Experimental preretinal neovascularization by laser-induced venous thrombosis in rats.
PURPOSE: Retinal ischemia and neovascularization (NV) are important components of many retinal disorders. To facilitate further investigation of retinal ischemia and neovascularization, we sought to develop a reproducible in vivo experimental model of venous occlusion by photodynamic thrombosis in rats. METHODS: After anesthesia, 27 eyes of pigmented rats received an intraperitoneal injection of 0.2 ml of, 10% sodium fluorescein 15 minutes prior to laser treatment. With a blue-green argon laser, selected venous sites next to the optic nerve head were photocoagulated indirectly with a 78 diopter lens. Venous occlusion was accomplished using laser parameters of 1.0 second, 50 microns, and 50-100 mW. For a control group, 10 eyes were coagulated on the retina between major vessels using the same parameters after fluorescein injection. For a second control group, 1% sodium hyaluronate was injected into the subretinal space to make a long-standing retinal detachment in 5 eyes. RESULTS: With 1-8 laser impulses, each venous occlusion was obtained and was associated with extreme venous constriction and tortuousity. Retinal edema became evident 10-30 minutes after treatment in the sectors associated with the occluded veins. This edema became a bullous retinal detachment (RD) within 12 hours and intra-retinal hemorrhage was observed. The retinal edema continued for 3-10 days and the retinas reattached spontaneously. Prior to or after retinal reattachment 70% (19/27) of eyes developed retinal NV and tractional RD. Of these, 11 developed NV of the optic disc (NVD), 6 developed NV elsewhere (NVE), and 2 developed NVD and NVE. In 30% (8/27) of the eyes, retinal edema resolved without evidence of NV. In control groups no eyes showed either circulatory disorders or evidence of NV. CONCLUSIONS: This is a new model of retinal ischemia and associated neovascularization established by venous thrombosis that is easily reproducible. Many aspects of rat retinal physiology are known and this model has promise as an avenue for further investigation.