Abstract: PCP was administered as 1 g cyclophosphsmide/m2/body surface, once per month, with 0.5mg prednisolone/kg/day, to 71 patients with severe posterior uveitis (PU) and/or retinal vasculitis (RV) , A Disease Activity Index (DAI) was calculated for each section of each eye, before the first PCP and then after each one. When a remission was obtained the gap between, PCP was increased gradually and prednisolone was tapered. If the result remained stable, PCP was given only in the case of a new attack. A comparison between the DAI before PCP, and after the last one was made by the Student t-test. Our patients received a total number of 664 PCPs, with a mean number of 9.4 PCPs per patient. The mean follow-up time was 19 months. 82% of the eyes with PU improved, 6% were stabilized, while 13% were aggravated. The mean DA! improved from 2.5 to 1 with t=9.911 and p<0.001, 69% of the eyes with RV improved, 11% were stabilized, while 20% were aggravated. The mean DA! improved from 1.8 to 0.7 with t=4.564 and p<0.001. The Visual Acuity (VA), as an overall index, improved in 54% of eyes. VA was stabilized in 19% and decreased in 27% of eyes, The mean VA improved from 3.5 to 4.5/10 with t=3.960 and p<0.001. None of the major side effects of oral cyclophosphamide was observed with PCP. From these results we conclude that (i) PCP is a very useful method for severe ocular Behcet. (ii) overall results do not deteriorate with time, and (iii) side effects are minor.