Effects of Electrode Position on Vestibular Implant Performance in Rhesus Macaque
Brian J Morris1, Katherine N Mueller1, Celia C Fernandez Brillet1
1Department of Biomedical Engineering, Johns Hopkins School of Medicine, Baltimore, MD, USA.
Purpose:
Vestibular implants that target the three semicircular canal branches of the vestibular nerve can partially restore the 3-dimensional vestibulo-ocular reflex (3D VOR) in individuals disabled by bilateral vestibular hypofunction. A key goal of implant design is optimizing the number, spacing, and placement of stimulating and return electrodes to maximize response strength and selectivity. While computational models provide initial insights, empirical data are essential to validate performance.
Methods:
We unilaterally implanted stimulating electrodes in each semicircular canal and positioned return electrodes both inside the labyrinth and outside the temporal bone in three female rhesus macaques. Using mixed-effects ANOVA, we quantified how electrode location influenced 3D VOR response magnitude and misalignment.
Results:
We found that: (1) the deepest stimulating electrode in each canal generally yielded the strongest and most aligned responses; (2) a 600-750 µm difference in electrode position significantly impacted VOR outcomes; (3) return electrodes placed inside the labyrinth produced significantly larger VOR responses than those placed outside the temporal bone when stimulus current is constrained to levels that elicit no sign of facial nerve excitation; and (4) "near-bipolar" stimulation-using a return electrode in the same ampulla as the stimulating electrode-yielded better alignment.
Conclusion:
Although including multiple stimulating electrodes per canal may lower the risk of missing the target, a VI limited to one deep stimulating electrode per canal and one common return electrode can suffice if array design and surgical technique ensure placement near the crista ampullaris. Unused stimulator channels could be repurposed in designs intended to stimulate the utricle, saccule and/or cochlea. Moreover, spatial selectivity is improved by placing return electrodes within the labyrinth or ampullae, rather than outside the temporal bone. VIs that use a single common return electrode shared by all stimulation channels could achieve better performance in the future by incorporating multiple independent return electrode channels to permit near-bipolar stimulation.
