Using data from systematic reviews and randomised controlled trials, the evidence for managing steroid sensitive nephrotic syndrome (SSNS) is reviewed. In the initial episode, increased duration (3–7 months) of prednisone compared with 2 months significantly reduced the risk for relapse at 12–24 months [relative risk (RR) 0.70; 95% confidence intervals (CI) 0.58–0.84] without increase in adverse effects. Six months of prednisone was significantly more effective than 3 months (RR 0.57; 95% CI 0.45–0.71). Higher prednisone doses given for the same duration reduced the risk of relapse (RR 0.59; 95% CI 0.42–0.84) suggesting that both dose and duration of prednisone therapy lead to prolonged remission. In relapsing SSNS prolonged prednisone treatment, daily prednisone during infections, oral or intravenous cyclophosphamide, chlorambucil, levamisole and cyclosporin significantly reduced the risk of relapse. Comparative effects of these options remain uncertain because of the absence of head-to-head trials, but existing trial evidence is strongest for cyclophosphamide and cyclosporin. Further adequately powered multinational trials are required to determine the optimum induction dose and duration of prednisone in the initial episode of SSNS and to determine the relative efficacies of immunosuppressive agents and the efficacy of newer agents, including mycophenolate and tacrolimus, in relapsing SSNS.