The expanding indications for transfusions in patients with sickle cell disease raise the issues of appropriate measurement of body iron burden and optimal timing of iron chelation therapy. In this study, we obtained 42 biopsy specimens from 20 patients with sickle cell disease (mean age, 15.7 years) who received transfusions. In 12 patients whose mean age was 11.3 years at the time of liver biopsy, hepatic iron concentration was measured to provide information about the rate of iron accumulation in sickle cell disease, as well as to guide the initiation of chelating therapy. Mean hepatic iron concentration after an average of 15.4 transfusions administered over 21 months was 9.4 +/- 1.2 mg/g liver, dry weight, which did not correlate significantly with determinations of serum transferrin or ferritin levels. On Initial liver biopsy, hepatic portal fibrosis was noted in 4 of 12 patients. Twenty-nine biopsies in 16 patients were performed after variable periods of treatment with deferoxamine. These 16 patients had received a mean of 38.5 transfusions over 4 years. Hepatic iron was 14.1 +/- 1.9 mg/g of liver, dry weight, Indicating poor control of body iron in many patients. Cirrhosis was reported in one of 29 and portal fibrosis in 10 biopsy specimens. Hepatic iron concentration in patients in whom fibrosis was observed varied from 8.9 to 37.7 mg/g of liver, dry weight. These data show that after 1 to 2 years of conventional transfusions, variable tissue iron concentrations and tissue damage are observed in patients with sickle cell disease. In some patients, iron chelation therapy may not be appropriate after 1 year of transfusions; in others, therapy is clearly indicated by this time to prevent tissue injury. The data also suggest that patients with sickle cell disease develop increased portal fibrosis at the thresholds previously described in young patients with thalassemia (approximately 7 mg/g of liver, dry weight).