OBJECTIVES: This study investigated the clinical course of myeloperoxidase-antineutrophil cytoplasm autoantibody (MPO-ANCA)-associated vasculitis after starting dialysis. METHODS: A retrospective review was conducted of the clinical charts of dialysis-dependent patients with MPO-ANCA-associated vasculitis who attended one of 8 associated clinics over the past 21 years. RESULTS: Eighty-nine patients were included in the study; 88 had microscopic polyangiitis (MPA) and 1 had granulomatosis with polyangiitis. Of the 88 patients with MPA, 18 had renal-limited vasculitis. Twenty-one relapses occurred among 13 patients (frequency, 0.05 relapses/person-year; 95% confidence interval, 0.03-0.08). Mean time from start of dialysis to relapse was 65 ± 59 months. Cox multivariate analysis showed that pulmonary involvement was a predictor of relapse (hazard ratio [HR], 21.4) and mortality (HR, 4.60), and that patient age (HR, 1.10) and cyclophosphamide use (HR, 0.20) were significant predictors of mortality. Postdialysis 1- and 5-year survival rates were 83.0% and 65.6%, respectively; infection was the most frequent cause of death. CONCLUSION: Pulmonary involvement was a predictor of relapse and mortality. Although relapse can occur long after the start of dialysis, incidence was low among dialysis-dependent patients. Prolonged maintenance immunosuppressive therapy might be limited to patients with pulmonary involvement in dialysis-dependent ANCA-associated vasculitis.
OBJECTIVES: This study investigated the clinical course of myeloperoxidase-antineutrophil cytoplasm autoantibody (MPO-ANCA)-associated vasculitis after starting dialysis. METHODS: A retrospective review was conducted of the clinical charts of dialysis-dependent patients with MPO-ANCA-associated vasculitis who attended one of 8 associated clinics over the past 21 years. RESULTS: Eighty-nine patients were included in the study; 88 had microscopic polyangiitis (MPA) and 1 had granulomatosis with polyangiitis. Of the 88 patients with MPA, 18 had renal-limited vasculitis. Twenty-one relapses occurred among 13 patients (frequency, 0.05 relapses/person-year; 95% confidence interval, 0.03-0.08). Mean time from start of dialysis to relapse was 65 ± 59 months. Cox multivariate analysis showed that pulmonary involvement was a predictor of relapse (hazard ratio [HR], 21.4) and mortality (HR, 4.60), and that patient age (HR, 1.10) and cyclophosphamide use (HR, 0.20) were significant predictors of mortality. Postdialysis 1- and 5-year survival rates were 83.0% and 65.6%, respectively; infection was the most frequent cause of death. CONCLUSION:Pulmonary involvement was a predictor of relapse and mortality. Although relapse can occur long after the start of dialysis, incidence was low among dialysis-dependent patients. Prolonged maintenance immunosuppressive therapy might be limited to patients with pulmonary involvement in dialysis-dependent ANCA-associated vasculitis.