The administration of parenteral iron dextran to hemodialysis patients is typically intermittent. We sought to determine the most appropriate intervals for sampling iron parameters during intermittent need-based and continuous maintenance regimens and to quantify differences in efficacy between such regimens during long-term therapy. After a single course of 10 consecutive 100-mg iron doses administered to 14 patients on 16 occasions, transferrin saturation (TSAT) and ferritin were unreliable indices of iron status for the next 2 and 6 weeks, respectively. TSAT and ferritin levels at 1 week were virtually identical to those at 2 weeks after the administration of a single 50-mg or 100-mg iron dextran dose to 16 other patients. Twelve patients on maintenance iron therapy (25 to 100 mg/wk; TSAT, 30% to 50%) had a statistically significant decrease in the amount of recombinant human erythropoietin (rHuEPO) needed to maintain hemoglobin (Hb) levels between 10 and 11 g/dL compared with 12 patients receiving intermittent need-based dosing, an effect that persisted from week 16 to week 72 of the study. Maintenance iron was feasible even in a third group of eight patients targeted to sustain an Hb level of 14 g/dL. In both iron maintenance groups, iron indices could be measured at weekly intervals, and ferritin levels did not progressively increase over time. Continuous maintenance iron dextran used to maintain TSATs of 30% to 50% significantly reduced rHuEPO requirements and resulted in no adverse side effects in chronic hemodialysis patients. After weekly maintenance 25- to 100-mg iron dextran doses, iron indices can be measured after 1 week; a delay of 2 weeks is not necessary.