Surgical management of secondary glaucoma after pars plana vitrectomy and silicone oil injection for complex retinal detachment.

OBJECTIVE: To evaluate the outcomes of surgical intervention for secondary glaucoma after pars plana vitrectomy and silicone oil injection for repair of complex retinal detachment. DESIGN: Retrospective noncomparative interventional case series. PARTICIPANTS: Forty-three eyes of 43 patients who underwent incisional surgery for secondary glaucoma after pars plana vitrectomy and silicone oil injection for repair of complex retinal detachment over a 9-year period. MAIN OUTCOME MEASURES: Intraocular pressure (IOP), intraoperative and postoperative complications, visual acuity, and the need for further surgical intervention for glaucoma. Success was defined as IOP < or =21 mmHg and > or =5 mmHg with or without medication but without surgical reoperation for glaucoma. RESULTS: Findings associated with elevated IOP included emulsified oil in the anterior chamber (n = 14), pupillary block from silicone oil (n = 13), open-angle glaucoma without silicone oil in the anterior chamber (n = 9), and angle-closure glaucoma without pupillary block (n = 7). The mean (+/- standard deviation) IOP was 41.4 +/- 15.1 mmHg before surgery for glaucoma and 17.2 +/- 10.2 mmHg after an average follow-up of 19.6 months (P < 0.001). Cumulative success was 69%, 60%, 56%, and 48% at 6, 12, 24, and 36-months respectively. In patients who underwent silicone oil removal alone for surgical management of glaucoma (n = 32), 11 of 12 IOP failures (92%) were due to uncontrolled IOP, whereas most IOP failures in the group who underwent silicone oil removal plus glaucoma surgery (n = 8) failed because of hypotony (3 of 4, 75%, P = 0.027). Of three patients who underwent glaucoma surgery alone to control IOP, one failed because of hypotony. There was no significant change in visual function at last follow-up (logarithm of the minimum angle of resolution [logMAR] 2.01) compared with preoperative visual function (logMAR 2.07, P = 0.74). CONCLUSION: Surgical management of secondary glaucoma after silicone oil injection for complex retinal detachment may achieve good IOP control and stabilization of visual function in most patients. Patients who undergo silicone oil removal alone to control IOP are more likely to have persistent elevation of IOP and possibly undergo reoperation for glaucoma, whereas patients who undergo concurrent silicone oil removal and glaucoma surgery are more likely to have hypotony.

The possibility of gene therapy for the treatment of choroidal neovascularization.

PURPOSE: Choroidal neovascularization (CNV) is responsible for most cases of severe visual loss in age-related macular degeneration. Recently, the possibility of gene therapy has been proposed for the treatment of CNV. The purpose of this study was to examine the feasibility of ex vivo and in situ gene therapy approaches for CNV. DESIGN: Experimental study. METHODS: Human retinal pigment epithelial (RPE) cells were transduced with a retroviral vector coding for beta-galactosidase. Transduced cells were grown on type II collagen sheets and transplanted under the retina of 20 rabbits. Animals were observed for 3 to 56 days, and transplanted cells were examined histologically and with X-gal staining. Bovine choroidal endothelial cells (CEC) were transduced with retroviral vectors coding for tissue inhibitor of metalloproteinase-2 (TIMP-2) or control vector. Production of TIMP-2 by transduced cells was determined by immunohistochemical analysis and enzyme-linked immunosorbent assay. Effect of transduction on in vitro proliferation, migration, and tube formation was examined in response to vascular endothelial growth factor (VEGF). Four CNV lesions were induced in one cynomolgus monkey by laser photocoagulation. Two days later, retroviral vector coding for TIMP-2 or control vector was injected into the subretinal space overlying the CNV lesions. The monkey was observed for 12 weeks using fluorescein angiography. RESULTS: Transplantation of transduced RPE cells was technically achieved in 10 of 20 animals. In these animals, RPE cells at the site of transplantation formed a monolayer and expressed beta-galactosidase for 14 days. beta-Galactosidase-positive cells were not identified at 56 days. Choroidal endothelial cells transduced with TIMP-2 secrete TIMP-2 into the media and show decreased migration and tube formation in vitro. In the in vivo monkey model, the control CNV lesions (n = 2) showed prominent leakage, whereas the experimental lesions (n = 2) showed minimal hyperfluorescence. CONCLUSIONS: Retrovirally transduced RPE cells survive in the subretinal space for at least 14 days and continue to express the gene product coded for by the vector. Choroidal endothelial cells retrovirally transduced for TIMP-2 produce TIMP-2 in vitro and show decreased angiogenic responses in vitro in response to VEGF. A preliminary study attempting in situ delivery of TIMP-2 vector to CNV lesions in a monkey eye supports the feasibility of this approach and encourages further study.

Retinal pigment epithelial cells are heterogeneous in their expression of MHC-II after stimulation with interferon-gamma.

The pattern of interferon-gamma-induced major histocompatibility complex Class II antigen expression was evaluated on the retinal pigment epithelium. Experiments were performed in vitro using explant cultures of aged and fetal human eyes and in vivo in albino rabbits. The human explants were stimulated with 50 U ml-1 interferon-gamma for 3 days prior to immunostaining for Class II. The rabbit eyes were subretinally injected in vivo with 50 microl of interferon-gamma (500 U ml-1) and analyzed immunohistochemically 3 days later. A heterogeneous pattern of Class II expression was present in the interferon-gamma-stimulated retinal pigment epithelial cells, in both the in vivo and the in vitro experiments. In aged human eyes the percent of Class-II positive cells was higher in the periphery than in the posterior pole (macular region) after interferon-gamma stimulation (P<0.01). No such difference was found in the fetal eyes. These data demonstrate that retinal pigment epithelial cells are heterogeneous in their response to interferon-gamma. The results are supportive of previous studies demonstrating the structural and proliferative heterogeneity of the retinal pigment epithelium. Together, these studies provide support for the possibility of functional retinal pigment epithelial heterogeneity.