Methylprednisolone (MP) has been prescribed for the treatment of solar retinopathy presumably because of its anti-inflammatory effect. Recently, high doses of MP have been shown to ameliorate light-induced photoreceptor degeneration, and the mechanism of action was suggested to be the inhibition of lipid peroxidation. In this study we examined the dose-response effect and the effect of delayed treatment with MP in an established rat model of retinal photic injury. Animals received intraperitoneal injections of either MP (8, 160, or 320 mg/kg/day) or saline solution (as a control) for 2 days. Injections were started simultaneously with the commencement of light exposure or delayed for 6 or 24 h. The animals were sacrificed 6 days after light exposure, and the retinal damage was assessed by light microscopy and morphometric measurement of the outer nuclear layer (ONL) thickness. Morphologically and morphometrically, treatment with 8 mg/kg/day of MP was not effective, while treatment with 160 mg/kg/day caused better preserved photoreceptors and a thicker ONL compared with the controls (P less than 0.001). Animals administered with a dose of 320 mg/kg/day showed more severe damage to photoreceptors, resulting in a thinner ONL (P less than 0.05). When treatment with 160 mg/kg/day was delayed for 6 h, a similar efficacy as in the no-delay group was noted, but when treatment was delayed for 24 h, no beneficial effect was observed (P = 0.19). Our results demonstrated that early treatment with high doses of MP ameliorated retinal photic injury in rats.