To identify the role of iron overload in the natural history of liver fibrosis, we reviewed serial hepatic biopsy specimens taken annually from patients cured of thalassemia major by bone marrow transplantation. The patients underwent transplantation between 1983 and 1989 and did not receive any chelation or antiviral therapy. Two hundred eleven patients (mean age, 8.7 +/- 4 years) were evaluated for a median follow-up of 64 months (interquartile range, 43-98 months) by a median number of 5 (interquartile range, 3-6) biopsy samples per patient. Hepatic iron concentration was stratified by tertiles (lower, 0.5-5.6 mg/g; medium, 5.7-12.7 mg/g; upper, 12.8-40.6 mg/g dry weight). Forty-six (22%) patients showed signs of liver fibrosis progression; the median time to progression was 51 months (interquartile range, 36-83 months). In a multivariate Cox proportional hazard model, the risk for fibrosis progression correlated to medium hepatic iron content (hazard rate, 1.9; 95% confidence interval [CI], 0.74-5.0), high hepatic iron content (hazard rate, 8.7; 95% CI, 3.6-21.0) and hepatitis C virus (HCV) infection (hazard rate, 3.1; 95% CI, 1.5-6.5). A striking increase in the risk for progression was found in the presence of both risk factors. None of the HCV-negative patients with hepatic iron content lower than 16 mg/g dry weight showed fibrosis progression, whereas all the HCV-positive patients with hepatic iron concentration greater than 22 mg/g dry weight had fibrosis progression in a minimum follow-up of 4 years. Thus, iron overload and HCV infection are independent risk factors for liver fibrosis progression, and their concomitant presence results in a striking increase in risk.