Iron therapy maintains iron stores and optimizes the response to recombinant human erythropoietin (r-HuEPO) in patients with end-stage renal failure. Information is limited, however, regarding the preferential route of iron administration in pediatric patients receiving hemodialysis. Therefore, we prospectively randomized 35 iron-replete patients (aged >1 to <20 years) to receive up to 16 weeks of maintenance i.v. ( n=17) or daily oral ( n=18) iron. Eligible patients had received hemodialysis for >2 months, had a baseline transferrin saturation [TSAT] >20%, and were receiving maintenance r-HuEPO. Treatment arms were evenly distributed with respect to baseline demographic and clinical characteristics, with no statistically significant differences in baseline hemoglobin (Hb), hematocrit (Hct), reticulocyte Hb content (CHr), serum ferritin (SF), TSAT, or r-HuEPO dose. In the 35 patients, i.v. iron dextran and not oral iron was associated with a significant increase (138.5 to 259.1 ng/ml, P=0.003) in SF. A comparison of the change in SF between the i.v. iron group and the oral iron group was also significant ( P=0.001). Whereas only i.v. iron was associated with a significant decrease in the dose of r-HuEPO (234.0 to 157.6 U/kg per week, P=0.046) and an increase of the CHr (29.2 to 30.1 pg, P=0.049), these changes were not significantly different from those experienced by patients in the oral iron group. In both groups, the Hct remained stable and in neither group was there a significant change in the TSAT. In summary, although both oral and i.v. iron maintained patients in an iron-replete state in this short-term study, only i.v. therapy allowed for a significant improvement in iron stores.