A Pre-clinical Model for Safe Retinal Stimulation

Abstract

Purpose : Safety limits for electrical stimulation of the retina are unclear. We developed a model to determine safe retinal stimulation limits using a clinical electrode array driven by two independent current sources.

Methods : Normally sighted adult cats (n=13) were implanted, suprachoroidally, with a medical grade electrode array. The array contained 44 platinum (Pt) disc electrodes (1 mm diameter) and 2 large Pt return electrodes. The lead assembly was tunnelled to an exteriorised connector. Following recovery, the animals were fitted with a wearable harness containing two multichannel stimulators controlled by a custom designed and built processor allowing simultaneous (paired) electrode activation. Stimulation parameters were fully controllable, and electrode impedances monitored, wirelessly using custom software running on a laptop PC base station. Biphasic current pulses were delivered continuously at a range of defined charge levels (250-1450 nC), phase widths (145-580 µs) and rates (50-200 pps) to randomly assigned electrode pairs for up to 4 months (mean: 54 days). Eye health was assessed monthly with standard clinical testing. Pathohistological assessments were performed using our published protocols.

Results : Chronic stimulation was well tolerated by the subjects, as determined by veterinary consultation. The wireless control and monitoring system significantly reduced subject interactions. Following a clinically relevant stimulus of 500 nC per pair of electrodes at 50 Hz (phase width: 400 µs), retinal imaging revealed local increased reflectivity and a build-up of material in the outer retinal layers. Histology confirmed localised acute inflammation limited to the suprachoroidal pocket (Fig). However, no overt structural changes were observed in the retina by clinical imaging or histopathology. Higher charge levels and/or rates induced progressively more significant retinal tissue reaction including, ultimately, destruction of the photoreceptor outer segments and reorganisation of the outer nuclear layer (Fig).

Conclusions : An animal model for assessing the safety limits of chronic electrical stimulation of the retina has been developed and validated with clinical and histopathological outcomes. Future studies using this model will refine the safe stimulus parameter space and provide clinical imaging guidelines to facilitate management of retinal prosthesis recipients.

This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.

 

Spectrum of Stimulus Induced Retinal Tissue Reaction