waves III and V, respectively. Generally, the estimated eABR THR decreased with increasing number of pulses. Over all subjects, the estimated eABR THRs for 1- and 2-pulse conditions were still well-above clinical THR (wave V analysis; median difference = 147 and 101 µA, respectively). For the 4-, 8-, and 16-pulse conditions, the estimated eABR THRs were much closer to the clinical THR (wave V analysis; median difference = -22, 4, and 12 µA, respectively). Results of this study showed that, although the inter-subject variability was large, the THRs estimated by multi-pulse conditions were close to the clinical THRs. Therefore, it might be possible to use the estimated eABR THRs values as CTHR for CI fitting in people who cannot cooperate, e.g. babies and young children.