Objectives: Glucocorticoids (GCs) are a mainstay of treatment for patients with ANCA-associated vasculitis (AAV) but are associated with significant adverse effects. Effective remission induction in severe AAV using extremely limited GC exposure has not been attempted. We tested an early rapid GC withdrawal induction regimen for patients with severe AAV. Methods: Patients with active MPO- or PR3-ANCA vasculitis or ANCA-negative pauci-immune glomerulonephritis were included. Induction treatment consisted of two doses of rituximab, 3 months of low-dose CYC and a short course of oral GC (for between 1 and 2 weeks). Clinical, biochemical and immunological outcomes as well as adverse events were recorded. Results: A total of 49 patients were included, with at least 12 months of follow-up in 46. All patients achieved remission, with decreases observed in creatinine, proteinuria, CRP, ANCA level and BVAS. Three patients requiring dialysis at presentation became dialysis independent. Two patients required the introduction of maintenance GC for treatment of vasculitis. Overall outcomes were comparable to those of two matched cohorts (n = 172) from previous European Vasculitis Society (EUVAS) trials, but with lower total exposure to CYC and GCs (P < 0.001) and reduced rates of severe infections (P = 0.02) compared with the RITUXVAS (rituximab versus cyclophosphamide in AAV) trial. We found no new cases of diabetes in the first year compared with historic rates of 8.2% from the EUVAS trials (P = 0.04). Conclusion: Early GC withdrawal in severe AAV is as effective for remission induction as the standard of care and is associated with reduced GC-related adverse events.