Twenty-nine of the 129 patients with RP seen at the Mayo Clinic between 1943 and 1984 had renal involvement. These patients were older, had arthritis and extrarenal vasculitis more frequently, and had a significantly worse survival rate than those without renal involvement. Renal biopsies were obtained in 11 of these 29 patients. The predominant lesions were mild mesangial expansion and cell proliferation, and segmental necrotizing glomerulonephritis with crescents. Small amounts of electron-dense deposits, predominantly mesangial, were noted on electron microscopy. Immunofluorescence revealed faint deposition of C3 and/or IgG or IgM, predominantly in the mesangium. Autopsies were obtained in 13 of the 47 patients who had died. Information regarding the renal pathology was available in 10 of these 13 autopsies. At the time of the initial evaluation at the Mayo Clinic, 6 of these 10 patients had evidence of renal involvement. At autopsy, none of these 10 patients had evidence of active renal vasculitis or segmental necrotizing glomerulonephritis, but 8 of the 10 patients exhibited variable degrees of vascular and glomerular sclerosis, segmental mesangial proliferation, tubular loss, and interstitial lymphocytic infiltrates. These observations expand the limited information available in the literature, which is based on 11 previously published case reports of renal involvement in RP. In only a few of our patients and previously reported patients were the manifestations of the disease limited to the systems characteristically involved in pure RP. The frequent coexistence of other autoimmune and connective tissue diseases supports the role of immune mechanisms in the pathogenesis of this syndrome. Deposition of immune complexes is likely to play a role in the pathogenesis of the glomerular lesions associated with RP. Administration of corticosteroids alone is sufficient to induce a complete remission in some cases, while in others the addition of a cytotoxic agent is necessary to control the activity of the disease or to spare corticosteroid side effects and maintain a remission. Immunosuppression-related infectious complications and undetected relapses after discontinuation of immunosuppressive therapy are largely responsible for the morbidity and mortality observed in these patients.