Abstract by Madeline Evers Olufsen

Background and objective: The macula is an essential part of vision. Loss of retinal cells within the macular region is a common feature for age-related macular diseases, such as refractory macular hole (MH) and age-related macular degeneration (AMD). In recent years, several surgical techniques have been proposed for MHs that are refractory to standard surgical treatment. Many of the techniques involve the insertion of a tissue graft into the MH to facilitate closure. These grafts include a human amniotic membrane (hAM) sheet and an autologous neurosensory retinal transplant (ART) that are harvested from the periphery of the same eye. Case studies have shown improved anatomical hole closure and visual gain. However, the mechanism underlying hole closure is not yet fully understood. We hypothesize that the observed visual improvement may be explained by regeneration of retinal cells, neuro-retinal integration or simply the effect of hole closure itself. This thesis explores the morphological changes during retinal healing in an in vivo porcine model of a retinal hole, both with and without treatment using a tissue graft.

Methods: Experiments were performed on 38 female domestic Danish landrace pigs distributed across three studies. All pigs underwent standard vitrectomy, including induction of a posterior vitreous detachment (PVD). A subretinal bleb was created temporally within the major arcades, followed by the creation of a retinal hole of approximately 800-4000 µm using a vitrector. The outcomes included the anatomical status of the retinal hole observed on optical coherence tomography (OCT) scans and histological examination of retinal morphology. Study I: This study investigated spontaneous hole closure and the significance of hole size in untreated retinal holes. The outcomes were OCT and histology after 1, 2, and 4 weeks. Study II: The mechanisms underlying retinal hole closure following subretinal hAM sheet transplantation were explored. The outcomes were OCT and histology after 2 and 4 weeks. Study III: A neurosensory retinal graft was harvested from the nasal side of the optic disc within the major arcades and subsequently inserted into the retinal hole. The mechanism underlying hole closure, as well as survival and integration of the retinal graft, was investigated. The outcomes were OCT and histology after 2 and 6 weeks.

Results: Study I: Retinal holes < 1380 µm closed spontaneously, whereas larger holes remained open with minimal signs of scar tissue or inflammation. Hole closure was facilitated by a centripetal movement of the edges of the hole, and in most cases, the residual gap was sealed by a retinal plug mainly composed of astrocytes. Study II: The subretinal hAM sheet facilitated closure of large retinal holes by inducing gliosis and serving as a scaffold for the centripetal migration of the edges of the hole. The retinal plug formed above the hAM sheet was primarily comprised of astrocytes. While retinal layering was not restored within the plug, the surrounding retina - which had moved centrally during hole closure - remained preserved above the hAM sheet. Study III: OCT scans indicated preservation of the outer retinal layers of the graft in four cases, and in two of these cases, there was apparent integration with the retinal layers of the adjacent host retina. Corresponding histology confirmed preservation of the photoreceptor layer in three cases with no evidence of neural integration of the graft. There was shrinkage of the retinal grafts during the follow-up period in all cases.

Conclusion: The mechanism underlying retinal hole closure involves glial cell proliferation and centripetal migration of the edges of the hole. While spontaneous closure of iatrogenic retinal holes in the posterior pole of the porcine retina can occur, larger holes require additional treatment to achieve closure. The hAM acts as a stimulator for hole closure by inducing gliosis and providing a scaffold for the centripetal migration of the edges of the hole. While the outer retina of a retinal graft may survive, the inner retina degenerates. There is no evidence of neural integration of the graft. The retinal graft primarily serves as a scaffold for centripetal migration of the edges of the hole and thereby assemble the hole closure mechanism seen in the hAM.