A new model of proliferative vitreoretinopathy.

PURPOSE To design a new model of proliferative vitreoretinopathy (PVR) that would not rely on the addition of exogenous cells. The release of endogenous cells from surrounding attachments seems to be an early event in the pathogenesis of PVR. Because the proteolytic enzyme dispase dissociates tissues, the hypothesis was that an intraocular injection of dispase could trigger events that would cause PVR. The requirement for a surgical retinal break at the time of dispase injection was also examined. METHODS One eye of Dutch Belted rabbits was injected with 0.003 U to 1.0 U dispase in the subretinal space or vitreous cavity. Control rabbits received a saline injection. An intentional retinal tear was created in animals in some groups. Observations were made for at least 10 weeks after surgery. RESULTS Proliferative vitreoretinopathy developed in response to subretinal or intravitreal dispase, with or without creation of a controlled retinal break. Increased severity of PVR correlated with increasing doses of dispase. Evidence of PVR included preretinal membranes, distortion of myelin wings and retinal vessels, fixed retinal folds, and traction retinal detachment. Proliferative vitreoretinopathy did not develop in saline-treated control animals. CONCLUSIONS Dispase initiated the development of PVR without the addition of exogenous cells, growth factors, or cytokines typically found in PVR membranes. A cascade of events was probably triggered by dispase, causing native cells and factors to produce PVR. The dispase model of PVR was technically easy to perform, permitted a clear view of the retina, and had a high success rate in development of PVR.