In 1992, we published the results of a prospective, controlled trial of aggressive therapy (high-dose prednisone plus oral cyclophosphamide alone or with plasmapheresis) in 86 patients with severe lupus nephritis. During this study, remission (serum creatinine < or =1.4 mg/dL [< or =123 micromol/L] and proteinuria < or =330 mg/d of protein) in renal disease occurred in 37 patients (43%). To assess the long-term effect of remission on patient and renal survival, we now report the results of our extended follow-up of these patients. After an average of 10 years of follow-up in the 86 patients, patient survival rates at both 5 and 10 years were 95% in the group that had a remission and 69% at 5 years and 60% at 10 years in the no-remission group (P < 0.001). Renal survival rates were 94% at both 5 and 10 years in the remission group compared with 46% at 5 years and 31% at 10 years in the no-remission group (P < 0. 0001). Features predictive of remission included stable renal function after 4 weeks on therapy, category IV lesion, lower chronicity index, white race, lower urine protein excretion level at baseline, and lower baseline serum creatinine level. The features predictive of end-stage renal disease were higher baseline serum creatinine level, presence of anti-Ro antibodies, and failure to attain a remission. Thus, in patients with the most severe forms of lupus nephritis, a remission of clinical renal abnormalities is associated with dramatic improvement in long-term patient and renal survival.