Background: In epoetin-treated dialysis patients, currently iron is administered by the intravenous route to maintain optimum erythropoiesis. However, rapid infusion of iron in excess of transferrin binding capacity can lead to the availability of unbound iron that can theoretically catalyze peroxidation of lipids, such as low-density lipoprotein (LDL), which when oxidatively modified is proinflammatory and promotes atherogenesis.
Methods: To address this issue, our study used one of the most specific measures of lipid peroxidation available, namely gas chromatography/mass spectometry (GC/MS) analysis of F2-isoprostanes. Using a prospective design, blood samples were collected 15 minutes before (pre) and 30 minutes after (post) a one-hour infusion of 700 mg bolus of intravenous iron in 22 adult home-hemodialysis patients on a non-hemodialysis day.
Results: With iron-dextran infusion, serum iron markedly increased (mean +/- SE, 42 +/- 4 vs. 311 +/- 92 microg/dL, P < 0.0001) and exceeded the transferrin saturation of 100% in 22 out of 22 patients (pre 23 +/- 3 vs. post 165 +/- 8%, P < 0.0001). Plasma concentrations of free F2-isoprostanes did not change significantly following infusion of iron (pre 40 +/- 5 vs. post 39 +/- 6 pg/mL). However, levels of F2-isoprostanes esterified in plasma lipoproteins increased significantly in the postinfusion samples (pre 199 +/- 19 vs. post 233 +/- 25 pg/mL, P < 0.004). Pre-infusion levels of serum iron correlated directly with pre-infusion levels of esterified F2-isoprostanes (r = 0.56, P = 0.008), which persisted in the postinfusion period (r = 0.43, P = 0.04). However, there was no correlation between esterified F2-isoprostanes and serum ferritin levels. In the last four patients in whom blood samples were collected five hours after the intravenous iron infusion, there were further increases in esterified F2-isoprostanes that very closely correlated with postinfusion serum iron levels (r = 0.99, P = 0.013). In a control study, the in vitro addition of iron dextran to blood samples did not increase free or esterified F2-isoprostanes, suggesting that the increase in esterified F2-isoprostanes seen in vivo after iron infusion in patients is not due to a procedural artifact.
Conclusion: Collectively our data suggest that high levels of serum iron appearing soon after a large bolus of iron infusion is associated with significant, albeit modest, increases in levels of F2-isoprostanes esterified in plasma lipoproteins that tended to increase with time. Although it is uncertain whether this degree of lipid peroxidation may have deleterious effects, it may be sagacious to explore whether this can be prevented by slow infusion of frequent smaller doses of iron and, if necessary, along with administration of antioxidants.