Corticosteroids and daunomycin in the prevention of experimental proliferative vitreoretinopathy induced by macrophages.

An experimental model of proliferative vitreoretinopathy (PVR) induced by macrophages simulates a special form of wound healing process in the eye and mimics the development of PVR from its initial stage. We used this model for the evaluation of drug efficacy in the prevention of PVR. One mg triamcinolone acetonide (TA), 10 micrograms daunomycin-liposome (DL), 5 micrograms free daunomycin (FD) and 0.1 ml saline or empty liposomes (as controls) were injected into the vitreous in four groups of animals (30 or 40 rabbit eyes each) after macrophage injection. Retinal detachment developed in 77.5% of the control eyes on day 28, compared to 13.3% of the TA-treated eyes (P < 0.01), to 33.3% of the eyes treated with DL (P < 0.01), and 50% of the FD-treated eyes (P < 0.05). TA cleared up from the vitreous within 35-63 days (average 45.5 days). The half-time of FD clearance was 145.5 min. Although DL declined rapidly during the first 2 days, there was an average of 0.64 microgram/ml daunomycin in the vitreous on day 14. Transmission electron microscopy showed that FD at a dosage of over 5 micrograms or DL over 20 micrograms was toxic to the retina and that up to 4 mg TA was nontoxic. These results suggest that steroids such as TA, given at the inflammatory stage, can effectively and safely prevent the development of PVR, and that encapsulation in liposomes of cytotoxic agents such as daunomycin can enhance drug efficacy and reduce toxicity. The time course of initiation and development of PVR is important in the selection of particular drugs.