Iron supplementation in the intensive care unit: when, how much, and by what route?

The development of anemia during critical illness is hematologically similar to the anemia of chronic disease/inflammation (ACD), except that the onset is generally acute. Blunting of endogenous erythropoietin production, mediated through the action of inflammatory cytokines, is thought to be the most important cause of this common syndrome [1]. Because low serum iron and elevated serum ferritin levels are also clinical features of ACD, investigators once believed that altered iron metabolism was responsible for this condition. This early hypothesis has merit because iron utilization plays a key role in anemia and can affect the response to endogenous or exogenously administered erythropoietin [2]. Nutritional deficiencies (folic acid, vitamin B₁₂, and iron) can possibly contribute to the etiology of early anemia in the intensive care unit (ICU). In a study conducted by Rodriguez and colleagues [3], 13% of critically ill patients were identified as having potentially correctable nutritional abnormalities, with 9% having laboratory values consistent with iron deficiency (i.e. iron/total iron binding capacity <15%, with a ferritin level <100 ng/ml).

Regardless of whether iron supply is limited as a result of nutritional or functional factors, when the iron available to developing erythroblasts is insufficient to meet the demands of heme synthesis, maturing red cells may contain suboptimal hemoglobin [8]. To prevent the synthesis of hypochromic red cells, administration of iron in conjunction with erythropoietin therapy should be considered in the critically ill patient, but when, how much, and by what route should iron be given?

In a recent review of best practice and research in clinical hematology [8], it was noted that iron uptake into developing erythroblasts begins at an early stage and is completed before late erythroblastosis. Precisely when to administer iron with erythropoietin therapy has not been systematically investigated. In a study examining health-related quality of life in rheumatoid arthritis patients receiving treatment with erythropoietin, all 28 patients who completed the study responded to treatment with erythropoietin [9]. However, 82% developed functional iron deficiency and were well supported with a mean absolute dose of 710 ± 560 mg intravenous iron sucrose. In that 12-week study, iron support was not initiated until the deficiency was noted (after a mean of 4.5 weeks of treatment). In other small studies conducted in elderly or critically ill patients, iron support was initiated concurrently with erythropoietin therapy [10]. Of note, daily intravenous iron therapy alone or in combination with folic acid did not result in a significant increase in reticulocyte count among critically ill patients. The timing of iron administration with erythropoietin clearly warrants further study in larger populations.

Administration of iron in critically ill patients is regarded by some as undesirable because free iron may promote bacterial growth and have detrimental effects in patients who are immunosuppressed or susceptible to infection and sepsis [11, 12]. Iron overload is most commonly associated with chronic transfusions in patients with refractory anemia [4] because there is no physiological mechanism to excrete iron; chelation therapy is generally required to remove it. This constitutes an important reason for careful consideration of iron dosing and route of administration in the critically ill patient. When given by the parenteral route, iron has been implicated in increased susceptibility to bacterial infection [11–14]. This effect is believed to be related to the influence of iron on the immune system [15]. The cytokine-mediated defect in iron release from macrophages in humans is said to have evolved as a primitive mechanism of defense against microbial pathogens to limit their access to iron [12]. Studies on the effects of iron on host defense suggest that increased iron availability, seen with increased transferrin saturation, leads to less efficient bacterial killing of various pathogens by plasma [16]. Elevated serum ferritin level is a risk factor for bacterial infection in hemodialysis patients [17]. This finding was corroborated by Hoen and colleagues at serum ferritin levels greater than 500 ng/ml [18], but this was later refuted when they found no association between serum ferritin and infection in a large prospective trial [19]. In these studies it is not clear whether high ferritin levels are associated with iron overload or an acute phase reaction, or both. In light of this rather limited evidence, some advocate that these observations should not limit iron utilization in patients with chronic kidney disease (CKD) [20]. Additional studies in critically ill patients are needed to determine whether there is a clinically relevant link between infection, septicemia, and iron availability.

Studies of iron supplementation are plentiful in the CKD literature, in pregnancy, and in the pediatric setting. However, even in the well studied CKD population, the dosing, route, and frequency of iron administration are controversial [21–23]. Concomitant use of intravenous iron was deemed superior to oral iron in patients with chronic renal failure treated with erythropoietin [24]. Total dose infusion therapy versus low dose intravenous iron replacement were equally effective in hemodialysis patients receiving erythropoietin therapy [25]. In pregnancy, the intravenous route of iron administration was similar to the oral route in terms of measured hemoglobin increases and toxicity, but intravenous iron increased ferritin concentrations significantly more than did oral iron [26].

