Epiretinal implants are designed to electrically stimulate the ganglion cells with electrodes attached to the inner retina. Commonly, the stimulating part of epiretinal implants containing electrodes is fixed to the posterior pole with a retinal tack (Majji et al. 1999; Walter et al. 1999; Ivastinovic et al. 2010). Injuries to the retina, however, provoke up-regulation of glial fibrillary acidic protein (GFAP) associated with subsequent impairment of retinal function (Lewis & Fisher 2003; Wallentén et al. 2008). The aim of this study is to explore the extent and distribution of GFAP accumulation in the retina induced by retinal tack insertion at the posterior pole.
Therefore, we performed an experimental animal study in five domestic pigs weighting 20–25 kg. This animal trial was approved by Austrian legal authorities and the Medical University of Graz. The animals were treated according to guidelines of the Association for Research in Vision and Ophthalmology. The surgeries were performed under sterile conditions in general anaesthesia. All animals underwent vitrectomy using standard surgical parameters and insertion of a titanium retinal tack (Geuder GmbH, Heidelberg, Germany) with a retinal forceps (Geuder GmbH) at the posterior pole above the upper vessel arcade (Fig. 1A). After 4 weeks, the operated eyes of the animals were examined with indirect ophthalmoscopy in general anaesthesia followed by the painless euthanization and enucleation. The globes were then fixed with Davidson solution consisting of glacial acetic acid, 95% ethyl alcohol, 10% neutral buffered formalin and distilled water (mixed in the relation 1:3:2:3). Prior to fixation, the anterior segment of the eyes was removed to facilitate the entrance of Davidson solution into the vitreous cavity. Before embedding the globes in paraffin, the retinal tacks were removed from the posterior pole and the globes were placed in ethanol 70%, 90% and 100% for 1 day each. The samples were stained with Elastica van Gieson; GFAP was detected using GFAP antibodies. In each eye, the area around the retinal tack, the macular region and the optic nerve disc were examined.
Clinically, no pathological changes of the retina were observed; the retinal tacks remained in place throughout the duration of the study. The histological examination revealed thickening of the inner plexiform layer around the retinal tack measuring approximately 1000 μm in diameter (Fig. 1B); no collagen fibres could be identified. In the corresponding area, intense GFAP accumulation was observed (Fig. 1C). The macular region and the optic nerve head did not show any pathological thickening (Fig. 2A). In these areas, only slight GFAP expression was noted (Fig. 2B,C).
GFAP is an intermediary filament expressed by Müller cells as a response to retinal injury (Lewis & Fisher 2003). Previously, it has been reported that postoperative GFAP accumulation is associated with impaired retinal function (Wallentén et al. 2008). In our study, intense GFAP accumulation was confined to the retinal area of 500 μm surrounding the retinal tack. The overall diameter consequently measured 1000 μm. However, in the areas predestined for electrical stimulation and transfer of electrical impulses including the macular region and the optic nerve, only slight GFAP accumulation was observed. The associated impairment of the retinal function plays a minor role since the GFAP up-regulation resolves over several months and the retinal function completely restores (Wallentén et al. 2008). Hence, the electrical stimulation would not be hampered after an adequate recovery period. In conclusion, the insertion of a retinal tack induces intense up-regulation of GFAP in the close vicinity of the retinal tack, whereas in regions predestined for electrical stimulation and its transfer, only slight GFAP accumulation was observed.