Abstract
Abstract: :
Purpose: Fundamental to the effort to build an effective visual prosthesis is an understanding of the most efficient parameters for stimulation of retinal neurons. Using ganglion cell recordings, stimulus threshold and response latency were studied in both normal and retinal degenerate (rd) isolated mouse retina, along with the effect of altering the site of stimulus delivery on these parameters. Methods: Twenty retinas each were isolated from either normal or rd mice for stimulation. Electrical stimulation was delivered via two 125 µm macroelectrodes positioned at either the photoreceptor (PR) or ganglion cell (GGL) surface. A cathodic first, charge balanced, biphasic, square wave stimulus with a phase duration of 1mS and a 3mS interphase delay was used for all stimulation cycles. Recording of induced responses was performed using penetrating tungsten wire microelectrodes positioned in the ganglion cell layer. Results: Stimulation of the ganglion cell side in normal retina yielded a significantly (P<0.05) lower mean threshold (30.04 µA/ 77.26 µA) and latency (2.11 mS/ 2.77 mS) when compared to similar stimulation of the photoreceptor side in normal retina. The latency of the ganglion cell side stimulation also proved to be significantly lower (P<0.05) than the latency for stimulation of the ganglion cell side in RD retina. (2.11 mS/ 2.69 mS) No other statistically significant differences were observed amongst the other subgroups. Conclusion: Stimulation of normal retina yields lower thresholds and latencies when the electrodes are placed on the ganglion cell side versus the photoreceptor side. This relationship is not seen in the stimulation of RD retina. Stimulation of the ganglion cell side of RD retina also had a higher latency and tended to have a higher threshold than ganglion side stimulation in normal retina.
Keywords: 554 retina • 394 electrophysiology: non-clinical • 316 animal model