Bone disease related to aluminum toxicity (aluminum-related bone disease) presents with variable clinical and biochemical findings in patients with renal failure. Bone pain and muscle weakness are common, although afflicted patients can be asymptomatic. Bone pain can be generalized or localized to the hips, back, feet, or ankles; proximal muscle weakness is common. Most cases in the United States arise from the ingestion of aluminum-containing gels by patients on long-term dialysis treatment. Patients at increased risk for developing aluminum-related bone disease include those with earlier parathyroidectomy, failed renal transplant, previous bilateral nephrectomy, and diabetes mellitus. Biochemical features that are common with aluminum-related bone disease include plasma aluminum levels greater than 100 to 150 micrograms/L, serum parathyroid hormone (PTH) levels equal to or lower than those in dialysis patients without bone disease, and normal or slightly elevated serum calcium levels. Plasma alkaline phosphatase levels are often elevated. In our experience, microcytic anemia has been uncommon. An increase in plasma aluminum levels greater than 200 micrograms/L 24 to 48 hours after the infusion of the chelating agent deferoxamine (DFO) correlates with an increased bone aluminum content, and an increment greater than 400 micrograms/L suggests marked aluminum accumulation. Radiographs are usually nonspecific. When results from indirect diagnostic procedures are equivocal, a bone biopsy is necessary. After a diagnosis of aluminum-related bone disease is established, therapy with DFO may be useful. DFO increases both the total plasma aluminum level and its ultrafilterable fraction. After an infusion of DFO, the removal of aluminum increases from 50 to 300 micrograms to 4 to 8 mg per dialysis session. Aluminum removal is similar during continuous ambulatory peritoneal dialysis after either intravenous (IV) or intraperitoneal (IP) administration of DFO. Usually, 2 to 4 g of DFO is administered once weekly, but the optimal dose and duration of therapy have not been determined. Symptoms usually improve after 4 to 12 weeks, and bone biopsies show improvement after treatment for 6 to 12 months. Further experience with DFO is needed, both to identify the optimal dosage and to clarify the risks of long-term therapy in patients with renal failure.