Stable, patchy RPE loss was observed with a low concentration of NaIO(3). The OKR measurement showed changes in visual function earlier than the cued WM test and before histologic findings were observed.

This study investigated the anatomical consequences of a photoreceptor toxin, iodoacetic acid~IAA!, in the rabbit retina. Retinae were examined 2 weeks, 1, 3, and 6 months after systemic IAA injection. The retinae were processed using standard histological methods to assess the gross morphology and topographical distribution of damage, and by immunohistochemistry to examine specific cell populations in the retina. Degeneration was restricted to the photoreceptors and was most common in the ventral retina and visual streak. In damaged regions, the outer nuclear layer was reduced in thickness or eliminated entirely, with a concomitant loss of immunoreactivity for rhodopsin. However, the magnitude of the effect varied between animals with the same IAA dose and survival time, suggesting individual differences in the bioavailability of the toxin. In all eyes, the inner retina remained intact, as judged by the thickness of the inner nuclear layer, and by the pattern of immunoreactivity for protein kinase C-a~rod bipolar cells! and calbindin D-28~horizontal cells!. Müller cell stalks became immunoreactive for glial fibrillary acidic protein~GFAP! even in IAA-treated retinae that had no signs of cell loss, indicating a response of the retina to the toxin. However, no marked hypertrophy or proliferation of Müller cells was observed with either GFAP or vimentin immunohistochemistry. Thus the selective, long lasting damage to the photoreceptors produced by this toxin did not lead to a reorganization of the surviving cells, at least with survival as long as 6 months, in contrast to the remodeling of the inner retina that is observed in inherited retinal degenerations such as retinitis pigmentosa and retinal injuries such as retinal detachment.

In a rabbit model, the safety of using an intravitreous injection of thermo-setting gel as artificial vitreous was confirmed by ophthalmoscopic, electrophysiological, and histological studies for a relatively short observation period. However, TG injection cannot be expected to provide a tamponade effect.