We compared, in a randomized controlled trial, the efficacy of a regimen based on intravenous (i.v.) cyclophosphamide therapy with a combination of i.v. dexamethasone and oral cyclophosphamide therapy in inducing remission in patients with steroid-resistant nephrotic syndrome (SRNS). During April 2001 to December 2003, 52 consecutive patients with idiopathic SRNS, normal renal function and renal histology findings showing minimal change disease, focal segmental glomerulosclerosis or mesangioproliferative glomerulonephritis were enrolled into the study. Patients in group I received i.v. injection of cyclophosphamide once a month for 6 months and prednisolone on alternate days. Those in group II received i.v. treatment with dexamethasone (initially on alternate days, later fortnightly and monthly; total 14 doses), oral cyclophosphamide therapy (for 3 months) and prednisolone on alternate days. Data from 49 patients (26 in group I, 23 in group II) were analyzed; their clinical and biochemical features were similar at inclusion. Following treatment, complete remission was seen in 53.8% and 47.8% patients in groups I and II, respectively (P = 0.6). Long-term follow up showed favorable outcome in 14 (53.8%) patients in group I, and 9 (39.1%) in group II. Chief adverse effects, including cushingoid features and serious infections, were similar in both groups. Patients receiving i.v. dexamethasone therapy commonly showed hypertension and hypokalemia, while vomiting and reversible alopecia occurred in those receiving i.v. treatment with cyclophosphamide. In patients with SRNS, the efficacy of treatment intravenously with cyclophosphamide and orally with prednisolone was similar to the combination of dexamethasone intravenously, orally administered cyclophosphamide and prednisolone.