Introduction

Iron deficiency is the most important cause of poor response to erythropoiesis-stimulating agents in patients with end-stage renal disease, and its judicious use in pediatric patients for the correction of anemia has been recently reviewed [1]. Extrapolation of data from adults to children is necessary, since few data exist from children. There is little doubt that iron causes oxidative stress in cells and animals. Zager et al. studied the effects of different iron preparations (iron dextran, iron sucrose, iron gluconate and iron oligosaccharide) in mouse proximal tubular segments, cultured human proximal tubular cells and bovine aortic endothelial cells [2]. Each of the iron preparations tested increased oxidative stress. An elegant set of experiments provided the following important observations:

  1. 1.

    The generation of oxidative stress by iron sucrose was dependent on mitochondrial respiration, but not free iron-oxidative stress was not blocked by the iron chelator, desferrioxamine.

  2. 2.

    Cytotoxicity of parenteral iron paralleled the degree of cell iron uptake [3]. Cytotoxicity was protected by reduced glutathione, not by its antioxidant effect, but by providing cellular protection, similar to glycine, in the setting of ATP depletion.

  3. 3.

    The in vitro injury occurred at clinically relevant concentrations of parenterally administered iron [3].

  4. 4.

    Important ultrastructural differences emerged when kidneys were examined by electron microscopy. In clinically relevant concentrations (30 μg/ml), iron sucrose demonstrated greater toxicity to proximal tubules than other iron products did and was accompanied by glomerular iron deposition [3]. In particular, there was greater depletion of ATP and of cytochrome c—markers for toxicity to mitochondria [4]—by iron sucrose than by other iron preparations [3].

  5. 5.

    In vivo, intravenous injection of 2 mg iron sucrose resulted in oxidative stress in renal cortex and heart within 90 min. This was not seen with iron dextran [3]. Endothelial cell damage was seen in vivo with iron sucrose but not with iron dextran [3].

These experiments demonstrate toxicity to mitochondria, cytotoxicity, and lipid peroxidation with intravenously administered iron compounds and substantial heterogeneity amongst iron preparations and the need for further clinical investigations in patients with chronic kidney disease (CKD).

CKD is a pro-oxidant state, due to a variety of reasons that include inflammation and renin–angiotensin system activation [5]. The evidence for generation of oxidative stress with intravenous iron treatment in healthy volunteers and patients with CKD, including those on dialysis, has accumulated over the past decade. This evidence is summarized in Table 1.

Table 1 Examples of studies on humans, demonstrating oxidative stress from the intravenous (i.v.) administration of iron (HPLC high-performance liquid chromatography, MDA malondialdehyde)
Full size table

Taken together, these findings support the pre-clinical data. Importantly, iron potentiates oxidative stress above and beyond hemodialysis [12, 13]. Although vitamin C has been reported to improve erythropoietin (EPO) hyporesponsiveness [17], co-administration of iron and vitamin C may increase the oxidant potential of intravenous iron (IVIR) [18]; thus, caution is warranted when vitamin C is used in combination with IVIR.

Repeated IVIR exposure may not produce just transient, inconsequential injury. Repetitive acute renal injury may lead to chronic renal failure, as exemplified by acute cellular rejection followed by chronic allograft nephropathy and repeated acute obstruction followed by chronic renal failure. Even if such injury is mild, it can lead to the generation of sodium-sensitive hypertension [19]. Oxidative stress plays an important role in the pathogenesis and progression of renal disease [20, 21]. Data from animals suggest the association of iron exposure, generation of oxidative stress and renal injury [22]. For example, the glycerol model of acute renal failure is an established, well-characterized, model of acute renal failure that involves the iron-dependent, oxidative stress pathway that, with repeated injury at weekly intervals, leads to CKD [23]. Oxidants generated by free iron may be independently associated with proteinuria. For example, hydrogen peroxide, when co-infused with myeloperoxidase, causes profound proteinuria, endothelial cell swelling and effacement of epithelial cell foot processes [24]. The glomerular epithelial cell interacts with the basement membrane via integrin receptors, the distribution of which can be altered by oxidants, perhaps leading to functional changes [25]. Furthermore, oxidative modification of membrane proteins could directly impair the permselectivity of the glomerulus. In animals with the passive Heymann nephritis model of membranous nephropathy, treatment with the iron chelating agent desferrioxamine abrogates proteinuria [26]. Long-term exposure to iron in animals can cause tubulointerstitial fibrosis [27].

Renal biopsies from patients with nephrotic syndrome and/or CKD reveal higher iron accumulation in the proximal tubular lysosomes of patients with nephrotic syndrome than in non-nephrotic patients [28]. The extent of tubular damage correlates with the amount of iron in lysosomes and suggests the association of iron and morphological injury to the kidney [28]. Long-term exposure to high doses of iron sucrose may cause tissue damage and deleterious effects. Case reports from Japan have described toxicity to the proximal tubules, manifested by renal phosphate leakage, calcitriol deficiency and overt osteomalacia after several years of daily exposure to iron sucrose [29]. Epidemiologic data from patients with CKD on hemodialysis suggest the association of increased iron exposure and atherosclerosis, as assessed by carotid intima thickness [15]. Finally, antioxidant therapies in randomized controlled trials, administered to patients on hemodialysis, have been demonstrated to reduce the number of cardiovascular events [30, 31].