Studies in premature infants have attempted to address the dosage and timing of iron supplementation. In neonates with anemia of prematurity, high dose iron (16 mg/kg per day) was no more effective than low dose (8 mg/kg per day) during erythropoietin therapy [27]. Early iron supplementation in infants was safe and prevented red blood cell transfusions as compared with late iron supplementation [28]; no erythropoietin was given in that study. Because there is substantial need for iron in erythropoietin-stimulated erythroid progenitors, Tarng and colleagues [29] recommended maintenance of serum ferritin and transferrin saturation levels over 300 ng/ml and 30%, respectively. They also asserted that, in uremic patients, oral iron is unlikely to keep pace with iron demand, rendering the response to erythropoietin suboptimal. Thus, intravenous iron therapy is believed to be the most appropriate choice, leading to reductions in erythropoietin dosage and costs [29].

How effective is iron alone in the treatment of anemia or ACD type disorders? With regard to reducing the need for transfusions, iron has not proven very effective in improving the capacity for autologous blood donation or in reducing transfusion requirements [10, 13, 36, 37]. Iron was helpful in reducing erythropoietin requirements for the maintenance of hemoglobin in end-stage renal disease patients on hemodialysis, but a similar study has not been done in ICU patients [38]. After 1 month of treatment, intravenous iron alone was sufficient to increase hematocrit in patients with CKD, but the response was less than that noted in the group administered erythropoietin plus iron [39]. In orthopedic surgery patients, those treated with iron alone did not experience a significant increase in erythropoietic variables over baseline values, and the effects of erythropoietin were seen irrespective of the route (oral or intravenous) of iron administration [40]. Two studies of erythropoietin in the critically ill were conducted [36, 37] in which patients were treated with iron therapy, but neither of these two studies addressed the preferred route of iron administration. Oral iron (at least 150 mg/day elemental iron) or parenteral iron (if unable to take oral formulation or inadequate response to iron therapy defined by transferrin saturation < 20% and serum ferritin <100 ng/ml) were administered to all patients starting on study day 1 for 2–4 weeks.

The question has been raised as to whether iron therapy is really needed as an adjuvant to erythropoietin therapy. Iron supplementation was found unnecessary in one small, limited observational study of critically ill surgical patients undergoing erythropoietin treatment [41], because iron store deficiency was not noted during the course of their ICU stay. However, the majority of clinical evidence shows that, if properly balanced, iron therapy can help to optimize erythropoietin treatment. Furthermore, iron deficiency is also the most common cause of resistance to erythropoietin treatment in CKD [42]. Finally, adjuvant therapy, such as ascorbic acid, to increase oral iron absorption and physiologic utilization has been used in many clinical CKD settings and must be tested in critically ill patients [43].

Another clinical concern related to iron supplementation in the ICU is its risk to affect the physiological redox potential. Iron is a mineral that is essential to cellular homeostasis, but an excess of iron can affect physiology and lead to cell injury [44]. During biologic stress, free radicals are formed. These radicals can have detrimental effects at different cellular levels (such as nucleic acid modification) and are involved in many biologic processes that can damage lipid and protein membranes [45]. The role of oxidative stress secondary to iron-derived free radicals and depressed antioxidant reserves has also been associated with pathologic damage in humans [46]. Free hydroxyl radical formation is catalyzed by iron via the Fenton reaction, which involves ferrous iron and peroxide. Although the clinical significance of this biologic reaction and the formation of reactive oxygen species is still debated, stress oxidation is certainly a viable reason to avoid iron overload and to monitor iron parameters during iron supplementation and critical illness [47]. Biologic tissues exposed to higher concentrations of free iron may be prone to oxidative damage [48].

In summary, clinicians must first understand that the majority of the clinical data available are derived from the CKD population. The final decision to administer supplemental iron will depend ultimately on the etiology of anemia in the critically ill patient. Because iron deficiency is noted in a small subset of critically ill patients, iron supplementation following identification of such patients is appropriate. If anemia is multifactorial, as is usually the case in a critically ill patient, then iron alone may not be sufficient to stimulate erythropoiesis but may be useful as an adjunct with erythropoietin. Although not specifically evaluated in the critically ill, studies in pregnant females, neonates, and patients with renal disease suggest that higher iron doses are not more effective than lower ones. Iron overload should be avoided because of its association with adverse effects. Therefore, until proven otherwise, clinicians should probably monitor iron parameters on a regular basis if they elect to administer iron and erythropoietin therapies concomitantly. Based on the CKD population, these parameters may help in optimizing therapy while preventing iron overload. Specific populations of ICU patients (i.e. those with severe infection or sepsis) may not benefit from iron therapy because of possibly increased risk for iron-mediated infection, although this association is still very controversial and is not based on robust clinical evidence. The optimal dose, route, and timing of iron administration in critically ill patients, especially when given concurrently with erythropoietin therapy, remains an open issue that requires further study.