OBJECTIVE. To investigate the effectiveness and side effects of oral versus pulse cyclophosphamide (CYC) in combination with corticosteroids (CS) in the treatment of systemic Wegener's granulomatosis (WG). METHODS. Patients with newly diagnosed systemic WG were enrolled in a prospective, randomized trial. At the time of diagnosis, prior to randomization, every patient received a daily injection of methylprednisolone for 3 days, followed by daily oral prednisone (1 mg/kg/day) and a 0.7-gm/m2 pulse of CYC. Patients were then randomly assigned to receive either prednisone plus intravenous pulse CYC (group A) or prednisone plus oral CYC (group B) as first-line treatment. CYC was given for at least 1 year and was then progressively tapered and discontinued. RESULTS. Fifty patients were included in the study: 27 in group A and 23 in group B. At 6 months, 24 group A patients (88.9%) were in remission, versus 18 group B patients (78.3%). At the end of the trial, 18 group A patients (66.7%) and 13 group B patients (56.5%) were in remission. In group A, 66.7% of the patients experienced side effects, versus 69.6% in group B. Infectious side effects were significantly more frequent in group B (69.6%) than in group A (40.7%) (P < 0.05). The incidence of Pneumocystis carinii pneumonia was higher in oral CYC-treated patients (30.4%) than in pulse CYC-treated patients (11.1%). Nine group A patients (33.3%) and 10 group B patients (43.5%) died. Actuarial curves showed that relapses were significantly more frequent in group A (59.2%) than in group B (13%) (P = 0.02). CONCLUSION. Our results indicate that pulse CYC is as effective as oral CYC in achieving initial remission of WG and is associated with fewer side effects and lower mortality. However, in the long term, treatment with pulse CYC does not maintain remission or prevent relapses as well as oral CYC.