Direct Electrical Stimulation of the Cochlear Nucleus: Surface vs. Penetrating Stimulation

Prosthetic stimulation of the cochlear nucleus (CN) has been used for rehabilitation of profoundly deaf patients who are not suitable candidates for cochlear implants. The goal of this article was to assess the relative effectiveness of surface vs. penetrating stimulation of the CN. Electrophysiologic and autoradiographic measures were used to study central auditory system activation elicited by direct stimulation of the CN. Eighteen pigmented guinea pigs, divided into three groups, underwent acute implantation of bipolar electrodes in the CN. One group was not stimulated and acted as a control (n = 7). Electrodes were placed on the surface of the CN in one test group (n = 4) and within the CN in a second test group (n = 7). Thresholds for electrically evoked middle latency responses (EMLR) were determined and input/output (I/O) functions were obtained. The two test groups were then pulsed with [14C]-2-Deoxyglucose (2-DG) intramuscularly and stimulated for 1 hour with biphasic; charge-balanced pulses having a total duration of 400 μsec, a repetition rate of 100/sec, and an amplitude of 200 μA. After stimulation, animals were killed and brains were harvested and prepared for autoradiography using standard techniques. Threshold current for EMLRs in the surface-stimulated group had a mean of 67.5 ± 23.9 μA (range, 40 to 100 μA). Thresholds for in-depth stimulated group had a mean of 11.4 ± 3.5 μA (range, 10 to 20 μA). The saturation level of the I/O function for the surface-stimulated group had a mean of 287.5 ± 41.5 μA (range, 250 to 350 μA). The saturation level for the in-depth stimulated group had a mean of 192.9 ± 49.5 μA (range, 100 to 250 μA). The dynamic range for the surface electrodes had a mean of 13.1 ± 2.7 dB (range, 9.9 to 15.9 dB), whereas the dynamic range for the penetrating electrodes had a mean of 24.5 ± 2.6 dB (range, 20 to 28.0 dB). Autoradiographs generated by CNS tissue from stimulated animals demonstrated no significant difference in metabolic activity of the CN between surface and in-depth stimulated groups. However, there were highly significant differences in 2-DG uptake in the contralateral superior olivary complex, contralateral inferior colllculus, and ipsilateral and contralateral lateral lemniscus, with greater uptake in in-depth stimulated preparations. Electrophysiologic and autoradiographic data suggest that a penetrating CN prosthesis is capable of activating the auditory tract at a lower threshold, with a relatively wider dynamic range than a surface prosthesis. This probably reflects greater efficiency of neurostimulation by virtue of direct access to larger populations of second-order neurons and avoiding current injection through surface pial layers.