Although elevated levels of tissue iron increase the risk for atherosclerosis, perhaps by favoring the formation of pro-atherogenic oxidized low-density lipoprotein (LDL) [15, 32], these conclusions appear to be premature. Apolipoprotein E (ApoE)-deficient mice, a model of atherosclerosis, fed a high iron diet, had twice as much iron in their plasma, a nine-fold increase in bleomycin-detectable free iron in their plasma, and ten-times as much iron in their livers as control mice at 24 weeks [33]. Somewhat surprisingly, this regimen did not exacerbate, but rather reduced the severity of atherosclerosis, by 50%, and it failed to elevate hepatic levels of heme oxygenase mRNA, which is induced by many different oxidative insults in vitro. Moreover, hepatic levels of protein-bound dityrosine and ortho-tyrosine, two markers of metal-catalyzed oxidative damage in vitro, failed to rise in iron-overloaded animals. These observations suggest that elevated serum and tissue levels of iron are not atherogenic in apoE-deficient mice. Moreover, they call into question the hypothesis that elevated levels of tissue iron promote LDL oxidation and promote atherosclerosis in vivo.

The role of intravenously administered iron in patients not on hemodialysis is uncertain. A review of the published literature shows that there have been five randomized, controlled, clinical trials comparing orally administered iron with intravenously given iron in CKD patients not on dialysis. Stoves et al. randomly allocated 45 patients to treatment with either ferrous sulfate, 200 mg taken orally three times daily (t.i.d.), or iron sucrose, 300 mg given intravenously over 2 h once a month, and followed them for an average of 5.2 months. Iron sucrose was not superior to ferrous sulfate in improving hemoglobin (Hgb) or decreasing the doses of erythropoietin [34]. Aggarwal et al. randomly allocated 40 erythropoietin-treated patients to receive either ferrous sulfate, 200 mg orally t.i.d., or iron dextran, 100 mg intravenously twice a month, and followed them for up to 3 months [35]. The investigators found that the intravenously administered iron dextran was more effective in improving hematological and iron parameters than the orally taken ferrous sulfate was. However, these patients were severely anemic, with a mean Hgb <6.5 g/dl at baseline, which very likely increased their iron requirements. Charytan et al. randomly assigned 96 erythropoietin-treated patients to receive either ferrous sulfate, orally at 325 mg t.i.d. for 29 days, or iron sucrose, 200 mg intravenously given once a week × five doses, and followed them up to 43 days from the onset of treatment [36]. There was no significant difference between the two groups in Hgb levels. Van Wyck et al. randomly selected 188 erythropoietin-treated and non-erythropoietin-treated patients to receive either 1 g of iron sucrose intravenously, in two to five divided doses over a 2-week period, or ferrous sulfate orally, 325 mg t.i.d. for 56 days [37]. The iron sucrose given intravenously was more effective than the ferrous sulfate given orally in improving iron indices at 56 days and in increasing Hgb levels by at least 1 g/dl at any time during the study. Hypotension, nausea, taste disturbances, myalgia, and headaches were associated with i.v. iron sucrose, while constipation, diarrhea, nausea, dyspepsia, and vomiting were associated with the oral ferrous sulfate. Agarwal et al. randomly assigned iron-deficient non-dialysis CKD patients not receiving erythropoietin to receive either ferric gluconate 250 mg intravenously weekly × 4 or ferrous sulfate 325 mg t.i.d. × 42 days [38]. The data from 75 patients were analyzed [i.v. iron n = 36, per os (p.o.) iron n = 39]. Change from baseline in Hgb was similar in the two groups [i.v. iron 0.4 g/dl vs p.o. iron 0.2 g/dl, P = not significant (NS)]. The most common side effect reported for i.v. iron administration was hypotension, while constipation was more common with iron given orally. Thus, the bulk of the evidence suggests that iron given orally can be first-line therapy for patients with CKD not on dialysis. If this therapy fails, intravenously administered iron can be used, particularly if the patient requires erythropoietin.

The balance of risk and benefit of iron given intravenously to patients with CKD is unclear. Iron has benefits that may be independent of the correction of anemia [39]. Furthermore, given the concerns that surround the use of high doses of erythropoietin, it would be prudent for iron to continue to be used judiciously in patients who require erythropoietin [40]. Whereas there is little doubt that iron causes oxidative stress, it is unclear whether oxidative stress so generated leads to poor clinical outcomes and needs to be evaluated in clinical trials. Till these trials are performed, I feel that there is one clinical situation where special caution should be exercised when one is using any type of iron that is to be given intravenously. In animals with sepsis, Zager et al. have demonstrated that i.v. administration of iron can compound the inflammatory response and increase mortality rates [41]. Thus, in patients with sepsis, i.v. administration of iron (and perhaps even orally given iron) should be avoided.